SlideShare a Scribd company logo

Visual Perception Key Readings 1st Edition Steven Yantis

C
cockswasiqi

Visual Perception Key Readings 1st Edition Steven Yantis Visual Perception Key Readings 1st Edition Steven Yantis Visual Perception Key Readings 1st Edition Steven Yantis

Education
1 of 85
Download to read offline
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
Visual Perception Key Readings 1st Edition Steven Yantis download https://ebookbell.com/product/visual-perception-key-readings-1st- edition-steven-yantis-43686586 Explore and download more ebooks at ebookbell.com
Here are some recommended products that we believe you will be interested in. You can click the link to download. Visual Perception Of Emotional And Conversational Facial Expressions 1st Edition Kathrin Kaulard https://ebookbell.com/product/visual-perception-of-emotional-and- conversational-facial-expressions-1st-edition-kathrin-kaulard-51627100 Visual Perception Part 1 Fundamentals Of Vision Low And Midlevel Processes In Perception 1st Edition Susana Martinezconde https://ebookbell.com/product/visual-perception-part-1-fundamentals- of-vision-low-and-midlevel-processes-in-perception-1st-edition-susana- martinezconde-2192184 Visual Perception An Introduction 2nd Ed Michael T Swanston https://ebookbell.com/product/visual-perception-an-introduction-2nd- ed-michael-t-swanston-37316980 Visual Perception And Control Of Underwater Robots 1st Edition Yu https://ebookbell.com/product/visual-perception-and-control-of- underwater-robots-1st-edition-yu-38373170
Visual Perception Through Video Imagery Michel Dhome https://ebookbell.com/product/visual-perception-through-video-imagery- michel-dhome-4313878 Visual Perception Fundamentals Of Awareness Multisensory Integration And Highorder Perception S Martinezconde https://ebookbell.com/product/visual-perception-fundamentals-of- awareness-multisensory-integration-and-highorder-perception-s- martinezconde-4423928 Visual Perception From A Computer Graphics Perspective William Thompson https://ebookbell.com/product/visual-perception-from-a-computer- graphics-perspective-william-thompson-4748526 Visual Perception Physiology Psychology And Ecology 4th Vicki Bruce https://ebookbell.com/product/visual-perception-physiology-psychology- and-ecology-4th-vicki-bruce-5151800 Visual Perception For Humanoid Robots Environmental Recognition And Localization From Sensor Signals To Reliable 6d Poses 1st Ed David Israel Gonzlez Aguirre https://ebookbell.com/product/visual-perception-for-humanoid-robots- environmental-recognition-and-localization-from-sensor-signals-to- reliable-6d-poses-1st-ed-david-israel-gonzlez-aguirre-7324776
Visual Perception Key Readings 1st Edition Steven Yantis
KEY READIN GS sNGst GG Nt kOe ‘Visual Perception Steven Yantis —
LANCHESTER LIBRARY IANIOMIIAMUM 3 8001 00299 0111 lob PERCEPTION
Key Readings in Cognition The aim of this series is to make available to senior undergraduate and graduate students key articles in each area of cognition in an attractive, user-friendly format. Many professors want to encourage their students to engage directly with research in their fields, yet this can often be daunting for students coming to detailed study of a topic for the first time. Moreover, declining library budgets mean that articles are not always readily available, and course packs can be expensive and time-consuming to produce. Key Readings in Cognition aims to address this need by providing compre- hensive volumes, each one of which will be edited by a senior and active researcher in the field. Articles will be carefully chosen to illustrate the way the field has developed historically as well as current issues and research directions. Each volume will have a similar structure to include: ¢ An overview chapter, as well as introductions to sections and articles ¢ Questions for class discussion e Annotated bibliographies e Full author and subject indexes Titles in Preparation: The Psychology of Human Memory Edited by Henry L. Roédiger, Ill and Kathleen McDermott Cognitive Neuroscience Edited by Marie Banich and Neal Cohen The Psychology of Language Edited by Michael Tanenhaus
VloUHL PERCEPTION Essential Headings Edited by Steven Yantis Johns Hopkins University Baltimore, MD LOG oe & »y Gs ~n Nn Ay ~” Mw
USA Publishing Office: » PSYCHOLOGY PRESS A member of the Taylor & Francis Group 325 Chestnut Street Philadelphia, PA 19106 Tel: (215) 625-8900 Fax: (215) 625-2940 Distribution Center: PSYCHOLOGY PRESS A member of the Taylor & Francis Group 7625 Empire Drive Florence, KY 41042 Tel: 1-800-634-7064 Fax: 1-800-248-4724 PSYCHOLOGY PRESS A member of the Taylor & Francis Group 27 Church Road Hove E. Sussex, BN3 2FA Tel: +44 (0) 1273 207411 Fax: +44 (0) 1273 205612 VISUAL PERCEPTION: Essential Readings Copyright © 2001 Taylor & Francis. All rights reserved. Printed in the United States of America. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written permission of the publisher. 12534. 567-89. 0 Printed by Sheridan Books - Braun-Brumfield, Ann Arbor, MI, 2001. A CIP catalog record for this book is available from the British Library. € 6 The paper in this publication meets the requirements of the ANSI Standard Z39.48-1984 (Permanence of Paper) Vasily Kandinsky, Composition 8, July 1923, Oil on canvas, Solomon R. Guggenheim Museum, New York. Photograph by David Heald © The Solomon R. Guggenheim Foundation, New York, FN 37.262 Library of Congress Cataloging-in-Publication Data Visual perception : essential readings / edited by Steven Yantis. p. cm. -- (Key readings in cognition) Includes bibliographical references and index. ISBN 0-86377-597-7 (case : alk. paper) -- ISBN 0-86377-598-5 (paper : alk. paper) 1. Visual perception. I. Yantis, Steven. II. Series. QP491 . V577 2000 152.14--dce21 CIP 00-042311 ISBN: 0-86377-597-7 (case) ISBN: 0-86377-598-5 (paper) overtly University
Contents About the Editor xi Acknowledgements _ xiii Visual Perception: An Overview 1 PART 1 Theoretical Perspectives 19 READING1 Introduction to Reading 1 21 Concerning the Perceptions in General 24 H, von Helmholtz READING 2 Introduction to Reading 2. 45 A Decision-Making Theory of Visual Detection 48 W. P. Tanner, Jr. and J. A. Swets READING 3 Introduction to Reading 3. 57 Single Units and Sensation: A Neuron Doctrine for Perceptual Psychology? 60 H. B, Barlow READING 4 Introduction to Reading 4 81 The Theory of Information Pickup and its Consequences 84 J, J. Gibson
vi m_ Visual Perception READING 5 Introduction to Reading 5 101. The Philosophy and the Approach 104 D. Marr i Part 1: Discussion Questions and Suggested Readings 124 PART 2 Early Vision 125 READING 6 Introduction to Reading 6 127 An Opponent-Process Theory of Color Vision 129 L. M. Hurvich and D, Jameson READING 7 Introduction to Reading 7 145 Receptive Fields and Functional Architecture of Monkey Striate Cortex 147 D. H. Hubel and T. N. Wiesel READING 8 Introduction to Reading 8 169 On the Existence of Neurones in the Human Visual System Selectively Sensitive to the Orientation and Size of Retinal Images 172 C, Blakemore and F. W. Campbell READING 9 Introduction to Reading 9 191 A Direct Demonstration of Functional Specialization in Human Visual Cortex 193 S, Zeki, J. D. G. Watson, C. J. Lueck, K. J, Friston, C. Kennard, and R, S. J. Frackowiak READING 10 Introduction to Reading 10 203 Neuronal Correlates of a Perceptual Decision 205 W. T. Newsome, K. H. Britten, and J. A. Movshon Part 2: Discussion Questions and Suggested Readings 209
Contents m vii PART 3 Perceptual Organization and Constancy 211 READING 11 Introduction to Readings 11 and 12 213 Laws of Organization in Perceptual Forms 216 M. Wertheimer READING 12 Figure and Ground 9225 E, Rubin READING 13 Introduction to Reading 13 231 The Moon Illusion 233 L. Kaufman and |, Rock READING 14 Introduction to Readings 14 and 15 243 Brightness Constancy and the Nature of Achromatic Colors 246 H. Wallach READING 15 Grouping Based on Phenomenal Similarity of Achromatic Color 256 |,Rock, R. Nijhawan, S. Palmer, and L. Tudor ° Part 3: Discussion Questions and Suggested Readings 266 PART 4 Object and Spatial Vision 269 READING 16 Introduction to Reading 16 271 A Case of Visual Agnosia with a Contribution to Theory 274 H, Lissauer READING 17 Introduction to Reading 17 293 Object Vision and Spatial Vision: Two Cortical Pathways 296 M. Mishkin, L. G. Ungerleider, and K. A. Macko
viii m Visual Perception READING 18 Introduction to Reading 18 303 Phenomenal Coherence of Moving Visual Patterns 306 E, H. Adelson and J. A. Movshon READING 19 Introduction to Reading 19 311 Mental Rotation of Three-Dimensional Objects 313 R, N. Shepard and J. Metzler READING 20 Introduction to Reading 20 317 Recognition-by-Components: A Theory of Human Image Understanding 320 |, Biederman Part 4: Discussion Questions and Suggested Readings 341 PART 5 Visual Attention and Awareness 343 READING 21 Introduction to Reading 21 345 A Feature-Integration Theory of Attention 347 A.M. Treisman and G. Gelade READING 22 Introduction to Readings 22 and 23 359 Selective Attention Gates Visual Processing in the Extrastriate Cortex 361 J. Moran and R. Desimone READING 23 Voluntary Attention Modulates fMRI Activity in Human MT-MST 365 K, M. O'Craven, B. R. Rosen, K. K. Kwong, A. Treisman, and R, L. Savoy READING 24 Introduction to Readings 24 and 25 375 Visual Capacity in the Hemianopic Field Following a Restricted Occipital Ablation 378 L. Weiskrantz, E. K. Warrington, M. D, Sanders, and J. Marshall
Contents m ix READING 25 The Role of Temporal Cortical Areas in Perceptual Organization 393 D, L. Scheinberg and N. K. Logothetis Part 5: Discussion Questions and Suggested Readings 402 Appendix: Reading Journal Articles in Cognitive Psychology 405 H. L. Roediger, Ill and D. A. Gallo Author Index 417 Subject Index 425 The following color plates follow page 210 Color Plate 9.1 for Reading 9 Color Plate 9.2 for Reading 9 Color Plate 9.3 for Reading 9 Color Plate 9.4 for Reading 9 Color Plate 23.5 for Reading 23 Color Plate 23.6 for Reading 23
— “=e, a hive me ia nal cpa umes es nsilesins gOlandQa20 4 AD te fctiesadti >: egebaah bateaiglieBinsano a Re if: 20k ancnn aapga — s ~- *5 Secu awl > Aas a) on SF xa onto, eases 38 eSh >xebat Wwirbtties, on to Ailing £0 Lally Ranggnits aeoagetinge . “igen Sear pie = Pal = weary Lake we oa jones poe ional eae ae r £5nities 101 APS.stat yu), 1 etd Mection ane Awlwraness 7. ~— i" wire 1 ee . ; — a <a ; Poa pecoinoereat vache k 93 7 —— = : ators inteciation Maury of Agron. ja , -_ ; oO ee ent , Gitex ~~. 7 py divvk gt : ‘ io "WEES Netliny: oa dag BD] ge ae er ptetFakes.ina Pend «ete S a ; ae tay et”. eee - i; “on 9 a cin it ae ao as | - hares eae ll ica ~ 2° oe vere 2 <a. 7a si i ye teal a artes Wi Prides eh Orcas, Wii Sens Ration mY.
Hhout the Editor eo Steven Yantis is a Professor of Psychology and Cognitive Sciences at The Johns Hopkins University. He received a B.S. in Psychology from the University of Washington in 1978 and a PhD in Experimental Psychology from the University of Michigan in 1985. He is the recipient of the Troland Research Award from the National Academy of Sciences and the Early Career Award from the American Psychological Association. Professor Yantis’s research is concerned with the psychological and neural mechanisms of visual attention and perceptual organization. x1
o ~ i — ns A sat sicamcmmacmmmcatnt ee A A ALLO D is : ’ _ 7 4 Ke ‘ s 7 a ogee «gs be Yooh so wo 8 NY ge ah inertygolodnae thaa 5 (Sven sti gent acto a a _ rot mi vig hein mtSA truOTBACORA a, 1 en a renee brat ooh YoBrigg, orf ofoH .eBer rn é : cent awh ees aidott tafe naan Ws ytd whan ’ : : ihe reovuen) riaang vaneten conc Mieka 7% . a e an eae Mile < ar reali tenraty ioervaings cea ipo : ‘| ae rae ; y "a * = ‘ a e A yn | see = » ; re e > : ; | ‘ ¥ ; ’ » 3 ' 7
Acknowledgements The Authors and Publishers are grateful to the fol- lowing for permission to reproduce the articles in this book: Reading 1: H. von Helmholtz (1925). Physiologi- cal Optics. (Vol III, $26. Concerning the percep- tions in general. pp. 1-36). Translated from the Third German Edition and Edited by J.P.C. Southall. New York: Optical Society of America. (Original work published 1896). Reading 2: H. B. Barlow, Single units and sensa- tion: A neuron doctrine for perceptual psychology? Perception, 1, 371-394. Copyright © 1972 by Pion Limited, London. Reprinted with permission. Reading 3: W. P. Tanner, Jr. and J. A. Swets, A decision-making theory of visual detection. Psy- chological Review, 61, 401-409. Copyright ©1954 by the American Psychological Association. Re- printed with permission. Reading 4: J. J. Gibson, The ecological approach to visual perception. Chapter 14: The theory of information pickup and its consequences (pp. 238- 263). Boston: Houghton Miflin Co., 1979. Re- printed with permission. Reading 5: D. Marr, Vision. Chapter 1: The phi- losophy and the approach (pp. 8-38). San Fran- cisco: W. H. Freeman and Co., 1982. Reprinted with permission. Reading 6: L. M. Hurvich and D. Jameson, An opponent-process theory of color vision. Psycho- logical Review, 64, 384-404. Copyright © 1957 by the American Psychological Association. Re- printed with permission. Reading 7: D. H. Hubel and T. N. Wiesel, Recep- tive fields and functional architecture of monkey striate cortex. Journal of Physiology, 195, 215-243. Copyright © 1968 by Cambridge University Press. Reprinted with permission. Reading 8: C. Blakemore and F. W. Campbell, On the existence of neurones in the human visual sys- tem selectively sensitive to the orientation and size of retinal images. Journal of Physiology, 203, 237- 260. Copyright © 1969 by Cambridge University Press. Reprinted with permission. Reading 9: S. Zeki, J. D. G. Watson, C. J. Lueck, K. J. Friston, C. Kennard, & R. S. J. Frackowiak, A direct demonstration of functional specialization in human visual cortex. Journal of Neuroscience, 11, 641-649. Copyright © 1991 by the Society for Neuroscience. Reprinted with permission. Reading 10: W. T. Newsome, K. H. Britten, & J. A. Movshon, Neuronal correlates of a perceptual decision. Nature, 341, 52-54. Copyright © 1989 by Macmillan Magazines Ltd. Reprinted with per- mission. Reading 11: M. Wertheimer, Untersuchungen zur Lehre von der Gestalt, II [Laws of organization in perceptual forms]. Psycholoche Forschung, 4, 301- 350. Excerpts translated and reprinted in W. D. Ellis (Ed.), A source book of Gestalt psychology (pp. 71-88). New York: Routledge, 1999. Reprinted with permission. Xili
xiv m_ Visual Perception Reading 12: E. Rubin, Visuaell wahrgenommene Figuren [Figure and Ground]. Copenhagen: Gyldendalske, 1921. [excerpts translated and re- printed in D. C. Beardslee & M. Wertheimer (Eds.), Readings in Perception (pp. 194-203). Princeton, NJ: D. Van Norstrand Co., Inc.]. Reading 13: L. Kaufman and I. Rock, The Moon Illusion. Scientific American, 207, 120-132. Copy- right © 1962 by Scientific American, Inc. Re- printed with permission. All rights reserved. Reading 14: H. Wallach, Brightness constancy and the nature of achromatic colors. Journal of Experi- mental Psychology, 38, 310-324. Copyright © 1948 by the American Psychological Association. Reprinted with permission. Reading 15: I. Rock, R. Nijhawan, S. Palmer, & L. Tudor, Grouping based on phenomenal simi- larity of achromatic color. Perception, 21, 779-789. Copyright © 1992 by Pion Limited, London. Re- printed with permission. Reading 16: H. Lissauer (1890), Ein Fall von Seelenblindheit nebst einem Beitrag zur Theorie derselben. Archiv fiir Psychiatrie, 21, 222-270. [translated and reprinted in 1988 as “A case of vi- sual agnosia with a contribution to theory.” Cog- nitive Neuropsychology, 5, 157-192; commentary by T. Shallice & M. Jackson. Lissauer on agnosia. Cognitive Neuropsychology, 5, 153-156. Copy- right © 1988 by Taylor and Francis. Reprinted with permission. Reading 17: M. Mishkin, L. G. Ungerleider, & K. A. Macko, Object vision and spatial vision: Two cortical pathways. Trends in Neurosciences, 6, 414-417. Copyright © 1982 by Elsevier Science. Reprinted with permission. Reading 18: E. H. Adelson and J. A. Movshon, Phenomenal coherence of moving visual patterns. Nature, 300, 523-525. Copyright © 1982 by Macmillan Magazines, Ltd. Reprinted with per- mission. Reading 19: R. N. Shepard and J. Metzler, Men- tal rotation of three-dimensional objects. Science, 171, 701-703. Copyright © 1971 by the American Association for the Advancement of Science: Re- printed with permission. Reading 20: I. Biederman, Recognition-by-com- ponents: A theory of human image understanding. Computer Vision, Graphics, and Image Process- ing, 32, 29-73. Copyright © 1985 by Academic Press. Reprinted with permission. Reading 21: A. M. Treisman and G. Gelade, A feature-integration theory of attention. Cognitive Psychology, 12, 97-136. Copyright ©1980 by Academic Press. Reprinted with permission. Reading 22: J. Moran and R. Desimone, Selec- tive attention gates visual processing in the extrastriate cortex. Science, 229, 782-784. Copyright © 1985 by American Association for the Advancement of Science. Reprinted with permission. Reading 23: K. M. O’Craven, B. R. Rosen, K. K. Kwong, A. Treisman, & R. L. Savoy, Voluntary attention modulates fMRI activity in human MT- MST. Neuron, 18, 591-598. Copyright © 1997 by Cell Press. Reprinted with permission. Reading 24: L. Weiskrantz, E. K. Warrington, M. D. Sanders, & J. Marshall, Visual capacity in the hemianopic field following a restricted occipital ablation. Brain, 97, 709-728. Copyright © 1974 by Oxford University Press. Reprinted with permission. Reading 25: D. L. Sheinberg and N. K. Logothetis, The role of temporal cortical areas in perceptual organization. Proc. National Academy of Sciences USA, 94, 3408-3413. Copyright © 1997 by The National Academy of Sciences of the USA. Re- printed with permission. Appendix: From Research Methods in Psychol- ogy (with InfoTrac), 6" edition, by D.G. Elmes, B.H. Kantowitz, and H.L. Roediger, III © 1999. Reprinted with permission of Wadsworth, a divi- sion of Thomson Learning. Fax 800-730-2215.
V Visual Perception: finOverview Steven Yantis isual perception allows organisms to obtain knowledge of their surroundings by sensing light reflected from surfaces. Its ultimate purpose is to allow one to know what objects are present so as to behave appropriately and in accordance with one’s current behavioral goals. Although recognizing objects is subjectively effortless and nearly instantaneous, the steps that it requires are amazingly complex. Understanding visual perception requires an understanding of a diverse and multifaceted body of knowledge, including physics and optics, neuroanatomy and neurophysiology, psychology, and computation. This volume contains 25 influential chapters and articles from the vision science litera- ture that reflect the theoretical and empirical themes that have dominated research in visual perception over the last century. They are representative of a huge body of work that is truly interdisciplinary. The fertile crosstalk between psychology and neuroscience in par- ticular is made possible by the close correspondence between neural activity on the one hand and certain aspects of perceptual behavior on the other. This means that we can begin to grasp the neural implementation of vision more readily than that of other cognitive functions like memory, language, or planning and decision making where the link between neurophysiology.and behavioris-less.well.understood, What makes this interplay particu- larly exciting right now are technological advances that have brought new scientific tools to bear in the cognitive neuroscience of vision (e.g., technologies that permit us to collect images of brain activity in behaving human observers), so that new discoveries about how the human brain implements vision are now reported routinely in the scientific literature and the press. In this introductory chapter, I will outline of some of these theoretical and empirical themes and how they have been influenced by the tools that are available to practicing scientists. Of course, in this brief space it is not possible to adequately cover this topic in all its complexity and richness. At the end of this chapter, a list of several other sources that provide deeper treatments is provided. I started with the assertion that the purpose of vision is object recognition. Of course, one could quibble with this assertion, but it is a good starting point nevertheless. If we wish to understand how the visual system can take as its input a dynamic pattern of light and color at the retina and produce as its output a label that identifies or categorizes an object, we must first break the problem down into a smaller number of slightly more manageable parts. The selected readings have been grouped according to the following questions:
2 Visual Perception * The problem of early vision: How are elementary sensory attributes (e.g. color, motion) encoded? ¢ The problem of perceptual constancy: How are elementary attributes “corrected” for irrelevant viewing conditions? ¢ The problem of perceptual organization: How are sensory attributes conjoined into rep- resentations of surfaces and objects? * The problem of object recognition: How are surface representations used to categorize or identify objects? * The problem of attention: How does the observer select relevant objects and ignore irrel- evant objects? ° The problem of visual awareness: How is consciousness achieved? All of these problems are under active investigation; none has been solved. Of course, more progress has been made in some areas than others: we know a very great deal (but not everything) about how elementary sensory properties are represented in the visual system; we know very little about awareness. Although object recognition is arguably the most significant of these problems, there is little consensus about how this is achieved. This is not surprising, of course, given its complexity and that it relies to some degree on success in all of the other areas. The first section in this book is titled Theoretical Approaches. The choice of a theoreti- cal orientation is in some sense orthogonal to the substantive issues sketched above. Of course, one’s choice of a theoretical approach is utterly crucial in determining research strategies and identifying relevant questions. Indeed, according to certain theoretical tradi- tions, the questions outlined are simply not the important ones. Yet most scientists do not spend much time mulling possible theoretical approaches. In most sciences one main ap- proach tends to be dominant at any given time, and most scientists implicitly accept that approach. In contemporary vision research, the dominant theoretical approach can be traced to Helmholtz’s constructivism, which is the view that we construct internal representations of the objects in the visible scene that are most likely to have given rise to the pattern of light currently impinging on our retinas. This requires us to take into account the early sensory data, of course, but it also allows for the influence of past experiences (i.e., memory), expectation about what is likely in the current context, and what aspects of the scene are important given current behavioral goals (i.e., attention). In other words, the representation of objects in a scene is a joint result of both bottom-up information from the eyes, and top- down information from memory and attentional mechanisms. Almost all of the readings in this volume start from this common theoretical platform. (Gibson’s idea that perception is in an important sense direct constitutes the most notable alternative view, but his ideas have been extremely influential even among adherents of the constructivist perspective.) Two Principles of Visual Perception Before turning to the visual themes themselves, there are two central principles that char- acterize vision and that will help to orient the discussion. The first principle is that of functional specialization. In the section on methodology that concludes this chapter, the evidence for functional specialization emerged in the late 19" century. There are many ways in which functional specialization can be illustrated. Perhaps the first point at which specialization occurs in vision is at the very earliest point it could occur: in the photoreceptors themselves. There are four different types of photoreceptors grouped into two classes: the rods, and three types of cones. In addition to their shapes, there are at least three ways in which the rods and cones differ, and each of these has implications for their functions. First, the rods are poorly suited to represent
An Overview m 3 color information, while the three types of cones are ideally suited for this task (more on this later). Second, the rods are much more sensitive to low levels of illumination, and they are what. we must rely on for night vision; the cones are used under ordinary daylight viewing conditions. Third, the rods and cones are distributed throughout the retinal mosaic differently: the cones are packed densely in the center of gaze—the fovea—with no rods; both rods and cones are interspersed outside the fovea with density gradually decreasing into the periphery. The high density of the cones in central vision allows a very fine spatial sampling of the pattern of light in central vision, which in turn allows us to see fine spatial detail there. Functional specialization does not end with the rods and cones. The optic nerve, a bundle of fibers that carries visual information from the eye to the brain, consists of at least two distinct subpopulations of fibers that project to distinct regions in the brain. M fibers tend to favor information that varies temporally, such as motion or flicker, while P fibers tend to carry information about static properties such as color, orientation, or depth. This functional distinction continues as the information from the eyes makes its way to higher centers in the visual cortex, and it is intimately related to the distinction between the “what” and “where” pathways suggested by Mishkin et al. (Reading 17). As suggested by Zeki et al. (Reading 9), certain areas of cortex contain a predominance of cells that appear to be function- ally specialized for a single visual attribute (e.g., area V4 for color, area V5 for motion). A second principle of visual perception is that of distributed coding of perceptual at- tributes. There are (roughly speaking) two ways that any given attribute (e.g., color, edge orientation) can be represented in cortical tissue. According to a local coding approach, any given attribute or combination of attributes is represented by a single specialized cell. In the limit, we might arrive at a cell that responds if and only if your grandmother’s face appears somewhere in the visual field (a so-called “grandmother cell”). The advantage of this approach is that all one would need to do is monitor the cells that were most active at each location and these would reveal just what combination of features was present there. One major disadvantage of local coding is that it would require an implausibly large num- ber of specialized cells to adequately capture the richness of the visual world. An alternative approach, and this is the one the visual system tends to use, is that of distributed coding. Here, there exist cells that will respond to a range of feature values to different degrees. For example, a cell might respond best to an edge oriented vertically, and less well (in a graded fashion) as the orientation is displaced from the optimal one. Such a cell could be said to be tuned to orientation, and the location and width of the tuning function characterizes its sensory properties. A given edge would tend to drive a popula- tion of cells with different preferred orientations. The edge represented by the pattern of activity over that population of cells. This is precisely how orientation representation works, and it is also the basis for the trichromatic theory of color vision, where there are only three different types of color-sensitive cells with highly overlapping wavelength preference func- tions (called spectral sensitivity functions) that can with great precision encode a vast number of different colors by virtue of their relative levels of activity for any given wavelength. These two foundational and related principles of functional specialization and distrib- uted coding will reappear throughout the readings in this volume. In the following sections, some of the progress that has been made in each of the areas outlined earlier is reviewed. Palmer (1999) provides a detailed and highly recommended treatment of these and many other problems in vision. Early Vision Early vision refers to the part of vision that creates the initial representation of elementary sensory properties such as color, the orientation of edges, motion, and depth. These repre-
4 sm Visual Perception sentations tend to be local and veridical (in the sense that they closely mimic the properties that are present in the retinal image itself). This is not to say that these representations are simple, however. Before proceeding, it will be useful to define the concept of a receptive field. This is often a difficult concept to grasp, but it turns out to be absolutely crucial to any real under- standing of how the brain represents sensory information. A receptive field is a property of an individual sensory neuron. The receptive field of a visual neuron is that portion of the retina that, when stimulated, causes a change in the firing rate of the neuron in question. Imagine you are a visual neurophysiologist and you have placed an electrode next to the optic nerve, so you can “listen in” on the activity of a single optic nerve fiber. From time to time you may detect a stray signal from the cell. When you then present a stimulus on the retina (e.g., a small spot of red light, or a moving vertical bar, or even a toilet brush), you may eventually find a location that will cause the cell to increase its firing rate when illu- minated. If you repeat this experiment many times and “add up” all the locations that cause the cell to either increase or decrease its firing rate, you will have defined that cell’s recep- tive field. It is, in essence, the portion of the retina that the cell in question monitors. One also must specify the properties of the stimulus that effectively drives the cell, for example its color, direction of motion, and shape. The extent to which any given cell responds to a visual stimulus depends jointly on where on the retina the stimulus appears, and what the stimulus is. As one measures the receptive field properties of cells at higher levels of the visual system, one finds that more and more complex configurations of light are required to drive the cells. In the retina, cells will respond to small circular spots of light in a very specific location; in V1, cells respond best to small bars of light in a particular location and with a particular orientation. In V4, many cells prefer a corner with a particular color and orientation. In the inferotemporal cortex, many cells seems to prefer highly complex and seemingly arbitrary shapes (Tanaka, 1996). The following is a brief review of what we know about early vision. For present pur- poses, the discussion will be restricted to orientation, spatial scale or size, depth, color, and motion; these are the sensory properties that have received the most attention. The first three of these are aspects of spatial vision and contribute to representations of surface shape, which is crucial for object identification. As pointed out by Palmer (1999), knowing the shape of an object tells you a lot about what the object is, what it is for, even what its color is likely to be, but knowing the color or motion of an object tells you almost nothing about what the object is or about what its other attributes are likely to be. Color and mo- tion, of course, are themselves of great importance in enriching our visual experience and in helping us deal with the inherently dynamic world in which we live. Space The investigation of spatial properties like orientation, depth, and spatial scale began with Hubel and Wiesel’s pioneering investigations (Reading 7) in which they discovered (among many other things) that in primary visual cortex, most cells prefer an elongated, oriented, and specifically located region of stimulation to fire optimally. The representation in V1 was not of spots, but of something more useful for representing shape: oriented edges. It is useful to step back a moment and think about why an explicit representation of oriented edges is so important. The retinal image contains simply a pattern of light and dark distributed in two dimensions (we can ignore color, depth, and motion for now) and sensed by the mosaic of photoreceptors. One can think of this as a 2D array of numbers, each of which corresponds to the brightness of a small location on the retina. The shape, location, and orientation of objects, shadows, and surfaces is not explicitly given in the array of brightness values. In order to recover these properties, some computation must be performed on the array of brightnesses (e.g., taking the difference between adjacent num-
An Overview m 5 bers to ask whether the brightnesses there differ, as they would if an edge was present in that location). That is just the sort of computation that retinal ganglion cells (RGCs) and cortical cells carry out. RGCs specify whether a given location has more (or less) brightness than the surrounding locations in all directions. V1 cells specify whether a given location contains an edge (or part of an edge) with a particular orientation and aparticular polarity (e.g., dark to light or vice-versa). It does so by combining the outputs of a set of RGCs that happen to be aligned properly so as to provide a sample of the retina in the corresponding location and orientation. This V1 representation has essentially thrown away the brightnesses in the retinal image. What is left, however, is much more useful: a representation of localized edges. This is useful precisely because a shape is specified by the locations and orienta- tions of its boundaries. The discovery that cortex was in the business of representing ori- ented edge fragments was therefore a landmark in the effort to understand object recognition. Scenes contain (at least) two other sorts of elementary spatial information. One of these is information about the size (or spatial scale) of features. The visual system encodes infor- mation about spatial scale with cells whose receptive fields vary in size so as to be selective for different spatial frequencies. This turns out to be important because of the following mathematical fact: any spatial pattern can be represented as the sum of an appropriately chosen set of spatial frequency patterns called gratings. (This is a nonintuitive idea that is discussed in more detail in the introduction to Reading 8, Blakemore and Campbell, and in the reading itself.) This means that by representing spatial frequencies, the visual system provides some of the building blocks that are likely to be useful for representing shapes and objects. The last kind of spatial information that is contained in scenes is surface depth and orientation. Because objects are typically three-dimensional, it is often not sufficient to represent simply the shape of the object’s outer boundaries (in fact, it is quite difficult to recognize all but the simplest of objects based only on the object’s silhouette). Therefore, it is necessary to know something of the location of the surfaces that bound the object and their distances and orientations relative to the viewer. There are many redundant sources of information in the retinal image, including pictoral cues like linear perspective, overlap, or relative size, relative motjon cues (objects a different depths will have different motion trajectories as the observer moves in the environment), and, perhaps most importantly, stereoscopic depth cues. Stereopsis is the process of combining the two slightly different views of the visual scene that are available at the two eyes (because they view the world from approximately 6 cm apart) to recover information about the relative depth of features in the scene. This requires a process of matching corresponding features in the two eyes (this is the correspondence problem) and estimating the extent to which they fall on corre- sponding or noncorresponding points on the two retinae. By definition, the image feature at the current point of fixation falls on corresponding points in the two eyes (i.e., the two foveas). Any other image feature that falls on corresponding points on the two eyes (that is, points that are equidistant from the foveas in the two eyes) are said to fall on the horopter, an imaginary surface that is defined by all points at the same distance from the observer as the current point of fixation. Any image feature that falls on noncorresponding points in the two eyes exhibits binocular disparity; whenever binocular disparity is nonzero, the image feature in question is at a depth other than the depth of the current point of fixation (or, equivalently, it is either closer than or further away than the horopter). The magnitude of binocular disparity for any given image feature is proportional to that feature’s distance from the horopter. It is known that stereoscopic depth information (specifically, the magnitude of binocular disparity) is available early in cortex. Thus depth information can in principle be combined with information about orientation and spatial scale to provide a rich representation of the spatial layout of edges and surfaces in the scene.
6 @ Visual Perception Color The first major steps in coming to understand color vision were taken early in the 19" century by Thomas Young, and then, independently in the mid 19" century by Helmholtz. The story of the trichromatic theory of color vision and the debate about the alternative opponent-process theory advocated by Herring is contained in the introduction to Reading 6, Hurvich and Jamison. We now know in detail the spectral sensitivities of the three classes of cones and how the cone responses contribute to our ability to match mixtures of colors to one another. This is a theoretical success story of the first magnitude. We also know, based on the work of Hurvich and Jamison and of neurophysiologists since, that the oppo- nent color system that receives input from the cones subserves judgements about color appearance. These two mechanisms can together account for a wide range of color percep- tion phenomena, including color adaptation (the fact that after looking at a red patch fora few minutes, one sees green when subsequently viewing a white paper); simultaneous color contrast (the fact that a gray patch surrounded by bright red will appear slightly green, or a gray patch surrounded by white will appear darker than the same gray patch surrounded by black); Mach Bands, first described in the 19 century by Ernst Mach (the fact that the apparent brightness of the regions surrounding a black-white edge is nonuni- form: the dark side of the edge looks somewhat darker near the edge than do regions far- ther away, while the light side of the edge looks somewhat lighter near the edge than do regions further away); and many forms of color blindness (which can often be attributed to a genetic defect in which one or more of the three cone classes is missing). Motion The first and by far most common kind of motion experience arises when (a) a stationary observer views a moving object or (b) a moving observer views stationary objects (or some combination of these). The simplest case is for a stationary observer with eyes still, look- ing at a stationary scene containing a single smoothly moving object. For this case, simple models of motion representation such as that suggested by Reichert and discussed in the introduction to Reading 18 (Adelson and Movshon) are suitable. Needless to say, however, this kind of simple motion is extremely rare. If nothing else, people tend to move their eyes about 3 times a second, and this causes abrupt displacements of the visual scene even if it contains no motion at all. Furthermore, people tend to move through their environment, and yet they have no trouble experiencing the stable parts of the scene as such and tracking objects that are moving relative to them (e.g., the problem faced by an outfielder trying to catch a fly ball; see McBeath et al., 1995). These forms of motion require more complex systems for coordinating self-motion and retinal image motion. There are several additional motion phenomena, some fairly common, others laboratory inventions, all of which require explanation. These include the motion aftereffect (the fact that after viewing a motion stimulus for several minutes—for example, a waterfall—one sees motion in the opposite direction — in this case, upward—when the view is shifted to a static scene), autokinetic motion (the fact that a small stationary spot of light in a com- pletely dark room will appear to move randomly after a short while), and many forms of apparent motion (motion experiences when viewing a sequence of locations that are stimu- lated in turn but discretely, as in a motion picture). Perceptual Constancy and Organization The early sensory representation of a scene is inadequate for object recognition in at least two major respects. First, it contains representations of image properties and not represen-
An Overview m 7 tations of scene properties. Second, the representation is local in scope but any representa- tion of objects necessarily requires a more global perspective to conjoin parts into wholes. Perceptual Constancy First is the problem of perceptual constancy, which can be stated as follows: Most retinal measurements of scenic attributes are affected not only by properties of the object in ques- tion, but also by irrelevant and transient factors such as the current illumination, the dis- tance between the object and the observer, and the object’s orientation relative to the ob- server. To be useful, vision needs to represent object properties and to discount the irrelevant environmental conditions that are not intrinsic to the objects. In other words, how can we correctly perceive objects despite irrelevant variation in viewing conditions? To give just one example of this problem, consider color vision for a moment. Informa- tion about color is carried by the wavelength of light: the spectrum of colors in a rainbow depicts the portion of electromagnetic radiation with wavelengths from about 400 (blue) to 700 (red) nanometers. The color of a surface depends on the distribution of wavelengths in the light reflected from that surface from a light source such as the sun or a lightbulb. For example, a red apple will tend to reflect long wavelengths of light and absorb short wave- lengths of light. So far, so good. We simply need a way to encode different wavelengths, and we can recover the surface color of the apple. Unfortunately, there is a problem. The spectral composition of the illuminant (e.g., the sun) affects the distribution of light reflected from a surface. A white light such as sunlight has about equal energy at all visible wavelengths, but some sources, such as a flourescent lightbulb, have quite imbalanced spectral distributions. If a red apple is illuminated by a source that does not have much energy in the red portion of the spectrum, the distribution of light that itreflects into the eye will be very different than ifitis illuminated by a source, like the sun, with a fairly balanced spectral distribution. This suggests that our perception of the color of various surfaces will depend both on the (desired) surface properties of the object and on the (irrelevant) properties of the illuminant. An apple could appear to change color when we carry it from our yard into the kitchen where it is illuminated by incandescent light. Similar problems arise jn other stimulus domains. Visual information about object size depends on the object’s true size and the object’s distance from the observer, because the retinal image size of the object varies directly with distance. How can we come to know the relevant property (object size) and discount the irrelevant property (distance)? Visual in- formation about object shape depends on the orientation of the object relative to our line of sight (so that a square will project a parallelogram or a diamond onto the retina depending on its slant and tilt). How can we come to know the relevant property (shape) and discount the irrelevant property (orientation)? Each of these is a problem in perceptual constancy. We are, in fact, able to correctly know the color, size, and shape of objects despite variation in irrelevant factors such as illumi- nation, distance, and orientation. The question is how this is achieved. It turns out that the visual system has evolved different solutions in each domain, specialized for the particular properties in question. Although it is implied that the mechanisms of perceptual constancy are not part of early vision by placing them in separately labeled sections, most of percep- tual constancy is indeed achieved by what most vision scientists would call early vision. Color constancy, for example, begins as early as area V4, and certain aspects of color constancy, its precursors so to speak, arise even earlier (see Zeki, 1993, for a discussion). Perceptual Organization The second main problem of the early sensory representation is that it is local in scope. This is a problem because most real scenes contain multiple objects that are arranged in 3D
8 Visual Perception space and that both occlude one another and occlude parts of themselves. To take an eXx- treme example, consider the problem of looking at your garden through the partially open slats of a Venetian blind. Although the tree and foliage in the garden are broken up into horizontal strips by the blind, we are able to attend to and recognize the objects in the garden with little difficulty. Any understanding of this ability requires us to explain how the visible part of a branch in one horizontal strip are correctly conjoined with the corre- sponding part of the branch in the next horizontal strip, and more puzzlingly how the occluded parts of the branch are “filled in” to produce a representation of a single, coher- ent, solid branch despite extreme fragmentation in the retinal image. Many of the Gestalt principles of perceptual organization are discussed in Reading 11 by Wertheimer. These principles characterize how the visual system puts elements in the retinal image together so as to form coherent object representations. These principles in- clude similarity (the texture and color and depth of the branch are similar at all points along it, and furthermore those properties are different for nearby image regions; thus similarity of surface properties is a good clue that those regions all belong to the same object), proximity (the visible parts of a reasonably compact object tend to be in nearby regions of space, so all other things being equal, two image regions will tend to be near one another if they are part of the same object), and common motion (if the branch is blowing in the wind, all its parts will move together, and other non-branch objects, such as the fence behind the branch, will tend not to move with it; thus common motion in two regions is a good clue that they are part of the same object). The problem of perceptual completion or “filling in” has also been investigated by per- ceptual psychologists. There are two sides to this problem. One is that of what Michotte et al. (1991) called amodal completion. The partly occluded branch behind the blind is expe- rienced as complete and continuous despite the fact that parts of the branch are completely absent from the retinal image. This is called amodal completion, because there is no infor- mation coming from a sensory modality to support perceptual completion, and in fact there is a sense that one “knows” that the branch is complete without having an explicit visual experience of the branch in the occluded regions. Amodal completion depends on two factors: (1) the edges of the occluded objects must be “relatable” such that they could plausibly be considered part of a single common contour (this essentially requires that there are no implausibly large bends in the contour; see Kellman & Shipley, 1991), and (2) the boundary between the occluding surface and the occluded surface must “belong” to the former and not the latter (this essentially requires that the occluding surface be perceived as in front of the occluded surface; see Nakayama et al., 1995). The other form of perceptual completion is the perception of illusory contours, a phe- nomenon first described by Kanizsa (e.g., 1979). Illusory contours arise most famously in the context of the Kanizsa triangle at right. Here the seg- mented disks serve as inducers for the illusory shape. The shape is illusory because there is no physical contrast to & a define the edges of the triangle where the page is white, yet most observers report that they see a dim but clearly ” discernable edge there. Furthermore, the brightness of the figure is often reported as being greater than the brightness ae of the background. Clearly the relatability of the bound- ne aries of the segments (i.e., that they are nearly collinear) supports the perceptual conclusion that a curved white tri- os angle is partly occluding three black disks (which, by the way, are perceived as complete disks by virtue of amodal completion!). The visual system “fills in’ where the con- G8 trast of the perfectly white triangle with the perfectly white
An Overview m 9 background is zero, and makes it slightly greater than zero. Von der Heydt et al. (1984) have provided evidence about the neural basis of illusory contours in V2. Object and Spatial Vision Recognizing objects visually is a difficult problem for the following reason: the data upon which we must base our identification decision is a retinal projection that can vary dra- matically depending on our viewpoint relative to the object (not to mention the lighting, where we are looking, whether we or the object are moving, and so forth; we will ignore these complications because we have enough trouble as it is). Even something as simple as a cube can project a nearly infinite variety of retinal images depending on its orientation and its distance from us. So there is a highly underdetermined many-to-one mapping of retinal images onto the object’s true shape. Obviously the visual system has found a way to solve this problem with staggering efficiency and accuracy. The question is: how? There are two major routes one can take in approaching this problem. They can be broadly classified as view-based_.or part-based models. View-based (also called image-based) mod- els assume that internal representations of whole objects as they would appear to the ob- server are stored as such in memory. One approach to dealing with the many-to-one prob- lem discussed in the last paragraph is to store in memory, for every object one might wish eventually to recognize, all the possible retinal images that object could produce (or at least the images one is likely to encounter under normal circumstances, which reduces the number of possibilities from infinite to merely unimaginably large). This is the idea behind simple template matching accounts of object recognition. There are, of course, many prob- lems with this approach. First, it would require a very large number of stored templates: more than one could plausibly suppose are stored in the brain. New instances of a familiar category (a strange but recognizable chair, for instance) could not be recognized by this mechanism. Finally, there is no evidence for the sort of unguided search through a huge database of templates that would be required by this mechanism. So this simple brute- force account of object recognition will not work. A more sophisticated version of template matching, however, does have considerable merit. According to this ide&, there exist internal representations of objects depicted in one or, more often, several canonical views (that is, views that permit one to represent most of the critical parts of the object from a view that is often encountered in real life). The reason multiple views are necessary is because any one view usually will not contain information about the part of the object facing away from the viewpoint, and of course we know per- fectly well what the back of a chair looks like even when we can’t see it (or at least we can make a very good guess). When an object is to be recognized, it is first normalized, that is, it is transformed so as to match the size, location, and viewpoint that is likely to be stored for that object, and this normalized representation is then compared to stored object repre- sentations. This mechanism avoids some of the problems of the simple template matching models by allowing for normalization so that only a small number of views of each object need be stored. A very different approach to object recognition is offered by feature- or part-based mod- els of object recognition. According to these theories, object representations consist of lists of features (e.g., oriented lines, angles, the presence or absence of symmetry) or parts (e.g., volumetric primitives such as cylinders), together with some specification of the relations among the parts. These accounts deal with the many-to-one problem by virtue of the fact that they are inherently more abstract than view-based theories: they transform the retinal image into a set of symbolic descriptions that do not depend on rotation, size, and so forth. One major subclass of part-based accounts are structural description theories. According
10 ® Visual Perception to these theories (advocated by Marr and Nishihara, 1978, among others), stored object representations consist of 3D models of objects made up of a set of volumetric part primi- tives such as cylinders, cones, cubes, etc. that are arranged according to the structure of the object itself. The representation is not viewpoint dependent. They are called descriptions because they are not depictive (as in the multiple view plus transformation models). They are abstract and consist of a list of the parts together with a description of the structural relations among the parts (e.g., “cone on top of cylinder”). These models completely by- pass the problem of too many representations by using fairly abstract representations that are insensitive to such things as absolute size or viewpoint. As long as the parts and their relations can be recognized, a match to a stored representation is possible. Biederman (Reading 20) describes a detailed model of this type, and makes the case for it on both theoretical and empirical grounds. These alternative accounts of object recognition are fundamentally different in certain respects, and one might think that if one is the “correct” theory, the other must be wrong. However, it is possible, even likely, that both kinds of theories are correct to some extent (in a way analogous to the trichromatic and opponent-process theories of color vision), and that the challenge will be to specify the conditions under which each class of theories best accounts for recognition performance. Attention and Awareness To many people, attention and awareness are one in the same. We are aware of that to which we attend. Ever since Helmholtz and James first discussed these concepts, however, they have been treated differently by the scientific community. A look at the history of investigation of the two phenomena reveals a long and continuous line of scientific inves- tigation into attention. In contrast, awareness (and its related term, consciousness), while of sustained interest in psychology for the past century, has until recently been ignored by neuroscience. In the last 10 years, however, that imbalance has changed, and there has been a surge of interest in the neural substrates of visual awareness. Attention Attention is a term with a high potential for confusion, because it can mean different things to different people. One often hears the comment at scientific meetings that the term is meaningless because it has too many meanings. Perhaps the most common use of the term, and the meaning intended here, is that of perceptual selectivity, described below. Other meanings include vigilance, or the ability to focus for extended periods in anticipation of a signal (as when the operator of a power plant monitors for a particular warning light), and multi-tasking, or the ability to switch between two overlapping tasks, like driving and carrying on a conversation. Attention as perceptual selectivity refers to the problem first brought to prominence by William James at the end of the 19th century: we are confronted with far too much sensory information to grasp coherently at one time. We must select what is relevant and ignore what is irrelevant (which will tend to be almost everything). That means that until you started reading this sentence, you were unaware of the pressure of the chair you are sitting on against your skin, but now, with only a shift in attention to that source of sensory input, you are aware of it. Or of the faint sound of a vehicle out the window. And so forth. These kinds of attentional shifts are purely top-down and goal-driven (that is, you presumably set up, at least informally, a perceptual goal to give priority to certain tactile or auditory inputs and to ignore other inputs that might have been of higher priority previously, and this goal
An Overview m II serves somehow to modulate the relative strengths of the sensory inputs available to you). Of course, the stimulus itself may well be capable of modifying your attentive state—an unexpectedly loud handclap will capture your attention in a purely stimulus-driven fash- ion. In both instances, however, the result is that certain sensory inputs are selected and enter awareness, and others are excluded. As implied by the last sentence, the gatekeeping act of selection and its result (to permit sensory information to enter awareness) are dis- tinct; hence attention and awareness are distinct concepts. But why is there a need for attention? What is the limitation that requires us to select some objects and ignore others? Broadbent (1958), in his seminal book that brought the study of attention into the modern era, suggested simply that people are information chan- nels in the sense that we transmit sensory information into a behavioral response, and information theory states that any information channel must have an upper bound on its capacity. A more concrete answer was suggested by Treisman and Gelade (Reading 21), who were inspired by the finding of functional specialization by neurophysiologists in the 1960s and 70s. They pointed out that if visual objects are represented in terms of separate “feature maps,” one for color, another for orientation, and so forth, then at some point there must come a stage of reintegrating the separate features into a unified and coherent repre- sentation of the object we actually see. They proposed that the process of feature integra- tion was carried out by attending to one location at a time (and ignoring all other loca- tions), so that only the attended features there were strongly represented at that moment. This constitutes attentional selection, and results in feature integration. This was a groundbreaking idea, and it inspired a generation of experimental psycholo- gists to explore its consequences. The problem of feature integration came to be called the binding problem and it continues to be a subject of considerable interest 20 years later (see Roskies et al., 1999, and the introduction to Reading 21). Many of the details of Feature Integration Theory have been updated as a result of subsequent empirical work, but the basic idea remains nearly intact. Investigation of attention over the last decade or so has included a great deal of new work on the neural basis of attention inspired by Moran and Desimone (Reading 22), to- gether with advances in understanding how attention is controlled. This latter problem is a major challenge now, and begins to approach the problem of consciousness because it requires one to think about how goals are specified, and how acts of the will are imple- mented. Descartes proposed that communication between the Soul (where acts of the will originate) and the Body is mediated by the pineal gland. Today this idea is viewed as a quaint anachronism, and other approaches are being considered, including the notion that a working memory store that is probably located in the dorsolateral prefrontal cortex con- tains representations of goals that can cause feedback signals to sensory regions that in turn can modulate the strength of neural representations there. This remains an active area of investigation that is really just beginning. Awareness If attention is perceptual selection, then what is consciousness? Consciousness is generally broken down into smaller problems, some of which will be harder to solve than others. Philosophers (e.g., Chalmers, 1996) have identified the “hard problem” of consciousness: the problem of qualia. How do we experience the redness of red? As Newton said, there is nothing in the light that is what we would call red; there is only a wavelength that happens to have the ability to evoke a pattern of neural response in the three classes of retinal cones. How does that pattern get translated into a subjective experience like red? Nobody has any idea (some would say this is a fundamentally intractable problem), and so this will be left to the philosophers for now. But the problem of awareness, that is, of
12 m Visual Perception how certain kinds of neural activity correspond to something we experience and can report verbally or otherwise, seems more tractable (see Crick and Koch, 1998, for a discussion). The question can be stated as follows: some of the neural activity in the brain correlates well with our visual experience, that is, with what we are seeing at any given moment, and much of the neural activity in the brain is unrelated to what we are seeing. What is the difference between the parts of the brain (cells, layers, areas) that are and are not correlated with visual awareness. What is it about the type of cell (pyramidal?), the neurotransmitters it uses (serotonin? dopamine?), its connectivity pattern, its location in cortex (layer 4?; area V1?), some combination of these, or something else that endows it with the power to cause visual awareness? This is a new area of investigation; little is known. Investigators have had to be clever to devise experiments that will be revealing in this regard. Sheinberg and Logothetis (Read- ing 25) is a good recent example. Methods in Vision Science I conclude with a discussion of how different methods have contributed to advances in vision science. As in all sciences, theoretical and empirical advances go hand-in-hand with technological developments. The development of the telescope permitted astronomers to see celestial objects previously unimagined. The microscope provided biologists with a tool to investigate cellular phenomena at the foundation of that science. So it is with any science. Visual perception has been no exception. Psychophysics and Behavior The first methodological developments in perception research were behavioral. In his 1860 treatise Elements of Psychophysics, Fechner described a collection of techniques for mea- suring with precision the relation between physical stimulus magnitude (e.g., the bright- ness of a light) and perceptual experience by collecting behavioral responses from observ- ers in perceptual experiments (e.g., stating whether or not a dim flash was seen). These methods have withstood the test of time and continue to be used (in somewhat modified form) to this day (see Readings 6 and 8 for good examples). Psychophysical experiments provide the very bedrock of vision science, for they provide the precise measurements that are needed of the perceptual phenomena that require explanation. The criterion by which the usefulness and value of any psychophysical technique can be judged is whether that technique provides a measurement of experience that is as direct as possible and free of subjectivity and bias. Observers often have implicit theories of how vision works, and if these are permitted to contaminate psychophysical measurements, they are likely to lead to improper inferences. The development of signal detection theory (Reading 3 by Tanner and Swets) was inspired by precisely this concern. Further developments during the 20th century gave rise to psychophysical methods that allowed for more precise control of the stimulus than those available to Fechner and his contemporaries. For example, the development of the tachistoscope provided a means of presenting one or more visual stimuli of precisely known brightness and duration to an observer. Such a device was essential for investigating temporal phenomena such as visual masking, where two objects are presented in rapid succession and the effect of one on the visibility of the other is measured. Many other devices were invented to serve a huge vari- ety of special purposes in the investigation of visual function. In the 1960s and 70s, digital computers entered the laboratories of vision scientists and eventually replaced the tachistoscope and other special-purpose devices. The reason is simple: general-purpose computers permit almost infinite flexibility in creating and pre-
An Overview m 13 senting stimuli, and largely (but not completely) did away with the necessity to invent and build a new apparatus for each new experimental question, greatly accelerating research efforts. Neuropsychology All of the developments described in the previous paragraphs contributed to characterizing how variation in the properties of physical stimuli caused changes in visibility or discriminability as measured by behavioral methods. In parallel with these, a multitude of methods to investigate the neural basis of visual perception emerged. The earliest modern approach to brain-behavior relations was the proposal by Franz Joseph Gall that the brain consists of a number of functionally distinct compartments each of which was responsible for a particular kind of behavior (including such faculties as parental love, hope, and de- structiveness). He argued that when a given faculty was exercised, the corresponding part of the brain grew larger (just as a muscle, when exercised, grows larger), and that this in turn was reflected in bumps on the skull, which grew to accommodate the brain tissue beneath them. Gall attempted to assign function to different brain areas by correlating personality traits to bumps on the head, a field of investigation known as phrenology. Pierre Flourens tested these notions by removing parts of the brains of experimental animals, and he concluded that the brain is not functionally compartmentalized at all, but is instead functionally undifferentiated. According to Flourens, any part of the brain can carry out any function as well as any other. Although Flourens’ conclusion has the distinct merit that it opposes the highly implau- sible theory of Gall, it was soon shown to be wrong. In 1861, Pierre Paul Broca (1824— 1880) reported his observations of a patient who had suffered damage to his left frontal lobe (discovered after autopsy upon the patient’s death) that had prevented him from speak- ing, while sparing most other cognitive functions, including the ability to understand speech. Broca asserted that the frontal lobes contained the faculty of speech production. This idea of functional localization was met with much skepticism at the time (a continued reaction against Gall), but it was corroborated with subsequent observations. For example, in 1876, Carl Wernicke described a patient who had difficulty understanding speech but not in pro- ducing it, a pattern complementary to that reported by Broca. Wernicke found that this patient had a lesion in the posterior temporal lobe, now called Wernicke’s area. This double dissociation provided strong evidence that the perception and production of speech are carried out by functionally distinct brain regions. Among the first evidence for functional specialization in vision was reported by the French ophthalmologist Louis Verrey who, in 1888, described his observations of a patient who had suffered a stroke in her left visual cortex. Among other less dramatic effects, the patient lost her ability to see color in the right visual field—all the objects there were seen in shades of gray. Color vision was normal in the left visual field. The acquired loss of color vision due to cortical damage (in this case, stroke) is known as achromatopsia. The existence of this syndrome strongly suggests that there is a local brain region that is neces- sary for the experience of color. It is important to keep in mind that other parts of the visual system are also necessary for normal color vision (for example, the cone photoreceptors in the retina are certainly necessary for color vision!), but this brain region appears to be specialized for color because damage there does not much affect other visual abilities, such as form or motion perception. Inspired by the findings of the neurologists, Korbinian Brodmann analyzed the anatomy of the human brain around the turn of the century using the method of cytoarchitectonics, which relies on differences in the form and structure of cortical neurons. He identified 52 distinct areas and suggested that these areas may well be functionally distinct as well. Brodmann’s classification scheme continues in wide use to- day.
14m Visual Perception Functional specialization in the brain is now as close to a fact as anything in science can be. Brain-damaged patients have been described with specific deficits in motion percep- tion, color vision, the perception of faces or objects, and in the ability to attend to specific regions of space, among many other syndromes, suggesting that each of these functions may rely selectively on the damaged region of the brain. Lissauer (Reading 16) provided an insightful early analysis of a patient with cortical damage that caused not blindness, but an inability to recognize objects that could nevertheless be seen, a condition known as visual agnosia. The blindsight patient of Weiskrantz et al. (Reading 24) revealed a surpris- ing dissociation between explicit visual awareness and measured visual performance. Subtle deficits such as these have turned out to be extremely informative in developing accounts of how the normal visual system works. Neurophysiology Vision is an emergent property of neural activity in the brain. Acomplete understanding of visual perception, therefore, will require an understanding of how neurons subserve vi- sion. Of course, there is much to learn by analyzing visual perception through purely be- havioral means using the methods of visual psychophysics and cognitive neuropsychol- ogy. But despite its almost infinite complexity, the brain is a concrete medium in which visual behavior is implemented and it can therefore provide important clues about the principles of visual perception. The development of the microscope provided physiologists with a crucial tool that led to the idea, first promoted by Santiago Ramon y Cajal, that the brain consists not of continu- ous and undifferentiated cellular tissue but of billions of separate nerve cells that commu- nicate with one another at specialized junctions called synapses. Johannes Miller, Hermann von Helmholtz, and others learned that axons carry electrical signals from the cell body to the synaptic gap, which can in turn activate the next cell. Neurons are exquisitely complex biological devices whose functional properties are a result of a multitude of molecular, chemical, and electrical forces acting in concert. Mod- ern neuroscience is dedicated to uncovering how the interactions among these forces gives rise to patterns of activity in single neurons. However, for purposes of understanding how neural activity produces visual function, one can view a neuron as a device that produces a discrete signal (an action potential that can be measured with a microelectrode) in response to light (in the case of the photoreceptors) or to the release of neurotransmitters at the synapse (in the case of all other neurons). Cellular neurophysiologists place electrodes in or near the axon of a neuron and record action potentials as they occur. The rate with which action potentials are produced by the cell provide a compact measure of the activity of that cell, and this in turn can reveal how that cell participates in any given perceptual act. For example, as suggested by Barlow (Reading 2), one can characterize a cell’s function by determining what sorts of visual stimuli cause that cell to fire most effectively. The neuro- physiological studies of Hubel and Weisel (Reading 7), Newsome et al. (Reading 10), Moran and Desimone (Reading 22), and Scheinberg and Logothetis (Reading 25) each provides an excellent example of the insights to be gained from single-cell recording com- bined with carefully designed stimuli and behavioral tasks. Functional Neuroimaging Until the 1970s, direct measurements of brain function in behaving organisms was only possible in two ways. First, one could analyze patterns of impaired performance in brain- damaged patients, relying on “experiments of nature” such as a stroke or closed-head in- jury to reveal what the intact system must be like in order that it could be damaged in such a way as to have produced the observed functional deficit (e.g., damage causing cortical
An Overview m 15 achromatopsia but not affecting the perception of shape or motion suggested that there exists a distinct “color center” in the brain). The main disadvantage here is that true experi- ments are not an option (one had to seek out patients that happened to have had a stroke in a focal region that happened to correspond to a functional area). Second, one could per- form experiments using animals, recording from single cells or producing focal lesions in cats or monkeys. The disadvantages of the single-cell approach is that although single cell responses are an extremely rich source of data, the evidence is quite local so that global cortical circuits are difficult to study without great effort. The difficulty with lesions is that the interpretation of a pattern of impaired performance is not always unambiguous. Fur- thermore, there are limitations in what tasks monkeys can be trained to do (for example, language tasks are not available). Finally, the anatomy and function of monkey and cat brain differ from those of human brain in both known and unknown ways. Over the last several decades, neuroscientists perfected the use of the electroencephalo- gram (or EEG) to measure electrical field potentials at the scalp that are caused by the activity of millions of neurons in cortex several millimeters below the electrode’s scalp location. This technique provides a highly detailed record of the timecourse of brain activ- ity, and has been used very effectively (e.g., Hillyard et al., 1996). The data one obtains from this technique are similar in many respects to single cell recording in monkeys, en- joying the same very high temporal resolution of that technique. It has several advantages over single cell recording: it can be carried out in humans, and it provides measurements over the entire surface of the scalp, potentially overcoming the inherently local measure- ment that emerges from single cell recording. However, the event-related potential suffers from its comparatively low and sometimes uncertain spatial resolution. Magnetoencephalography (or MEG) has recently emerged as a new technology that may one day replace the EEG. It operates on principles that are analogous to the EEG, but instead of measuring electrical potentials, it measures extremely small magnetic field in- homogeneities produces by the electrical activity of neurons in cortex. This technique ar- guably has better spatial resolution than EEG, and similar temporal resolution. It also can provide a map of activity throughout the brain. One of its drawbacks is that it is currently a fairly expensive technology that is available at only a few sites. This is perhaps the new- est of the brain-imaging technologies and therefore the least is known about it. Finally, two techniques that have emerged over the last two decades have generated great interest although their full potential as scientific tools to investigate vision has not yet been reached. Both methods are based on changes in blood flow or in the composition of oxygenated or deoxygenated hemoglobin within the brain that is caused by changes in neural activity. In rough outline, when neurons produce action potentials, they require oxygen and other nutrients that are carried by the blood. They signal their need for blood, and this in turn causes changes in blood flow in those regions. Thus both of these methods are indirect, although surprisingly precise, measures of the location of neural activity. The first method is positron emission tomography or PET, which was used by Zeki et al. (Reading 9). It requires an injection of a radioactively labeled agent (usually oxygen in water) that is carried by the blood to the brain. The PET scanner can detect the location and concentration of the radioactive agent, and thereby pinpoint the region of neural activity. A major advantage of PET over EEG is its improved spatial precision. The disadvantages of PET are the invasive use of a radioactive substance, which is potentially harmful, and the very low temporal resolution (on the order of several minutes). The second of the two blood-based methods is functional magnetic resonance imaging, or fMRI, used by O’Craven et al. (Reading 23). It exploits the fact that deoxygenated hemoglobin is paramagnetic and can be detected by an appropriately programmed MRI scanner (see Haxby et al., 1998, fora tutorial review). This is relevant because the concen- tration of deoxygenated hemoglobin changes systematically where there is neural activity. fMRI has better spatial and much better temporal resolution than PET (on the order of a
16 # Visual Perception few seconds) and it does not require an invasive injection. It also has better spatial resolution than EEG. However, {MRI has much poorer temporal resolution than EEG or MEG, and it is currently quite expensive to carry out a study. On the other hand, most modern medical cen- ters own MRI scanners for clinical purposes, and so they are often more readily available than MEG scanners. } The techniques for studying vision, ranging from behavior to brain imaging, are comple- mentary rather than competing. Each represents a tradeoff between cost, spatial and temporal resolution, invasiveness, and other factors. Each requires a unique approach to designing ex- periments, and the range of questions that can be addressed differs markedly across the meth- ods. The only generalization that can be made is that vision science can only benefit from the development of new methods and new technologies that will allow new questions to be ad- dressed and old questions to be re-examined in new ways. About this Volume The articles selected for this volume are among the most influential writings on visual percep- tion published in the last 125 years. These 25 articles were selected from an initial list of more than 200 classic articles culled from many sources. In the end, the selection reflects a need to provide reasonably complete coverage of the major themes and methods in visual perception, and a desire to include articles that are accessible by non-experts. Also included are some articles from the last decade to reflect some recent trends in the field (e.g., the use of func- tional magnetic resonance imaging to measure human brain activity in vision). Obviously many important topics and many influential works have been omitted. It would not be diffi- cult to compile three more volumes of readings of equivalent quality and impact. Each of the five sections concludes with a set of questions designed to stimulate discussion of the articles contained in that section. In addition, several additional articles are listed that could easily have been included in this volume if only there had been room. These, together with the references following each of the article introductions, constitute a larger set of essen- tial readings in visual perception that readers are urged to seek out on their own. In order to include as many articles as are included here, it was necessary to reduce the length of some articles through editing. In each case where this was done, the omitted mate- rial included experiments that, while they contributed importantly to the theoretical point that was being made, were not essential to the reader’s understanding of that point. I benefitted greatly from the advice of many colleagues during the process of selection. Stephen Palmer provided an exceptionally detailed and scholarly evaluation of an early list of proposed articles for which I am grateful. Richard Abrams, Jim Enns, Bruce Goldstein, Patrick Green, Glyn Humphreys, Eileen Kowler, William Newsome, Harvey Schiffman, and Nick Wade offered valuable feedback and ideas for readings I hadn’t considered. My colleagues at Johns Hopkins — Ed Connor, Howard Egeth, Mike McCloskey, Brenda Rapp, and Trish Van Zandt — were helpful and patient in our conversations over the months I was engaged in this project. To all I offer my thanks. REFERENCES Broadbent, D. E. (1958). Perception and communication. London: Pergamon Press. Chalmers, D. J.(1996). The conscious mind: In search of a fundamental theory. New York: Oxford University Press. Crick, F., & Koch, C. (1998). Consciousness and neuroscience. Cerebral Cortex, 8, 97-107. Fechner, G. T. (1860). Elemente des psychophysik. Leipzig: Breitkopf and Hartel. (Translated by H. E. Adler, Elements of psychophysics. New York: Holt, Rinehart, & Winston, 1966) Haxby, J. V., Courtney, S. M., & Clark, V. P. (1998). Func- tional magnetic resonance imaging and the study of atten- tion. In R. Parasuraman (Ed.), The attentive brain (pp. 123— 142). Cambridge, MA: MIT Press. Hillyard, S.A., Anllo-Vento, L., Clark, V. P., Heinze, H. J., Luck, S. J., & Mangun, G. R. (1996). Neuroimaging ap-
proaches to the study of visual attention: A tutorial. In A. F. Kramer, M. G. H. Coles, & G. D. Logan (Eds.). Converg- ing operations in the study of visual selective attention (pp. 107-138). Washington, DC: American Psychological As- sociation. Kanizsa, G. (1979). Organization in vision: Essays on Ge- stalt perception. New York: Praeger. Kellman, P. J., & Shipley, T. F. (1991). A theory of visual interpolation in object perception. Cognitive Psychology, 23, 141-221. Marr, D., & Nishihara, H. K. (1978). Representation and rec- ognition of the spatial organization of three-dimensional shapes. Proceedings of the Royal Society ofLondon B, 200, 269-294, McBeath, M. K., Shaffer, D. M., & Kaiser, M. K. (1995). How baseball outfielders determine where to run to catch fly balls. Science 268, 569-573. Michotte, A., Thinés, G., & Crabbé, G. (1991). Amodal suggested Readings | An Overview m 17 completion of perceptual structures (E. Miles & T.R. Miles, Trans.). In G. Thinés, A. Costall, & G. Butterworth (Eds.), Michotte’s experimental phenomenology ofperception (pp. 140-167). Hillsdale, NJ: Erlbaum. (Original work published 1964) Nakayama, K., He., Z. J., & Shimojo, S. (1995). Visual sur- face representation: A critical link between lower-level and higher-level vision. In S. M. Kosslyn & D. N. Osherson (Eds.), An invitation to cognitive science. Vol. 2: Visual Cognition (pp. 1-70). Cambridge, MA: MIT Press. Roskies, A. L., et al. (1999) The binding problem. Neuron, 24, 7-125. Tanaka, K. (1996). Inferotemporal cortex and object vision. Annual Review of Neuroscience, 19, 109-139. von der Heydt, R., Peterhans, E., & Baumgartner, G. (1984). Illusory contours and cortical neuron responses. Science, 224, 1260-1261. Zeki, S. (1993). A vision of the brain. Oxford: Blackwell. The following books provide excellent overviews of vision science from various thematic and methodological perspectives. Farah, M. J. (2000). The cognitive neuroscience ofvision. Malden, MA: Blackwell Publish- ers, Inc. Gordon, LE. (1997). Theories of visual perception (2nd Ed.). New York: John Wiley & Sons. Gregory, R. L. (1970). The intelligent eye. New York: McGraw-Hill. Hubel, D. H. (1988). Eye, brain, and vision. New York: W. H. Freeman & Co. Marr, D. (1982). Vision. San Francisco: W. H. Freeman & Co. Palmer, S. E. (1999). Vision science: Photons to phenomenology. Cambridge, MA: MIT Press. Rock, I. (1984). Perception., New York: Scientific American Books. Zeki, S. (1993). A vision of the brain. Oxford: Blackwell.
ii om woer Pb ie : ott hh colt Sica “pe hi Iepbs ; ; Pope ss irs Lee Wes ie « . | A) north Aa Aa 'sirheae, Vinee noe aa Leadbatliny wire Rantidlby Torgeen Gee praca tee a siete wea _ ae ’ ye Cap ‘nda — don leper? § MS.Ee ure = nil J S antipier “A co. a tie wag Lvond raver inroe 7" ‘ ely spy ‘ i : : wate a VAP IA ST aes cane 10 He ze bus wan 6 MOP ieting D4 vb, FE andy WEUICpre agers nate) ay Be a> “ASotlheLapeantitagea>’ ors. Ep fA SEQUEN BY penta Ansa) gate =| at OO Lake Be gtate ite: cote hate ‘ ny = eee seta bagi ae pe in a A a apart 5 wine i~ tae mr WPA 2), per teh ue ie saese (ety Leslidenwtiiadl Jy 1) cere Gr hacen aimete Bd ae ee : ern will gaonn dite “a eeedGhD §nares An GO?atwia CMe ots £78 : sain ee ed 4 ey Ae ; iMitatety BAT: Tinie aia Lh hgh rckalt @ ynvaly & Atv I) 1 ss tenet a LfetHD : ; ‘ ya Tapert dire Vala tytareomtcag det ape intheiad ig ;Dwll et ta rent SSS year. These 2S egies peg ‘nigh is Tae = <t Thou 2h a) ied Sneen ans NY IOs rh ify (crete « ratty “amg? of Catia ibeines weg Te oitertiadt Coppa) ee vewmsiere Se TKR “ailsaes, hyts : Dad & Pale 4A* feat sas Pa sas inal iaantaalt! ¢Ati. cues sits ardenHORA ny icp u LK : oy eal Udy whigre rad Dior Peeve: Green CMW padyaadabed, --y 8 4 “3£0) sotaensG Sty alah Aas E- a es was featths Warha aot One bei te. ; c Ped Lary gt Bt. std Leas ae 0G 4 aan Moho wetupneia SAB rye ra af With tc. 2 ely 59 Yh AMY poet pac it TIM LAA yapsaresios “) Topi PAL sph ethos torn pom ~. ie? wp many ox UG LU oere, 1 vee Legeds oP ics 2 igMange)5 Ugieio® BTavs vai Wyn MA} eh, Wy Rey ia dass) thee yy : veres Conds} - xe, Spas he vader s anhkey peters om cht pg | ical ice of tne) ioaget drink,the ema Sey, 1 SB WPRMIEAL ii Weil asterwarly ¢ Bak rea rt Wyss : 2 Rn, Pita dere Ji Ba) Og (in ‘ j + niles ee wah (Seer nestled oe ty Wate © — we ae s oy a | ancadeaesl. ge * Lotuye 3 ‘ pe Vers, ihr eta eekosDac Rap “be é : _ a id ait ise. ube iacen f
Theoretical Perspectives
x i Nie
Introduction toReading | oq Hermann von Helmholtz (1821-1894) was responsible for half a dozen truly seminal contributions to science, any one of which might have been worthy of a Nobel Prize. This includes the trichromatic theory of color vision (which he discovered independently of Thomas Young; see the introduction to Reading 7 by Hurvich and Jamison for more details), the place theory of pitch perception (which was corroborated a century later by the Nobel Prize-winning work of Georg von Békesy), and the first measurement of the conduction velocity of a neural signal (about 100 meters/sec). Among his most influential contributions is the monumental Treatise on Physiological Optics, in which he analyzes the functioning of the visual system at many levels in exquisite detail. This selection is‘taken from Volume Ill of that work, subtitled The Perceptions of Vision. It contains one of Helmholtz’s most enduring theoretical contributions: his articulation of the constructivist perspective (what he terms the empirical theory) in perception. According to this view, perception is not merely an “internal picture” of the retinal image; instead, it is a joint result of sensory input on the one hand and what the observer knows and expects on the other. Each of these provide evidence about what is actually present in the world, and a form of perceptual thought that Helmholtz called unconscious inference gives rise to veridical experiences despite the limitations and distortions of our senses. As Helmholtz puts it, “such objects are always imagined as being present in the field of vision as would have to be there in order to produce the same impression on the nervous mechanism, the eyes being used under ordinary normal conditions” (p. 25). The alternative to the empirical theory Helmholtz calls the intuition theory, according to which perception does not depend at all on experience or expectation, but is wholly innate Pil
22 Visual Perception and stimulus-driven, He argues persuasively at various points in this selection against this position. He notes, for example, that one’s ability to navigate about a room in twilight is dramatically improved if one has had previous experience viewing the layout of the room under good illumination. One’s memory of the layout, combined with the indistinct impressions of the dimly illuminated furniture, combine to produce an accurate perceptual experience. A standard example of unconscious inference is its manifestation in size constancy. The problem solved by size constancy is this: the retinal image size of an object depends directly on the distance between it and the observer: the closer the observer, the larger the retinal image size. Indeed, any particular retinal image is consistent with an infinite number of possible size/ distance combinations (small/near or large/far). Yet observers routinely make correct judgements about the true size of objects. This fact of size constancy implies that size judgements cannot be based only on the retinal image, but must also take into account the perceived distance to the object (see Reading 14 by Kaufman and Rock for more on size constancy). Helmholtz’s explanation for this and many other phenomena is that the observer unconsciously combines the retinal image size with an estimate of the distance to the object, based perhaps on stereoscopic or other depth cues. Helmholtz discusses some of the ideas behind the notion of unconscious inference (or, as translated here, unconscious conclusion) that were later to have such a significant impact on the field and continue to hold sway today. He uses as an initial example the effects accompa- nying mechanical stimulation of the retina (e.g., a finger press on the outer edge of the eyeball). The pressure activates a small set of photoreceptors, producing what are called “phosphenes.” The perceptual experience is not of an event occurring at the point of stimulation, however, but of light coming from the direction of the bridge of the nose; such a light would, if present, have stimulated the part of the retina that is being mechanically stimulated. The brain “concludes” that there must be light in the scene in a location that would stimulate the active part of the retina. In addition to laying out the principle of unconscious inference, Helmholtz brings to the argument examples from his own observations of a wide range of perceptual phenomena, anticipating by more than a century some issue of great contemporary interest. These include the “ohantom limb” syndrome in which an amputee experi- ences sensations (often pain) that appear to come from the limb that is no longer there (é.g., Ramachandran, 1993): the perception of biological motion (e.g., Johansson, 1973); and filling in of the blind spot (the location on the retina where the optic nerve leaves the eye and where, therefore, there are no photoreceptors; e.g., Ramachandran, 1993). He dismisses the notion of isomorphic brain representations, and what Dennett (1991) referred to as the “Cartesian theater” a century later: “Now |ask, what similarity can be imagined between the process in the brain that is concomitant with the idea of a table and the table itself? Is the form of the table to be supposed to be outlined by electric Currents? . .. such an electrical reproduction of the table in the brain would be simply another bodily object to be perceived, but no idea of the table” (p. 35). It should be emphasized that in Helmholtz’s view, unconscious inference occurs not only when there are several distinct sources of information that must be combined in an algorithmic way (as in the size constancy example); it plays a role in virtually every percept. A distinction is drawn between the sensation, that pure sensory signal that is not affected by experience or
expectation, and the perception, which is an “idea” of the objects the scene must contain. Sensations themselves are difficult to experience because they do not serve any useful purpose, other than to provide evidence for perceptual conclusions (“In the ordinary affairs of life the sensations have no other importance for us. Subjective sensations are of interest chiefly for scientific investiga- tions only,” p. 27). One must make a deliberate effort to experience sensations themselves. J. S, Bruner (1957) proposed a theory in this tradition that also explicitly implicates perceptual hypotheses in vision; Bruner’s contribution was part of what came to be termed the “New Look” in perception. REFERENCES Bruner, J. S. (1957). On perceptual readiness. Psychological Review, 64, 123-152. Dennett, D. (1991). Consciousness explained. Boston: Little, Brown. Gregory, R. (1980). Perceptions as hypotheses. Philosophi- cal Transactions of the Royal Society of London B, 290, 181-197. Johansson (1973). Visual perception of biological motion and Introduction to Reading 1 m 23 For example, when observers viewed briefly-flashed playing cards that occasionally were colored incorrectly (e.g., a black ten of diamonds), they would sometimes report the color as purple, as iftheir knowledge of the correct color (a perceptual hypothesis of red) was somehow combined with the sensory evidence (black) to yield a sort of perceptual compromise. Other modern constructivists include Irvin Rock (e.g., 1983) and Richard Gregory (e.g., 1974). The constructivist position is rarely discussed in the contemporary literature, however. This is because it is now part of the set of commonly held foundational assumptions of almost all vision scientists. a model for its analysis. Perception & Psychophysics, 14, 201-211. Ramachandran, V. S. (1993) Filling the gaps in perception: II. Scotomas and phantom limbs. Current Directions in Psychological Science, 2, 56-65. Rock, I. (1983). The logic of perception. Cambridge, MA: MIT Press.
PEA D EN G4 Concerning thePerceptions inGeneral H. von Helmholtz he sensations aroused by light in the nervous mechanism of vision enable us to form concep- tions as to the existence, form and position of ex- ternal objects. These ideas are called visual per- ceptions. In this third subdivision of Physiological Optics we must try to analyze the scientific re- sults which we have obtained concerning the con- ditions which give rise to visual perceptions. Perceptions of external objects being therefore of the nature of ideas, and ideas themselves being invariably activities of our psychic energy, percep- tions also can only be the result of psychic energy. Accordingly, strictly speaking, the theory of per- ceptions belongs properly in the domain of psy- chology. This is particularly true with respect to the mode of the mental activities in the case of the perceptions and with respect to the determination of their laws. Yet even here there is a wide field of investigation in both physics and physiology, in- asmuch as we have to determine, scientifically as far as possible, what special properties of the physi- cal stimulus and of the physiological stimulation are responsible for the formation of this or that particular idea as to the nature of the external ob- jects perceived. In this part of the subject, there- fore, we shall have to investigate the special prop- erties of the retinal images, muscular sensations, etc., that are concerned in the perception of a defi- nite position of the observed object, not only as to its direction but as to its distance; how the percep- tion of the form of a body of three dimensions depends on certain peculiarities of the images; and under what circumstances it will appear single or double as seen by both eyes, etc. Thus, our main 24 purpose will be simply to investigate the material of sensation whereby we are enabled to form ideas, in those relations that are important for the per- ceptions obtained from them. This problem can be solved entirely by scientific methods. At the same time, we cannot avoid referring to psychic activities and the laws that govern them, as far as they are concerned with the perception of the senses. But the discovery and description of these psychic activities will not be regarded as an es- sential part of our present task, because then we might run the risk of losing our hold of established facts and of not adhering steadily to a method founded on clear, well-recognized principals. Thus, for the present at least, I' think the psychological domain of the physiology of the senses should be kept separate from pure psychology, whose prov- ince really is to establish as far as possible the laws and nature of the processes of the mind. Still we cannot altogether avoid speaking of the mental processes that are active in the sense-per- ceptions, if we wish to see clearly the connection between the phenomena and to arrange the facts in their proper relation to one another. And hence, to prevent any misconception of the plan I have in mind, I intend to devote the latter part of this chap- ter to a discussion of the conclusions which Ithink can be inferred with respect to these mental pro- cesses. And yet we know by experience that people 'In this volume (contrary to the usage adopted in the two previous volumes of the English translation), the editor has deemed it best to retain the more intimate language of the original text, and let the author speak throughout in the first person. (J.P.C.S.)
very seldom come to any agreement as to abstract questions of this nature. The keenest thinkers, phi- losophers like Kant for instance, have long ago analysed these relations correctly and demon- strated them, and yet there is no permanent and general agreement about them among educated people. And, therefore, in the subsequent chapters devoted specially to the theory of the visual per- ceptions, I shall endeavour to avoid all reference to opinions as to mental activity, as involving ques- tions that always have been, and perhaps always will be, subjects of debate between the various metaphysical schools; so as not to distract the reader’s attention from those facts about which an agreement may possibly be reached, by wrangling over abstract propositions that are not necessarily involved in the problem before us. Here I shall merely indicate at the outset cer- tain general characteristics of the mental processes that are active in the sense-perceptions, because they will be constantly encountered in connection with the various subjects to be considered. With- out some previous explanation of their general sig- nificance and wide range of activity, the reader might be apt in some special case to regard them as paradoxical and incredible. The general rule determining the ideas of vi- sion that are formed whenever an impression is made on the eye, with or without the aid of optical instruments, is that such objects are always imag- ined as being present in the field ofvision gs would have to be there in order to produce the same im- pression on the nervous mechanism, the eyes be- ing used under ordinary normal conditions. To employ an illustration which has been mentioned before, suppose that the eyeball is mechanically stimulated at the outer corner of the eye. Then we imagine that we see an appearance of light in front of us somewhere in the direction of the bridge of the nose. Under ordinary conditions of vision, when our eyes are stimulated by light coming from outside, if the region of the retina in the outer cor- ner of the eye is to be stimulated, the light actu- ally has to enter the eye from the direction of the bridge of the nose. Thus, in accordance with the above rule, in a case of this kind we substitute a luminous object at the place mentioned in the field of view, although as a matter of fact the mechani- cal stimulus does not act on the eye from in front of the field of view nor from the nasal side of the Concerning the Perceptions in General m 25 eye, but, on the contrary, is exerted on the outer surface of the eyeball and more from behind. The general validity of the above rule will be shown by many other instances that will appear in the following pages. In the statement of this rule mention is made of the ordinary conditions of vision, when the visual organ is stimulated by light from outside; this out- side light, coming from the opaque objects in its path that were the last to be encountered, and hav- ing reached the eye along rectilinear paths through an uninterrupted layer of air. This is what is meant here by the normal use of the organ of vision, and the justification for using this term is that this mode of stimulation occurs in such an enormous major- ity of cases that all other instances where the paths of the rays of light are altered by reflections or refractions, or in which the stimulations are not produced by external light, may be regarded as rare exceptions. This is because the retina in the fundus of the firm eyeball is almost completely protected from the actions of all other stimuli and is not easily accessible to anything but external light. When a person is in the habit of using an optical instrument and has become accustomed to it, for example, if he is used to wearing spectacles, to a certain extent he learns to interpret the visual images under these changed conditions. Incidentally, the rule given above corresponds to a general characteristic of all sense-perceptions, and not simply to the sense of sight alone. For example, the stimulation of the tactile nerves in the enormous majority of cases is the result of in- fluences that affect the terminal extensions of these nerves in the surface of the skin. It is only under exceptional circumstances that the nerve-stems can be stimulated by more powerful agencies. In ac- cordance with the above rule, therefore, all stimu- lations of cutaneous nerves, even when they af- fect the stem or the nerve-centre itself, are perceived as occurring in the corresponding pe- ripheral surface of the skin. The most remarkable and astonishing cases of illusions of this sort are those in which the peripheral area of this particu- lar portion of the skin is actually no longer in ex- istence, as, for example, in case of a person whose leg has been amputated. For a long time after the operation the patient frequently imagines he has vivid sensations in the foot that has been severed. He feels exactly the places that ache on one toe or
26 Visual Perception the other. Of course, in a case of this sort the stimu- lation can affect only what is left of the stem of the nerve whose fibres formerly terminated in the amputated toes. Usually, it is the end of the nerve in the scar that is stimulated by external pressure or by contraction of the scar tissue. Sometimes at night the sensations in the missing extremity get to be so vivid that the patient has to feel the place to be sure that his limb is actually gone. Thus it happens, that when the modes of stimu- lation of the organs of sense are unusual, incor- rect ideas of objects are apt to be formed; which used to be described, therefore, as illusions of the senses. Obviously, in these cases there is nothing wrong with the activity of the organ of sense and its corresponding nervous mechanism which pro- duces the illusion. Both of them have to act ac- cording to the laws that govern their activity once for all. It is rather simply an illusion in the judg- ment of the material presented to the senses, re- sulting in a false idea of it. The psychic activities that lead us to infer that there in front of us at a certain place there is a certain object of a certain character, are generally not conscious activities, but unconscious ones. In their result they are equivalent to a conclusion, to the extent that the observed action on our senses enables us to form an idea as to the possible cause of this action; although, as a matter of fact, it is invariably simply the nervous stimulations that are perceived directly, that is, the actions, but never the external objects themselves. But what seems to differentiate them from a conclusion, in the or- dinary sense of that word, is that a conclusion is an act of conscious thought. An astronomer, for example, comes to real conscious conclusions of this sort, when he computes the positions of the stars in space, their distances, etc., from the per- spective images he has had of them at various times and as they are seen from different parts of the orbit of the earth. His conclusions are based on a conscious knowledge of the laws of optics. In the ordinary acts of vision this knowledge of optics is lacking. Still it may be permissible to speak of the psychic acts of ordinary perception as unconscious conclusions, thereby making a distinction of some sort between them and the common so-called con- scious conclusions. And while it is true that there has been, and probably always will be, a measure of doubt as to the similarity of the psychic activity in the two cases, there can be no doubt as to the similarity between the results of such unconscious conclusions and those of conscious conclusions. These unconscious conclusions derived from sensation are equivalent in their consequences to the so-called conclusions to the so-called conclu- sions from analogy. Inasmuch as in an overwhelm- ing majority of cases, whenever the parts of the retina in the outer corner of the eye are stimulated, it has been found to be due to external light com- ing into the eye from the direction of the bridge of the nose, the inference we make is that it is so in every new case whenever this part of the retina is stimulated; just as we assert that every single in- dividual now living will die, because all previous experience has shown that all men who were for- merly alive have died. But, moreover, just because they are not free acts of conscious thought, these unconscious con- clusions from analogy are irresistible, and the ef- fect of them cannot be overcome by a better un- derstanding of the real relations. It may be ever so clear how we get an idea of a luminous phenom- enon in the field of vision when pressure is ex- erted on the eye; and yet we cannot get rid of the conviction that this appearance of light is actually there at the given place in the visual field; and we cannot seem to comprehend that there is a lumi- nous phenomenon at the place where the retina is stimulated. It is the same way in case of all the images that we see in optical instruments. On the other hand, there are numerous illustra- tions of fixed and inevitable associations of ideas due to frequent repetition, even when they have no natural connection, but are dependent merely on some conventional arrangement, as, for ex- ample, the connection between the written letters of a word and its sound and meaning. Still to many physiologists and psychologists the connection between the sensation and the conception of the object usually appears to be so rigid and obliga- tory that they are not much disposed to admit that, to a considerable extent at least, it depends on ac- quired experience, that is, on psychic activity. On the contrary, they have endeavoured to find some mechanical mode of origin for this connection through the agency of imaginary organic struc- tures. With regard to this question, all those expe- riences are of much significance which show how the judgment of the senses may be modified by experience and by training derived under various circumstances, and may be adapted to the new conditions. Thus, persons may learn in some mea- sure to utilize details of the sensation which oth-
erwise would escape notice and not contribute to obtaining any idea of the object. On the other hand, too, this new habit may acquire such a hold that when the individual in question is back again in the old original normal state, he may be liable to illusions of the senses. Facts like these show the widespread influence that experience, training and habit have on our perceptions. But how far their influence really does extend, it would perhaps be impossible to say pre- cisely at present. Little enough is definitely known about infants and very young animals, and the in- terpretation of such observations as have been made on them is extremely doubtful. Besides, no one can say that infants are entirely without expe- rience and practice in tactile sensations and bodily movements. Accordingly, the rule givén above has been stated in a form which does not anticipate the decision of this question. It merely expresses what the result is. And so it can be accepted even by those who have entirely different opinions as to the way ideas originate concerning objects in the external world. Another general characteristic property of our sense-perceptions is, that we are not in the habit of observing our sensations accurately, except as they are useful in enabling us to recognize exter- nal objects. One the contrary, we are wont to dis- regard all those parts of the sensations that are of no importance so far as external objects are con- cerned. Thus in most cases some special agsistance and training are needed in order to observe these latter subjective sensations. It might seem that nothing could be easier than to be conscious of one’s own sensations; and yet experience shows that for the discovery of subjective sensations some special talent is needed, such as Purkinje mani- fested in the highest degree; or else it is the result of accident of theoretical speculation. For instance, the phenomena of the blind spot were discovered by Mariotte from theoretical considerations. Simi- larly, in the domain of hearing, I discovered the existence of those combination tones which I have called summation tones. In the great majority of cases, doubtless it was accident that revealed this or that subjective phenomenon to observers who happened to be particularly interested in such matters. It is only when subjective phenomena are so prominent as to interfere with the perception of things, that they attract everybody’s attention. Once the phenomena have been discovered, it is gener- ally easier for others to perceive them also, pro- Concerning the Perceptions in General m 27 vided the proper precautions are taken for observ- ing them, and the attention is concentrated on them. In many cases, however—for example, in the phe- nomena of the blind spot, or in the separation of the overtones and combination tones from the fun- damental tones of musical sounds, etc.—such an intense concentration of attention is required that, even with the help of convenient external appli- ances, many persons are unable to perform the experiments. Even the after-images of bright ob- jects are not perceived by most persons at first except under particularly favourable external con- ditions. It takes much more practice to see the fainter kinds of after-images. A common experi- ence, illustrative of this sort of thing, is for a per- son who has some ocular trouble that impairs his vision to become suddenly aware of the so-called mouches volantes in his visual field, although the causes of this phenomenon have been there in the vitreous humor all his life. Yet now he will be firmly persuaded that these corpuscles have de- veloped as the result of his ocular ailment, although the truth simply is that, owing to his ailment, the patient has been paying more attention to visual phenomena. No doubt, also, there are cases where one eye has gradually become blind, and yet the patient has continued to go about for an indefinite time without noticing it, until he happened one day to close the good eye without closing the other, and so noticed the blindness of that eye.’ When a person’s attention is directed for the first time to the double images in binocular vision, he is usually greatly astonished to think that he had never noticed them before, especially when he re- flects that the only objects he has ever seen single were those few that happened at the moment to be about as far from his eyes as the point of fixation. The great majority of objects, comprising all those that were farther or nearer than this point, were all seen double. Accordingly, the first thing we have to learn is to pay heed to our individual sensations. Ordinarily we do so merely in case of those sensations that enable us to find out about the world around us. In the ordinary affairs of life the sensations have no other importance for us. Subjective sensations are of interest chiefly for scientific investigations only. If they happen to be noticed in the ordinary activ- 2 Nearly everybody has a dominant eye, which governs the other eye; and in which the vision is superior to that in the other eye. But not many persons are aware of the fact. (J.P.C.S.)
28 m Visual Perception ity of the senses, they merely distract the atten- tion. Thus while we may attain an extraordinary degree of delicacy and precision in objective ob- servation, we not only fail to do so in subjective observations, but indeed we acquire the faculty in large measure of overlooking them and of form- ing our opinions of objects independently of them, even when they are so pronounced that they might easily be noticed. The most universal sign by which subjective visual phenomena can be identified appears to be by the way they accompany the movement of the eye over the field of view. Thus, the after-images, the mouches volantes, the blind spot, and the “lu- minous dust” of the dark field all participate in the motions of the eye, and coincide successively with the various stationary objects in the visual field. On the other hand, if the same phenomena recur again invariably at the same places in the visual field, they may be regarded as being objec- tive and as being connected with external bodies. This is the case with contrast phenomena produced by after-images. The same difficulty that we have in observing subjective sensations, that is, sensations aroused by internal causes, occurs also in trying to ana- lyze the compound sensations, invariably excited in the same connection by any simple object, and to resolve them into their separate components. In such cases experience shows us how to recognize a compound aggregate of sensations as being the sign of a simple object. Accustomed to consider the sensation-complex as a connected whole, gen- erally we are not able to perceive the separate parts of it without external help and support. Many il- lustrations of this kind will be seen in the follow- ing pages. For instance the perception of the ap- parent direction of an object from the eye depends on the combination of those sensations by which we estimate the adjustment of the eye, and on be- ing able to distinguish those parts of the retina where light falls from those parts where it does not fall. The perception of the solid form of an object of three dimensions is the result of the com- bination of two different perspective views in the two eyes. The gloss of a surface, which is appar- ently a simple effect, is due to differences of colouring or brightness in the images of it in the two eyes. These facts were ascertained by theory and may be verified by suitable experiments. But usually it is very difficult, if not impossible, to discover them by direct observation and analysis of the sensations alone. Even with sensations that are much more involved and always associated with frequently recurring complex objects, the oftener the same combination recurs, and the more used we have become to regarding the sensation as the normal sign of the real nature of the object, the more difficult it will be to analyze the sensa- tion by observation alone. By way of illustration, it is a familiar experience that the colours of a land- scape come out much more brilliantly and defi- nitely by looking at them with the head on one side or upside down than they do when the head is in the ordinary upright position. In the usual mode of observation all we try to do is to judge correctly the objects as such. We know that at a certain dis- tance green surfaces appear a little different in hue. We get in the habit of overlooking this difference, and learn to identify the altered green of distant meadows and trees with the corresponding colour of nearer objects. In the case of very distant ob- jects like distant ranges of mountains, little of the colour of the body is left to be seen, because it is mainly shrouded in the colour of the illuminated air. This vague blue-grey colour, bordered above by the clear blue of the sky or the red-yellow of the sunset glow, and below by the vivid green of meadows and forests, is very subject to variations by contrast. To us itis the vague and variable colour of distance. The difference in it may, perhaps, be more noticeable sometimes and with some illumi- nations than at other times. But we do not deter- mine its true nature, because it is not ascribed to any definite object. We are simply aware of its variable nature. But the instant we take an unusual position, and look at the landscape with the head under one arm, let us say, or between the legs, it all appears like a flat picture; partly on account of the strange position of the image in the eye, and partly because, as we shall see presently, the bin- ocular judgment of distance becomes less accu- rate. It may even happen that with the head upside down the clouds have the correct perspective, whereas the objects on the earth appear like a paint- ing on a vertical surface, as the clouds in the sky usually do. At the same time the colours lose their associations also with near or far objects, and con- front us now purely in their own peculiar differ- ences.° Then we have no difficulty in recognizing that the vague blue-grey of the far distance may * This explanation in given also by O. N. Rood, Silliman’s Journ., (2) xxxii. 1861. pp.184, 185.
indeed be a fairly saturated violet, and that the green of the vegetation blends imperceptibly through blue-green and blue into this violet, etc. This whole difference seems to me to be due to the fact that the colours have ceased to be distinc- tive signs of objects for us, and are considered merely as being different sensations. Consequently, we take in better their peculiar distinctions with- out being distracted by other considerations. The connection between the sensations and ex- ternal objects may interfere very much with the perception of their simplest relations. A good il- lustration of this is the difficulty about perceiving the double images of binocular vision when they can be regarded as being images of one and the same external object. In the same way we may have similar experi- ences with other kinds of sensations. The sensa- tion of the timbre of a sound, as I have shown else- where,* consists of a series of sensations of its partial tones (fundamental and harmonics); but it is exceedingly difficult to analyze the compound sensation of the sound into these elementary com- ponents. The tactile sensation of wetness is com- posed of that of coldness and that of smoothness of surface. Consequently, on inadvertently touch- ing a cold piece of smooth metal, we often get the impression of having touched something wet. Many other illustrations of this sort might be ad- duced. They all indicate that we are exceedingly well trained in finding out by our sensations the objective nature of the objects around us, but that we are completely unskilled in observing the sen- sations per se and that the practice of associating them with things outside of us actually prevents us from being distinctly conscious of the pure sen- sations. This is true also not merely with respect to quali- tative differences of sensation, but it is likewise true with respect to the perception of space-rela- tions. For example, the spectacle of a person in the act of walking is a familiar sight. We think of this motion as a connected whole, possibly taking note of some of its most conspicuous singularities. But it requires minute attention and a special choice of the point of view to distinguish the up- ward and lateral movements of the body in a person’s gait. We have to pick out points or lines of reference in the background with which we can compare the position of his head. But look through an astronomical telescope at a crowd of people in motion far away. Their images are upside down, Concerning the Perceptions in General m 29 but what a curious jerking and swaying of the body is produced by those who are walking about! Then there is no trouble whatever in noticing the pecu- liar motions of the body and many other singularities of gait; and especially differences between individuals and the reasons for them, sim- ply because this is not the everyday sight to which we are accustomed. On the other hand, when the image is inverted in this way, it is not so easy to tell whether the gait is light or awkward, dignified or graceful, as it was when the image was erect. Consequently, it may often be rather hard to say how much of our apperceptions (Anschauungen) as derived by the sense of sight is due directly to sensation, and how much of them, on the other hand, is due to experience and training. The main point of controversy between various investiga- tors in this territory is connected also with this difficulty. Some are disposed to concede te the influence of experience as much scope as possible, and to derive from it especially all notion of space. This view may be called the empirical theory (empiristische Theorie). Others, of course, are obliged to admit the influence of experience in the case of certain classes of perceptions; still with respect to certain elementary apperceptions that occur uniformly in the case of all observers, they believe it is necessary to assume a system of in- nate apperceptions that are not based on experi- ence, especially with respect to space-relations. In contradistinction to the former view, this may per- haps be called the intuition theory (nativistische Theorie) of the sense-perceptions. In my opinion the following fundamental prin- ciples should be kept in mind in this discussion. Let us restrict the word idea (Vorstellung) to mean the image of visual objects as retained in the memory, without being accompanied by any present sense-impressions; and use the term ap- perception (Anschauung) to mean a perception (Wahrnehmung) when it is accompanied by the sense-impressions in question. The term immedi- ate perception (Perzeption) may then be employed to denote an apperception of this nature in which there is no element whatever that is not the result of direct sensations, that is, an apperception such 4 Helmholtz, Die Lehre von den Tonempfundungen. Braunschweig 1862. (See English translation by A. J. Ellis, entitled On the sensations of tone as a physiological basis for the theory of music. 3rd ed. London and New York, 1895.— JEP.G:S5)
30 @ Visual Perception as might be derived without any recollection of previous experience. Obviously, therefore, one and the same apperception may be accompanied by the corresponding sensations in very different mea- sure. Thus idea and immediate perception may be combined in the apperception in the most differ- ent proportions.° A person in a familiar room which is brightly lighted by the sun gets an apperception that is abun- dantly accompanied by very vivid sensations. In the same room in the evening twilight he will not be able to recognize any objects except the brighter ones, especially the windows. But whatever he does actually recognize will be so intermingled with his recollections of the furniture that he can still move about in the room with safety and lo- cate articles he is trying to find, even when they are only dimly visible. These images would be utterly insufficient to enable him to recognize the objects without some previous acquaintance with them. Finally, he may be in the same room in com- plete darkness, and still be able to find his way about in it without making mistakes, by virtue of the visual impressions formerly obtained. Thus, by continually reducing the material that appeals to the senses, the perceptual-image (Anschauungs- bild) can ultimately be traced back to the pure memory-image (Vorstellungsbild) and may gradu- ally pass into it. In proportion as there is less and less material appeal to the senses, a person’s move- ments will, of course, become more and more un- certain, and his apperception less and less accu- rate. Still there will be no peculiar abrupt transition, but sensation and memory will continually supple- ment each other, only in varying degrees. But even when we look around a room of this sort flooded with sunshine, a little reflection shows us that under these conditions too a large part of our perceptual-image may be due to factors of memory and experience. The fact that we are ac- customed to the perspective distortions of pictures of parallelopipeds and to the form of the shadows they cast has much to do with the estimation of the shape and dimensions of the room, as will be seen hereafter. Looking at the room with one eye shut, we think we see itjust as distinctly and defi- nitely as with both eyes. And yet we should get exactly the same view in case every point in the room were shifted arbitrarily to a different distance from the eye, provided they all remained on the same lines of sight. Thus in a case like this we are really consider- ing an extremely multiplex phenomenon of sense; but still we ascribe a perfectly definite explana- tion to it, and it is by no means easy to realize that the monocular image of such a familiar object nec- essarily means a much more meagre perception than would be obtained with both eyes. Thus too it is often hard to tell whether or not untrained observers inspecting stereoscopic views really notice the peculiar illusion produced by the instru- ment. We see, therefore, how in a case of this kind reminiscences of previous experiences act in con- junction with present sensations to produce a per- ceptual image (Anschauungsbild) which imposes itself on our faculty of perception with overwhelm- ing power, without our being conscious of how much of it is due to memory and how much to present perception. Still more remarkable is the influence of the comprehension of the sensations in certain cases, especially with dim illumination, in which a vi- sual impression may be misunderstood at first, by not knowing how to attribute the correct depth- dimensions; as when a distant light, for example, is taken for a near one, or vice versa. Suddenly it dawns on us what it is, and immediately, under the influence of the correct comprehension, the correct perceptual image also is developed in its full intensity. Then we are unable to revert to the previous imperfect apperception. This is very common especially with compli- cated stereoscopic drawings of forms of crystals and other objects which come out in perfect clear- ness of perception the moment we once succeed in getting the correct impression. Similar experiences have happened to every- body, proving that the elements in the sense-per- ceptions that are derived from experience are just as powerful as those that are derived from present sensations. All observers who have thoroughly investigated the theory of the sense-perceptions, even those who were disposed to allow experience as little scope as possible, have always admitted this. Hence, at all events it must be conceded that, “It is very difficult to find the precise English equivalents for these metaphysical terms, which will prove satisfactory to everybody. And it may not be quite possible to restrict the English word “idea,” for example, to the definition here given. It is doubtful whether the author himself is scrupulously careful throughout the remainder of this work to distinguish these shades of meaning always exactly. (J.P.C.S.)
even in what appears to the adult as being direct apperception of the senses, possibly a number of single factors may be. involved which are really the product of experience; although at the time it is difficult to draw the line between them. Now in my opinion we are justified by our pre- vious experiences in stating that no indubitable present sensation can be abolished and overcome by an act of the intellect; and no matter how clearly we recognize that it has been produced in some anomalous way, still the illusion does not disap- pear by comprehending the process. The attention may be diverted from sensations, particularly if they are feeble and habitual; but in noting those relations in the external world, that are associated with these sensations, we are obliged to observe the sensations themselves. Thus we may be un- mindful of the temperature-sensation of our skin when it is not very keen, or of the contact-sensa- tions produced by our clothing, as long as we are occupied with entirely different matters. But just as soon as we stop to think whether it is warm or cold, we are not in the position to convert the feel- ing of warmth into that of coldness; maybe be- cause we know that it is due to strenuous exertion and not to the temperature of the surrounding air. In the same way the apparition of light when pres- sure is exerted on the eyeball cannot be made to vanish simply by comprehending better the nature of the-process, supposing the attention is directed to the field of vision and not, say, to the ear or the skin. ‘ On the other hand, it may also be that we are not in the position to isolate an impression of sen- sation, because it involves the composite sense- symbol of an external object. However, in this case the correct comprehension of the object shows that the sensation in question has been perceived and used by the consciousness. My conclusion is, that nothing in our sense-per- ceptions can be recognized as sensation which can be overcome in the perceptual image and converted into its opposite by factors that are demonstrably due to experience. Whatever, therefore, can be overcome by fac- tors of experience, we must consider as being it- self the product of experience and training. By observing this rule, we shall find that it is merely the qualities of the sensation that are to be consid- ered as real; pure sensation; the great majority of space-apperceptions, however, being the product of experience and training. Concerning the Perceptions in General m 31 Still it does not follow that apperceptions, which persist in spite of our better conscious insight and continue as illusions, might not be due to experi- ence and training. Our knowledge of the changes of colour produced in distant objects by the hazi- ness of the atmosphere, of perspective distortions, and of shadow is undoubtedly a matter of experi- ence. And yet in a good landscape picture we shall get the perfect visual impression of the distance and the solid form of the buildings in it, in spite of knowing that it is all depicted on canvas. Similarly, our knowledge of the composite sound of the vowels is certainly obtained from experience; and yet we get the auditory-impres- sion of the vowel sound by combining the indi- vidual tones of tuning forks (as I have demon- strated) and grasp the sound in its entirety, although in this instance we know that it is really compound. Here we still have to explain how experience counteracts experience, and how illusion can be produced by factors derived from experience, when it might seem as if experience could not teach any- thing except what was true. In this matter we must remember, as was intimated above, that the sensa- tions are interpreted just as they arise when they are stimulated in the normal way, and when the organ of sense is used normally. We are not simply passive to the impressions that are urged on us, but we observe, that is, we adjust our organs in those conditions that enable them to distinguish the impressions most accu- rately. Thus, in considering an involved object, we accommodate both eyes as well as we can, and turn them so as to focus steadily the precise point on which our attention is fixed, that is, so as to get an image of it in the fovea of each eye; and then we let our eyes traverse all the noteworthy points of the object one after another. If we are interested in the general shape of the object and are trying to get as good an idea as we can of its relative di- mensions, we assume a position such that, with- out having to turn the head, we can survey the whole surface, enabling us at the same time to view as symmetrically as possible those dimensions we wish to compare. Thus, in looking at an object, as, for example, a building with prominent horizon- tal and vertical lines, we like to stand opposite to it with the centres of rotation of the two eyes ina horizontal line. This position of the eyes can be controlled at any moment by separating the double images; which in the case mentioned here are in the same horizontal plane.
32 ww Visual Perception Unquestionably, our reason for choosing this definite mode of seeing is because in this way we can observe and compare most accurately; and, consequently, in this so-called normal use of the eyes we learn best how to compare our sensations with the reality. And so we obtain also the most correct and most accurate perceptions by this method. But if, from necessity or on purpose, we em- ploy a different mode of looking at objects, that is, if we view them merely indirectly or without focusing both eyes on them, or without surveying them all over, or if we hold the head in some un- usual position, then we shall not be able to have as accurate apperceptions as when the eyes are used in the normal fashion. Nor are we so well trained in interpreting what we see under such circum- stances as in the other case. Hence there is more scope for interpretation, although, as a rule, we are not clearly aware of this uncertainty in the ex- planation of our sense-perceptions. When we see an object in front of us, we are obliged to assign it to some definite place in space. We cannot think of it as having some dubious intermediate posi- tion between two different places in space. With- out any recollections coming to our aid, we are wont to interpret the phenomenon as it would have to be interpreted if we had received the same im- pression in the normal and most accurate mode of observation. Thus certain illusions enter into the perception, unless we concentrate our eyes on the objects under observation, or when the objects are in the peripheral part of the visual field, or if the head is held to one side, or if we do not focus the object with both eyes at once. Moreover, the agree- ment between the images on the two retinas is most constant and regular in looking at distant objects. The fact that the horizontal floor usually happens to be in the lower part of the visual field, appar- ently influences the comparison of the fields of the two eyes in a peculiar manner. Thus, our judg- ment as to the position of near objects is not en- tirely correct when we observe them with the look tilted decidedly up or down. The retinal images presented in this way are interpreted just as if they had been obtained by looking straight ahead. We run across many illustrations of this sort. Our train- ing in interpreting immediate perceptions is not equally good in all directions of the eyes, but sim- ply for those directions which enable us to have the most accurate and most consistent perceptions. We transfer the latter to all cases, as in the instances just cited. Now.it is quite possible that the similarity be- tween a visual impression of this kind and one of the possible impressions obtained by normal ob- servation may not be so overwhelming and strik- ing as to preclude many other comparisons and corresponding interpretations of that impression. In such cases the explanation of the impression varies. Without any change of the retinal images, the same observer may see in front of him various perceptual images in succession, in which case the variation is easy to recognize. Or else one observer may incline more toward one comparison and in- terpretation, and another toward another. This has been a source of much controversy in physiologi- cal optics, because each observer has been disposed to consider the apperception which he obtained by the most careful observation he could make as being the only valid one. But supposing that we have such confidence in the observers as to as- sume that their observations were careful and un- prejudiced, and that they knew how to make them, it would not be proper in such cases to adopt one of the conflicting interpretations of the visual phe- nomenon as being the only correct one. And yet that is what they are disposed to do who try to derive the origin of perceptual images mainly from innate factors. The truth rather is, that in a case of this sort various perceptual images may be devel- oped; and we should seek rather to discover what circumstances are responsible for the decision one way or the other. It is true we meet with a difficulty here that does not exist in the other parts of the natural sciences. In many instances we have simply the assertions of individual observers, without being in the posi- tion to verify them by our own observation. Many idiosyncrasies are manifested in this region, some of which are doubtless due to the structure of the eyes, others to the habitual way of using the eyes, and others still perhaps to previous impressions and apperceptions. Of course, nobody save the person who has peculiarities of this nature can observe their effects, and nobody else can give an opinion about them. On the other hand, observa- tion in this region is by no means so easy as might be supposed at first. Steady fixation of a point for a long time while observations are being made in indirect vision; controlling the attention; taking the mind away from the ordinary objective interpre-
tation of sense-impression; estimation of differ- ence of colour and of difference of space in the visual field—all these things take much practice. And hence a number of facts in this region cannot be observed at all without having had previous long training in making observations in physiological optics. It cannot be done even by persons who are skilled in making other kinds of observations. Thus, with respect to many matters we have to depend on the observations of a very limited num- ber of individuals, and hence when the results found by somebody else are different, it is much harder in this subject than anywhere else to judge rightly whether secondary influences have not contributed in an observation of this sort. Accord- ingly, I must apprise the reader in advance that much of the material that is perhaps: new in the following chapters may possibly be due to indi- vidual peculiarities of my own eyes. Under such circumstances, there was no alternative for me except to observe as carefully as possible the facts as they appeared to my own eyes, and to try to ascertain their connection. Discrepancies that have been found by other observers have been noted. But how widespread this or the other mode of vi- sion may be, is something that has to be left to the future to determine. Incidentally, the more the visual impressions are unlike the normal ones, the greater will be the va- riety of interpretation as a rule. This is a natural consequence of the view which I hold, and is an essential characteristic of the activity of psychic influences. Heretofore practically nothing has been ascer- tained as to the nature of psychic processes. We have simply an array of facts. Therefore, it is not strange that no real explanation can be given of the origin of sense-perceptions. The empirical theory attempts to prove that at least no other forces are necessary for their origin beyond the known faculties of the mind, although these forces them- selves may remain entirely unexplained. Now gen- erally it is a useful rule in scientific investigation not to make any new hypothesis so long as known facts seem adequate for the explanation, and the necessity of new assumptions has not been dem- onstrated. That is why I have thought it incum- bent to prefer the empirical view essentially. Still less does the intuition theory attempt to give any explanation of the origin of our perceptual images; for it simply plunges right into the midst of the Concerning the Perceptions in General m 33 matter by assuming that certain perceptual images of space would be produced directly by an innate mechanism, provided certain nerve fibres were stimulated. The earlier forms of this theory im- plied some sort of self-observation of the retina; inasmuch as we were supposed to know by intu- ition about the form of this membrane and the positions of the separate nerve terminals in it. In its more recent development, especially as formu- lated by E. Hering, there is an hypothetical sub- jective visual space, wherein the sensations of the separate nerve fibres are supposed to be registered according to certain intuitive laws. Thus in this theory not only is Kant’s assertion adopted, that the general apperception of space is an original form of our imagination, but certain special ap- perceptions of space are assumed to be intuitive. The naturalistic view has been called also a spe- cial theory of identity, because in it the perfect fu- sion of the impressions on the corresponding places of the two retinas has to be postulated. On the other hand, the empirical theory is spoken of as a theory of projection,® because according to it the perceptual images of objects are projected in space by means of psychic processes. I should like to avoid this term, because both supporters and opponents of this view have often attached undue importance to the idea that this projection must take place parallel to the lines of direction; which was certainly not the correct description of the psychic process. And, even if this construction were admitted as being valid simply with respect to the physiological description of the process, the idea would be incorrect in very many instances. I am aware that in the present state of knowl- edge it is impossible to refute the intuition theory. The reasons why I prefer the opposite view are because in my opinion: 1. The intuition theory is an unnecessary hypoth- esis. 2. Its consequences thus far invariably apply to perceptual images of space which only in the fewest cases are in accordance with reality and with the correct visual images that are undoubt- edly present; as will be shown in detail later. The adherents of this theory are, therefore, obliged to make the very questionable assump- tion, that the space sensations, which accord- See remarks in Appendix I as to misunderstandings connected with the term “projection theory.’—K.
34 om Visual Perception ing to them are present originally, are continu- ally being improved and overruled by knowl- edge which we have accumulated by experi- ence. By analogy with all other experiences, however, we should have to expect that the sen- sations which have been overruled continued to be present in the apperception as a conscious illusion, if nothing else. But this is not the case. 3. It is not clear how the assumption of these origi- nal “space sensations” can help the explana- tion of our visual perceptions, when the adher- ents of this theory ultimately have to assume in by far the great majority of cases that these sen- sations must be overruled by the better under- standing which we get by experience. In that case it would seem to me much easier and sim- pler to grasp, that all apperceptions of space were obtained simply by experience, instead of supposing that the latter have to contend against intuitive perceptual images that are generally false. This is by way of justifying my point of view. A choice had to be made simply for the sake of get- ting at least some sort of superficial order amid the chaos of phenomena; and so I believed I had to adopt the view I have chosen. However, I trust it has not affected the correct observation and description of the facts. To prevent misunderstandings as to my mean- ing, and to make it clearer to the natural intelli- gence of those readers who have never thought much about their sense-perceptions, the follow- ing explanations will be added. Thus far the sensations have been described as being simply symbols for the relations in the ex- ternal world. They have been denied every kind of similarity or equivalence to the things they denote. Here we touch on the much disputed point as to how far our ideas agree in the main with their ob- jects; that is, whether they are true or false, as one might say. Some have asserted that there is such an agreement, and others have denied it. In favour of it, a pre-established harmony between nature and mind was assumed. Or it was maintained that there was an identity of nature and mind, by re- garding nature as the product of the activity of a general mind; the human mind being supposed to be an emanation from it. The intuition theory of Space-apperceptions is connected with these views to the fact that, by some innate mechanism and a certain pre-established harmony, it admits of the origin of perceptual images that are supposed to correspond with reality, although in a rather im- perfect.fashion. Or else the agreement between ideas and their objects was denied, the ideas being explained therefore as illusions. Consequently, it was neces- sary to deny also the possibility of all knowledge of any objects whatsoever. This was the attitude of certain so-called “sensational” philosophers in England in the eighteenth century. However, it is not my purpose here to undertake an analysis of the opinions of the various philosophical schools on this question. That would be much too exten- sive a task in this place. I shall confine myself therefore merely to inquiring what I think should be the attitude of an investigator toward these con- troversies. Our apperceptions and ideas are effects wrought on our nervous system and our consciousness by the objects that are thus apprehended and con- ceived. Each effect, as to its nature, quite neces- sarily depends both on the nature of what causes the effect and on that of the person on whom the effect is produced. To expect to obtain an idea which would reproduce the nature of the thing conceived, that is, which would be true in an ab- solute sense, would mean to expect an effect which would be perfectly independent of the nature of the thing on which the effect was produced; which would be an obvious absurdity. Our human ideas, therefore, and all ideas of any conceivable intelli- gent creature, must be images of objects whose mode is essentially codependent on the nature of the consciousness which has the idea, and is con- ditioned also by its idiosyncrasies. In my opinion, therefore, there can be no pos- sible sense in speaking of any other truth of our ideas except of a practical truth. Our ideas of things cannot be anything but symbols, natural signs for things which we learn how to use in order to regu- late our movements and actions. Having learned correctly how to read those symbols, we are en- abled by their help to adjust our actions so as to bring about the desired result; that is, so that the expected new sensations will arise. Not only is there in reality no other comparison at all between ideas and things—all the schools are agreed about this—but any other mode of comparison is entirely unthinkable and has no sense whatever. This latter consideration is the conclusive thing, and must be grasped in order to escape from the labyrinth of conflicting opinions. To ask whether the idea Ihave ~
of a table, its form, strength, colour, weight, etc., is true per se, apart from any practical use I can make of this idea, and whether it corresponds with the real thing, or is false and due to an illusion, has just as much sense as to ask whether a certain musical note is red, yellow, or blue. Idea and the thing conceived evidently belong to two entirely different worlds, which no more admit of being compared with each other than colours and musi- cal tones or than the letters of a book and the sound of the word they denote. Were there any sort of similarity of correspon- dence between the idea in the head of a person A and the thing to which the idea belongs, another intelligent person B, conceiving both the thing it- self and A’s idea of it, according to the same laws, might be able to find some similarity between them or at least to suppose so; because the same sort of thing represented (conceived) in the same way would have to give the same kinds of images (ideas). Now I ask, what similarity can be imag- ined between the process in the brain that is con- comitant with the idea of a table and the table it- self? Is the form of the table to be supposed to be outlined by electric currents? And when the per- son with the idea has the idea that he is walking around the table, must the person then be outlined by electric currents? Perspective projections of the external world in the hemispheres of the brain (as they are supposed to be) are evidently not suffi- cient for representing the idea of a bodily object. And granted that a keen imagination is not fright- ened away by these and similar hypotheses, such an electrical reproduction of the table in the brain would be simply another bodily object to be per- ceived, but no idea of the table. However, it is not simply persons with materialistic opinions who try to refute the proposed statement, but also persons with idealistic views. And for the latter I should think the argument would be still more forcible. What possible similarity can there be between the idea, some modification of the incorporeal mind that has no extension in space, and the body of the table that occupies space? As far as I am aware, the idealistic philosophers have never once inves- tigated even a single hypothesis or imagination in order to show this connection. And by the very nature of this view it is something that cannot be investigated at all. In the next place as to the properties of objects in the external world, alittle reflection reveals that all properties attributable to them may be said to Concerning the Perceptions in General m 35 be simply effects exerted by them either on our senses or on other natural objects. Colour, sound, taste, smell, temperature, smoothness, and firm- ness are properties of the first sort, and denote ef- fects on our organs of sense. Smoothness and firm- ness denote the degree of resistance either to the gliding contact or pressure of the hand. But other natural bodies may be employed instead of the hand. And the same thing is true in testing other mechanical properties such as elasticity and weight. Chemical properties are described by cer- tain reactions, that is, by effects exerted by one natural body on others. It is the same way with any other physical property of a body, optical, elec- trical, or magnetic. In every case we have to do with the mutual relations between various bodies and with the effects depending on the forces that different bodies exert on each other. For all natu- ral forces are such as are exerted by one body on others. When we try to think of mere matter with- out force, it is void of properties likewise, except as to its different distribution in space and as to its motion. All properties of bodies in nature are mani- fested therefore simply by being so situated as to interact with other bodies of nature or with our organs of sense. But as such interaction may oc- cur at any time, particularly too as it may be pro- duced by us voluntarily at any moment, and as then we see invariably the peculiar sort of interaction occurring, we attribute to the objects a permanent capacity for such effects which is always ready to become effective. This permanent capacity is a so- called characteristic property. The result is that in point of fact the character- istic properties of natural objects, in spite of this name, do not denote something that is peculiar to the individual object by itself, but invariably im- ply some relation to a second object (including our organs of sense). The kind of effect must, of course, depend always on the peculiarities both of the body producing it and of the body on which it is produced. As to this there is never any doubt even for an instant, provided we have in mind those properties of bodies that are manifested when two bodies belonging to the external world react on each other, as in the case of chemical reactions. But in the case of properties depending on the mutual relations between things and our organs of sense, people have always been disposed to forget that here too we are concerned with the reaction toward a special reagent, namely, our own nervous system; and that colour, smell, and taste, and feel-
Another Random Document on Scribd Without Any Related Topics
favourite tricks being that of making feint to assist a lady to descend and take her bag from her hand. “I don’t suppose,” she sighed, “that I shall ever see or hear of my ornaments again. Yet I think that if the thief but knew the truth concerning me he would regret. Perhaps he is without means, just as I am. Probably he became a thief of sheer necessity, as I have heard many men have become. Criminal instinct is not always responsible for an evil life. Many persons try to live honestly, but fate is ever contrary. Indeed, is it not so with my own self?” She turned, and her eyes fell upon the sleeping child. She was all she had now to care for in the whole wide world. Recollections of her last visit to Paris haunted her—that visit when Carl had so very indiscreetly followed her there, and taken her about incognito in open cabs to see the sights. There had been no harm in it whatsoever, no more harm than if he had been her equerry, yet her enemies had, alas! hurled against her their bitter denunciations, and whispered their lies so glibly that they were believed as truth. Major Scheel, the attaché at the Embassy, had recognised them, and being Leitolf’s enemy, had spread the report. It had been a foolish caprice of hers to take train from Aix-les- Bains to Paris to see her old French nurse Marie, who had been almost as a mother to her. The poor old woman, a pensioned servant of the Archducal family, had, unfortunately, died a month ago, otherwise she would have had a faithful, good friend in Paris. Marie, who knew Count Leitolf well, could have refuted their allegations had she lived; but an attack of pneumonia had proved fatal, and she had been buried with a beautiful wreath bearing the simple words “From Claire” upon her coffin.
As the sunset haze fell over Paris she still sat beside the sleeping child. If her enemies condemned her, then she would not defend herself. God, in whom she placed her fervent trust, should judge her. She had no fear of man’s prejudices or misjudgment. She placed her faith entirely in her Maker. To His will she bowed, for in His sight the pauper and the princess are equal. That evening she had a little soup sent to her room, and when Ignatia was again sleeping soundly she went forth upon the balcony leading from the corridor, and sitting there, amused herself by looking down upon the life and movement of the great salon below. To leave the hotel was impossible because of Ignatia, and she now began to regret that she had not brought the maid with her from Wartenstein. Time after time the misfortune of the loss of her jewels recurred to her. It had destroyed her independence, and it had negatived all her plans. Money was necessary, even though she were an Imperial Archduchess. She was incognito, and therefore had no credit. The gay, after-dinner scene of the hotel was presented below—the flirtations, the heated conversations, and the lazy, studied attitudes of the bloused English girl, who lolls about in cane lounge-chairs after dining, and discusses plays and literature. From her chair on the balcony above she looked down upon that strange, changeful world—the world of tourist Paris. Born and bred at Court as she had been, it was a new sensation to her to have her freedom. The life was entirely fresh to her, and would have been pleasant if there were not behind it all that tragedy of her marriage.
Several days went by, and in order to kill time she took little Ignatia daily in a cab and drove in the Bois and around the boulevards, revisiting all the “sights” which Leitolf had shown her. Each morning she went out driving till the luncheon-hour, and having once lunched with old Marie upstairs at the Brasserie Universelle in the Avenue de l’Opéra, she went there daily. You probably know the place. Downstairs it is an ordinary brasserie with a few chairs out upon the pavement, but above is a smart restaurant peculiarly Parisian, where the hors d’oeuvres are the finest in Eurorie and the vin gris a speciality. The windows whereat one sits overlook the Avenue, and from eleven o’clock till three it is crowded. She went there for two reasons—because it was small, and because the life amused her. Little Ignatia would sit at her side, and the pair generally attracted the admiration of every one on account of their remarkably good looks. The habitués began inquiring of the waiters as to who was the beautiful lady in black, but the men only elevated their shoulders and exhibited their palms. “A German,” was all they could answer. “A great lady evidently.” That she attracted attention everywhere she was quite well aware, yet she was not in the least annoyed. As a royalty she was used to being gazed upon. Only when men smiled at her, as they did sometimes, she met them with a haughty stare. The superiority of her Imperial blood would on such occasions assert itself, much to the confusion of would-be gallants. Thus passed those spring days with Paris at her gayest and best. The woman who had renounced a crown lived amid all that bright life, lonely, silent, and unrecognised, her one
anxiety being for the future of her little one, who was ever asking when Allen would return.
Chapter Fifteen. The Shy Englishman. One afternoon about four o’clock, as the Princess, leading little Ignatia, who was daintily dressed in white, was crossing the great hall of the Hotel Terminus on her way out to drive in the Bois, a rather slim, dark-haired man, a little under forty, well-dressed in a blue-serge suit, by which it required no second glance to tell that he was an Englishman, rose shyly from a chair and bowed deeply before her. At that hour there were only two or three elderly persons in the great hall, all absorbed in newspapers. She glanced at the stranger quickly and drew back. At first she did not recognise him, but an instant later his features became somehow familiar, although she was puzzled to know where she had met him before. Where he had bowed to her was at a safe distance from the few other people in the hall; therefore, noticing her hesitation, the man exclaimed in English with a smile,— “I fear that your Imperial Highness does not recollect me, and I trust that by paying my respects I am not intruding. May I be permitted to introduce myself? My name is Bourne. We met once in Treysa. Do you not recollect?” In an instant the truth recurred to her, and she stood before him open-mouthed. “Why, of course!” she exclaimed. “Am I ever likely to forget? And yet I saw so little of your face on that occasion that I
failed now to recognise you! I am most delighted to meet you again, Mr Bourne, and to thank you.” “Thanks are quite unnecessary, Princess,” he declared; whereupon in a low voice she explained that she was there incognito, under the title of the Baroness Deitel, and urged him not to refer to her true station lest some might overhear. “I know quite well that you are here incognito,” he said. “And this is little Ignatia, is it?” and he patted the child’s cheeks. Then he added, “Do you know I have had a very great difficulty in finding you. I have searched everywhere, and was only successful this morning, when I saw you driving in the Rue Rivoli and followed you here.” Was this man a secret agent from Treysa, she wondered. In any case, what did he want with her? She treated him with courtesy, but was at the same time suspicious of his motive. At heart she was annoyed that she had been recognised. And yet was she not very deeply indebted to him? “Well, Mr Bourne,” said the Princess, drawing herself up, and taking the child’s hand again to go out, “I am very pleased to embrace this opportunity of thanking you for the great service you rendered me. You must, however, pardon my failure to recognise you.” “It was only natural,” the man exclaimed quickly. “It is I who have to apologise, your Highness,” he whispered. “I have sought you because I have something of urgent importance to tell you. I beg of you to grant me an interview somewhere, where we are not seen and where we cannot be overheard.” She looked at him in surprise. The Englishman’s request was a strange one, yet from his manner she saw that he
was in earnest. Why, she wondered, did he fear being seen with her? “Cannot you speak here?” she inquired. “Not in this room, among these people. Are there not any smaller salons upstairs? they would be empty at this hour. If I recollect aright, there is a small writing-room at the top of the stairs yonder. I would beg of your Highness to allow me to speak to you there.” “But what is this secret you have to tell me?” she inquired curiously. “It surely cannot be of such a nature that you may not explain it in an undertone here?” “I must not be seen with you, Princess,” he exclaimed quickly. “I run great risk in speaking with you here in public. I will explain all if you will only allow me to accompany you to that room.” She hesitated. So ingenious had been the plots formed against her that she had now grown suspicious of every one. Yet this man was after all a mystery, and mystery always attracted her, as it always attracts both women and men equally. So with some reluctance she turned upon her heel and ascended the stairs, he following her at a respectful distance. Their previous meeting had indeed been a strange one. Fond of horses from her girlhood, she had in Treysa made a point of driving daily in her high English dogcart, sometimes a single cob, and sometimes tandem. She was an excellent whip, one of the best in all Germany, and had even driven her husband’s coach on many occasions. On the summer’s
afternoon in question, however, she was driving a cob in one of the main thoroughfares of Treysa, when of a sudden a motor car had darted past, and the animal, taking fright, had rushed away into the line of smart carriages approaching on the opposite side of the road. She saw her peril, but was helpless. The groom sprang out, but so hurriedly that he fell upon his head, severely injuring himself; while at that moment, when within an ace of disaster, a man in a grey flannel suit sprang out from nowhere and seized the bridle, without, however, at once stopping the horse, which reared, and turning, pinned the stranger against a tree with the end of one of the shafts. In an instant a dozen men, recognising who was driving, were upon the animal, and held it; but the next moment she saw that the man who had saved her had fallen terribly injured, the shaft having penetrated his chest, and he was lying unconscious. Descending, she gazed upon the white face, from the mouth of which blood was oozing; and having given directions for his immediate conveyance to the hospital and for report to be made to her as soon as possible, she returned to the palace in a cab, and telephoned herself to the Court surgeon, commanding him to do all in his power to aid the sufferer. Next day she asked permission of the surgeon that she might see the patient, to thank him and express her sympathy. But over the telephone came back the reply that the patient was not yet fit to see any one, and, moreover, had expressed a desire that nobody should come near him until he had quite recovered.
In the fortnight that went by she inquired after him time after time, but all that she was able to gather was that his name was Guy Bourne, and that he was an English banker’s clerk from London, spending his summer holiday in Treysa. She sent him beautiful flowers from the royal hot-houses, and in reply received his thanks for her anxious inquiries. He told the doctor that he hoped the Princess would not visit him until he had quite recovered. And this wish of his she had of course respected. His gallant action had, without a doubt, saved her from a very serious accident, or she might even have lost her life. Gradually he recovered from his injuries, which were so severe that for several days his life was despaired of, and then when convalescent a curious thing happened. He one day got up, and without a word of thanks or farewell to doctors, staff, or to the Crown Princess herself, he went out, and from that moment all trace had been lost of him. Her Highness, when she heard of this, was amazed. It seemed to her as though for some unexplained reason he had no wish to receive her thanks; or else he was intent on concealing his real identity with some mysterious motive or other. She had given orders for inquiry to be made as to who the gallant Englishman was; but although the secret agents of the Government had made inquiry in London, their efforts had been futile. It happened over two years ago. The accident had slipped from her memory, though more than once she had wondered who might be the man who had risked his life to save hers, and had then escaped from Treysa rather than be presented to her.
And now at the moment when she was in sore need of a friend he had suddenly recognised her, and come forward to reveal himself! Naturally she had not recognised in the dark, rather handsome face of the well-dressed Englishman the white, bloodless countenance of the insensible man with a brass- tipped cart-shaft through his chest. And he wanted to speak to her in secret? What had he, a perfect stranger, to tell her? The small writing-room at the top of the stairs was fortunately empty, and a moment later he followed her into it, and closed the door. Little Ignatia looked with big, wondering eyes at the stranger. The Princess seated herself in a chair, and invited the Englishman to take one. “Princess,” he said in a refined voice, “I desire most humbly to apologise for making myself known to you, but it is unfortunately necessary.” “Unfortunately?” she echoed. “Why unfortunately, Mr Bourne, when you risked your life for mine? At that moment you only saw a woman in grave peril; you were not aware of my station.” “That is perfectly true,” he said quietly. “When they told me at the hospital who you were, and when you sent me those lovely flowers and fruit, I was filled with—well, with shame.” “Why with shame?” she asked. “You surely had no need to be ashamed of your action? On the contrary, the King’s intention was to decorate you on account of your brave action, and had already given orders for a letter to be sent to your own King in London, asking his Majesty to allow you
as a British subject to receive and wear the insignia of the Order of the Crown and Sword.” “And I escaped from Treysa just in time,” he laughed. Then he added, “To tell you the truth, Princess, it is very fortunate that I left before—well, before you could see me, and before his Majesty could confer the decoration.” “But why?” she asked. “I must confess that your action in escaping as you did entirely mystified me.” “You were annoyed that I was ungentlemanly enough to run away without thanking your Highness for all your solicitude on my behalf, and for sending the surgeon of the royal household to attend to my injuries. But, believe me, I am most deeply and sincerely grateful. It was not ingratitude which caused me to leave Treysa in secret as I did, but my flight was necessary.” “Necessary? I don’t understand you.” “Well, I had a motive in leaving without telling any one.” “Ah, a private motive!” she said—“something concerning your own private affairs, I suppose?” He nodded in the affirmative. How could he tell her the truth? His disinclination to explain the reason puzzled her sorely. That he was a gallant man who had saved a woman without thought of praise or of reward was proved beyond doubt, yet there was something curiously mysterious about him which attracted her. Other men would have at least been proud to receive the thanks and decoration of a reigning sovereign, while he had utterly ignored them. Was he an anarchist?
“Princess,” he said at last, rising from his chair and flushing slightly, “the reason I have sought you to-day is not because of the past, but is on account of the present.” “The present! why?” “I—I hardly know what to say, Princess,” he said confusedly. “Two years ago I fled from you because you should not know the truth—because I was in fear. And now Fate brings me again in your path in a manner which condemns me.” “Mr Bourne, why don’t you speak more plainly? These enigmas I really cannot understand. You saved my life, or at least saved me from a very serious accident, and yet you escaped before I could thank you personally. To-day you have met me, and you tell me that you escaped because you feared to meet me.” “It is the truth, your Highness. I feared to meet you,” he said, “and, believe me, I should not have sought you to-day were it not of most urgent necessity.” “But why did you fear to meet me?” “I did not wish you to discover what I really am,” he said, his face flushing with shame. “Are you so very timid?” she asked with a light laugh. But in an instant she grew serious. She saw that she had approached some sore subject, and regretted. The Englishman was a strange person, to say the least, she thought. “I have nothing to say in self-defence, Princess,” he said very simply. “The trammels of our narrow world are so hypocritical, our laws so farcical and full of incongruities,
and our civilisation so fraught with the snortings of Mother Grundy, that I can only tell you the truth and offer no defence. I know from the newspapers of your present perilous position, and of what is said against you. If you will permit me to say so, you have all my sympathy.” And he paused and looked straight into her face, while little Ignatia gazed at him in wonder. “I wonder if your Highness will forgive me if I tell you the truth?” he went on, as though speaking to himself. “Forgive you? Why, of course,” she laughed. “What is there to forgive?” “Very much, Princess,” he said gravely. “I—I’m ashamed to stand here before you and confess; yet I beg of you to forgive me, and to accept my declaration that the fault is not entirely my own.” “The fault of what?” she inquired, not understanding him. “I will speak plainly, because I know that your good nature and your self-avowed indebtedness to me—little as that indebtedness is—will not allow you to betray me,” he said in a low, earnest tone. “You will recollect that on your Highness’s arrival at the Gare de l’Est your dressing-bag was stolen, and within it were your jewels—your most precious possession at this critical moment of your life?” “Yes,” she said in a hard voice of surprise, her brows contracting, for she was not yet satisfied as to the stranger’s bona fides. “My bag was stolen.” “Princess,” he continued, “let me, in all humility, speak the truth. The reason of my escape from Treysa was because your police held a photograph of me, and I feared that I might be identified. I am a thief—one of an international
gang. And—and I pray you to forgive me, and to preserve my secret,” he faltered, his cheeks again colouring. “Your jewels are intact, and in my possession. You can now realise quite plainly why—why I escaped from Treysa!” She held her breath, staring at him utterly stupefied. This man who had saved her, and so nearly lost his own life in the attempt, was a thief!
Chapter Sixteen. Light Fingers. Her Highness was face to face with one of those clever international criminals whose coups were so constantly being reported in the Continental press. She looked straight into his countenance, a long, intense look, half of reproach, half of surprise, and then, in a firm voice, said,— “Mr Bourne, I owe you a very great debt. To-day I will endeavour to repay it. Your secret, and the secret of the theft, shall remain mine.” “And you will give no information to the police?” he exclaimed quickly—“you promise that?” “I promise,” she said. “I admire you for your frankness. But, tell me—it was not you who took my bag at the station?” “No. But it was one of us,” he explained. “When the bag containing the jewels was opened I found, very fortunately, several letters addressed to you—letters which you evidently brought with you from Treysa. Then I knew that the jewels were yours, and determined, if I could find you, to restore them to you with our apologies.” “Why?” she asked. “You surely do not get possession of jewels of that value every day?” “No, Princess. But the reason is, that although my companions are thieves, they are not entirely devoid of the respect due to a woman. They have read in the newspapers
of your domestic unhappiness, and of your flight with the little Princess, and have decided that to rob a defenceless woman, as you are at this moment, is a cowardly act. Though we are thieves, we still have left some vestige of chivalry.” “And your intention is really to restore them to me?” she remarked, much puzzled at this unexpected turn of fortune. “Yes, had I not found those letters among them, I quite admit that, by this time, the stones would have been in Amsterdam and re-cut out of all recognition,” he said, rather shamefacedly. Then, taking from his pocket the three letters addressed to her—letters which she had carried away from Treysa with her as souvenirs—he handed them to her, saying,— “I beg of you to accept these back again. They are better in your Imperial Highness’s hands than my own.” Her countenance went a trifle pale as she took them, and a sudden serious thought flashed through her mind. “Your companions have, I presume, read what is contained in these?” “No, Princess; they have not. I read them, and seeing to whom they were addressed, at once took possession of them. I only showed my companions the addresses.” She breathed more freely. “Then, Mr Bourne, I am still more deeply in your debt,” she declared; “you realised that those letters contained a woman’s secret, and you withheld it from the others. How can I sufficiently thank you?”
“By forgiving me,” he said. “Remember, I am a thief, and if you wished you could call the hotel manager and have me arrested.” “I could hardly treat in that way a man who has acted so nobly and gallantly as you have,” she remarked, with perfect frankness. “If those letters had fallen into other hands they might, have found their way back to the Court, and to the King.” “I understand perfectly,” he said, in a low voice. “I saw by the dates, and gathered from the tenor in which they were written that they concealed some hidden romance. To expose what was written there would have surely been a most cowardly act—meaner even than stealing a helpless, ill-judged woman’s jewels. No, Princess,” he went on; “I beg that although I stand before you a thief, to whom the inside of a gaol is no new experience, a man who lives by his wits and his agility and ingenuity in committing theft, you will not entirely condemn me. I still, I hope, retain a sense of honour.” “You speak like a gentleman,” she said. “Who were your parents?” “My father, Princess, was a landed proprietor in Norfolk. After college I went to Sandhurst, and then entered the British Army; but gambling proved my ruin, and I was dismissed in disgrace for the forgery of a bill in the name of a brother officer. As a consequence, my father left me nothing, as I was a second son; and for years I drifted about England, an actor in a small travelling company; but gradually I fell lower and lower, until one day in London I met a well-known card-sharper, who took me as his partner, and together we lived well in the elegant rooms to which we inveigled men and there cheated them. The inevitable came
at last—arrest and imprisonment. I got three years, and after serving it, came abroad and joined Roddy Redmayne’s gang, with whom I am at present connected.” The career of the man before her was certainly a strangely adventurous one. He had not told her one tithe of the remarkable romance of his life. He had been a gentleman, and though now a jewel thief, he still adhered to the traditions of his family whenever a woman was concerned. He was acute, ingenious beyond degree, and a man of endless resource, yet he scorned to rob a woman who was poor. The Princess Claire, a quick reader of character, saw in him a man who was a criminal, not by choice, but by force of circumstance. He was now still suffering from that false step he had taken in imitating his brother officer’s signature and raising money upon the bill. However she might view his actions, the truth remained that he had saved her from a terrible accident. “Yours has been an unfortunate career, Mr Bourne,” she remarked. “Can you not abandon this very perilous profession of yours? Is there no way by which you can leave your companions and lead an honest life?” When she spoke she made others feel how completely the purely natural and the purely ideal can blend into each other, yet she was a woman breathing thoughtful breath, walking in all her natural loveliness with a heart as frail- strung, as passion-touched, as ever fluttered in a female bosom. “Ah, Princess!” he cried earnestly, “I beg of you not to reproach me; willingly I would leave it all. I would welcome work and an honest life; but, alas! nowadays it is too late.
Besides, who would take me in any position of trust, with my black record behind me? Nobody.” And he shook his head. “In books one reads of reformed thieves, but there are none in real life. A thief, when once a thief, must remain so till the end of his days—of liberty.” “But is it not a great sacrifice to your companions to give up my jewellery?” she asked in a soft, very kindly voice. “They, of course, recognise its great value?” “Yes,” he smiled. “Roddy, our chief, is a good judge of stones—as good, probably, as the experts at Spink’s or Streeter’s. One has to be able to tell good stuff from rubbish when one deals in diamonds, as we do. Such a quantity of fake is worn now, and, as you may imagine, we don’t care to risk stealing paste.” “But how cleverly my bag was taken!” she said. “Who took it? He was an elderly man.” “Roddy Redmayne,” was Bourne’s reply. “The man who, if your Highness will consent to meet him, will hand it back to you intact.” “You knew, I suppose, that it contained jewels?” “We knew that it contained something of value. Roddy was advised of it by telegraph from Lucerne.” “From Lucerne? Then one of your companions was there?” “Yes, at your hotel. An attempt was made to get it while you were on the platform awaiting the train for Paris, but you kept too close a watch. Therefore, Roddy received a telegram to meet you upon your arrival in Paris, and he met you.”
What he told her surprised her. She had been quite ignorant of any thief making an attempt to steal the bag at Lucerne, and she now saw how cleverly she had been watched and met. “And when am I to meet Mr Redmayne?” she asked. “At any place and hour your Imperial Highness will appoint,” was his reply. “But, of course, I need not add that you will first give your pledge of absolute secrecy—that you will say nothing to the police of the way your jewels have been returned to you.” “I have already given my promise. Mr Redmayne may rely upon my silence. Where shall we fix the meeting? Here?” “No, no,” he laughed—“not in the hotel. There is an agent of police always about the hall. Indeed, I run great risk of being recognised, for I fear that the fact of your having reported your loss to the police at the station has set Monsieur Hamard and his friends to watch for us. You see, they unfortunately possess our photographs. No. It must be outside—say at some small, quiet café at ten o’clock to- night, if it will not disturb your Highness too much.” “Disturb me?” she laughed. “I ought to be only too thankful to you both for restoring my jewels to me.” “And we, on our part, are heartily ashamed of having stolen them from you. Well, let us say at the Café Vachette, a little place on the left-hand side of the Rue de Seine. You cross the Pont des Arts, and find it immediately; or better, take a cab. Remember, the Vachette, in the Rue de Seine, at ten o’clock. You will find us both sitting at one of the little tables outside, and perhaps your Highness will wear a thick veil, for a pretty woman in that quarter is so quickly noticed.”
She smiled at his final words, but promised to carry out his directions. Surely it was a situation unheard-of—an escaped princess making a rendezvous with two expert thieves in order to receive back her own property. “Then we shall be there awaiting you,” he said. “And now I fear that I’ve kept you far too long, Princess. Allow me to take my leave.” She gave him her hand, and thanked him warmly, saying— “Though your profession is a dishonourable one, Mr Bourne, you have, nevertheless, proved to me that you are at heart still a gentleman.” “I am gratified that your Imperial Highness should think so,” he replied, and bowing, withdrew, and stepped out of the hotel by the restaurant entrance at the rear. He knew that the agent of police was idling in the hall that led out into the Rue St. Lazare, and he had no desire to run any further risk of detection, especially while that bag with its precious contents remained in the shabby upstairs room in the Rue Lafayette. Her Highness took little Ignatia and drove in a cab along the Avenue des Champs Elysées, almost unable to realise the amazing truth of what her mysterious rescuer of two years ago had revealed to her. She now saw plainly the reason he had left Treysa in secret. He was wanted by the police, and feared that they would recognise him by the photograph sent from the Prefecture in Paris. And now, on a second occasion, he was serving her against his own interests, and without any thought of reward! With little Ignatia prattling at her side, she drove along, her mind filled with that strange interview and the curious appointment that she had made for that evening.
Later that day, after dining in the restaurant, she put Ignatia to bed and sat with her till nine o’clock, when, leaving her asleep, she put on a jacket, hat, and thick veil— the one she had worn when she escaped from the palace— and locking the door, went out. In the Rue St. Lazare she entered a cab and drove across the Pont des Arts, alighting at the corner of the Rue de Seine, that long, straight thoroughfare that leads up to the Arcade of the Luxembourg, and walked along on the left- hand side in search of the Café Vachette. At that hour the street was almost deserted, for the night was chilly, with a boisterous wind, and the small tables outside the several uninviting cafés and brasseries were mostly deserted. Suddenly, however, as she approached a dingy little place where four tables stood out upon the pavement, two on either side of the doorway, a man’s figure rose, and with hat in hand, came forward to meet her. She saw that it was Bourne, and with scarcely a word, allowed herself to be conducted to the table where an elderly, grey-haired man had risen to meet her. “This is Mr Redmayne,” explained Bourne, “if I may be permitted to present him to you.” The Princess smiled behind her veil, and extended her hand. She recognised him in an instant as the gallant old gentleman in the bright red cravat, who, on pretence of assisting her to alight, had made off with her bag. She, an Imperial Archduchess, seated herself there between the pair of thieves.
Chapter Seventeen. In which “The Mute” is Revealed. When, in order to save appearances, Bourne had ordered her a bock, Roddy Redmayne bent to her, and in a low whisper said,— “I beg, Princess, that you will first accept my most humble apologies for what I did the other day. As to your Highness’s secrecy, I place myself entirely in your hands.” “I have already forgiven both Mr Bourne and yourself,” was her quiet answer, lifting her veil and sipping the bock, in order that her hidden face should not puzzle the waiter too much. “Your friend has told me that, finding certain letters in the bag, you discovered that it belonged to me.” “Exactly, and we were all filled with regret,” said the old thief. “We have heard from the newspapers of your flight from Treysa, owing to your domestic unhappiness, and we decided that it would be a coward’s action to take a woman’s jewels in such circumstances. Therefore we resolved to try and discover you and to hand them back intact.” “I am very grateful,” was her reply. “But is it not a considerable sacrifice on your part? Had you disposed of them you would surely have obtained a good round sum?” The man smiled. “We will not speak of sacrifice, your Highness,” the old fellow said. “If you forgive us and accept back your property, it is all that we ask. I am ashamed, and yet at the
same time gratified, that you, an Imperial Princess, should offer me your hand, knowing who and what I am.” “Whatever you may be, Mr Redmayne,” she said, “you have shown yourself my friend.” “And I am your friend; I’ll stand your friend, Princess, in whatever service you may command me,” declared the keen-eyed old man, who was acknowledged by the Continental police to be one of the cleverest criminals in the length and breadth of Europe. “We have discovered that you are alone here; but remember that you are not friendless. We are your friends, even though the world would call us by a very ugly name—a gang of thieves.” “I can only thank you,” she sighed. “You are extremely good to speak like this. It is true that misfortune has fallen upon me, and being friendless, it is reassuring to know that I have at least two persons in Paris ready to perform any service I require. Mr Bourne once rendered me a very great service, but refused to accept any reward.” And she added, laughing, “He has already explained the reason of his hurried departure from Treysa.” “Our departures are often hurried ones, your Highness,” he said. “Had we not discovered that the jewels were yours, we should in an hour have dispersed, one to England, one to Germany, and one to Amsterdam. But in order to discover you we remained here, and risked being recognised by the police, who know me, and are aware of my profession. To- morrow we leave Paris, for already Hamard’s agents, suspecting me of the theft, are searching everywhere to discover me.” “But you must not leave before I make you some reward,” she said. “Where are the jewels?”
“In that closed cab. Can you see it away yonder?” and he pointed to the lights of a vehicle standing some distance up the street. “Kinder, one of our friends, has it with him. Shall we get into the cab and drive away? Then I will restore the bag to you, and if I may advise your Highness, I would deposit it in the Crédit Lyonnais to-morrow. It is not safe for a woman alone to carry about such articles of great value. There are certain people in Paris who would not hesitate to take your life for half the sum they represent.” “Thank you for your advice, Mr Redmayne,” she said. “I will most certainly take it.” “Will your Highness walk to the cab with me?” Bourne asked, after he had paid the waiter. “You are not afraid to trust yourself with us?” he added. “Not at all,” she laughed. “Are you not my friends?” And she rose and walked along the street to where the cab was in waiting. Within the vehicle was a man whom he introduced to her as Mr Kinder, and when all four were seated within, Bourne beside her and Redmayne opposite her, the elder man took the precious bag from Kinder’s hand and gave it to her, saying,— “We beg of your Highness to accept this, with our most humble apologies. You may open it and look within. You will not, I think, find anything missing,” he added. She took the dressing-bag, and opening it, found within it the cheap leather bag she had brought from Treysa. A glance inside showed her that the jewels were still there, although there were so many that she, of course, did not count them. For a few moments she remained in silence; then thanking the two for their generosity, she said,—
“I cannot accept their return without giving you some reward, Mr Redmayne. I am, unfortunately, without very much money, but I desire you to accept these—if they are really worth your acceptance,” and taking from the bag a magnificent pair of diamond earrings she gave them into his hand. “You, no doubt, can turn them into money,” she added. The old fellow, usually so cool and imperturbable, became at once confused. “Really, Princess,” he declared, “we could not think of accepting these. You, perhaps, do not realise that they are worth at least seven hundred pounds.” “No; I have no idea of their value. I only command you to accept them as a slight acknowledgment of my heartfelt gratitude.” “But—” “There are no buts. Place them in your pocket, and say nothing further.” A silence again fell between them, while the cab rolled along the asphalte of the boulevard. Suddenly Bourne said,— “Princess, you cannot know what a weight of anxiety your generous gift has lifted off our minds. Roddy will not tell you, but it is right that you should know. The fact is that at this moment we are all three almost penniless—without the means of escape from Paris. The money we shall get for those diamonds will enable us to get away from here in safety.”
She turned and peered into his face, lit by the uncertain light of the street lamps. In his countenance she saw a deep, earnest look. “Then the truth is that without money to provide means of escape you have even sacrificed your chances of liberty, in order to return my jewels to me!” she exclaimed, for the first time realising the true position. He made no response; his silence was an affirmative. Kinder, who had spoken no word, sat looking at her, entirely absorbed by her grace and beauty. “Well,” she exclaimed at last, “I wonder if you would all three do me another small favour?” “We shall be only too delighted,” was Bourne’s quick reply. “Only please understand, your Highness, that we accept these earrings out of pure necessity. If we were not so sorely in need of money, we should most certainly refuse.” “Do not let us mention them again,” she said quickly. “Listen. The fact is this. I have very little ready money, and do not wish at this moment to reveal my whereabouts by applying to my lawyers in Vienna or in Treysa. Therefore it will be best to sell some of my jewellery—say one thousand pounds’ worth. Could you arrange this for me?” “Certainly,” Roddy replied, “with the greatest pleasure. For that single row diamond necklet we could get from a thousand to twelve hundred pounds—if that amount is sufficient.” She reopened the bag, and after searching in the fickle light shed by the street lamps she at length pulled out the necklet in question—one of the least valuable of the heap of
jewels that had been restored to her in so curious and romantic a manner. The old jewel thief took it, weighed it in his hand, and examined it critically under the feeble light. He had already valued it on the day when he had secured it. It was worth in the market about four thousand pounds, but in the secret channel where he would sell it he would not obtain more than twelve hundred for it, as, whatever he said, the purchaser would still believe it to be stolen property, and would therefore have the stones re-cut and reset. “You might try Père Perrin,” Guy remarked. “It would be quicker to take it to him than to send it to Amsterdam or Leyden.” “Or why not old Lestocard, in Brussels? He always gives decent prices, and is as safe as anybody,” suggested Kinder. “Is time of great importance to your Highness?” asked the head of the association, speaking with his decidedly Cockney twang. “A week or ten days—not longer,” she replied. “Then we will try Père Perrin to-morrow, and let you know the result. Of course, I shall not tell him whose property it is. He will believe that we have obtained it in the ordinary way of our profession. Perrin is an old Jew who lives over at Batignolles, and who asks no questions. The stuff he buys goes to Russia or to Italy.” “Very well. I leave it to you to do your best for me, Mr Redmayne,” was her reply. “I put my trust in you implicitly.” “Your Imperial Highness is one of the few persons—beyond our own friends here—who do. To most people Roddy
Redmayne is a man not to be trusted, even as far as you can see him!” and he grinned, adding, “But here we are at the Pont d’Austerlitz. Harry and I will descend, and you, Bourne, will accompany the Princess to her hotel.” Then he shouted an order to the man to stop, and after again receiving her Highness’s warmest thanks, the expert thief and his companion alighted, and, bowing to her, disappeared. When the cab moved on again towards the Place de la Bastille, she turned to the Englishman beside her, saying— “I owe all this to you, Mr Bourne, and I assure you I feel most deeply grateful. One day I hope I may be of some service to you, if,” and she paused and looked at him—“well, if only to secure your withdrawal from a criminal life.” “Ah, Princess,” he sighed wistfully, “if I only could see my way clear to live honestly! But to do so requires money, money—and I have none. The gentlemanly dress which you see me wearing is only an imposture and a fraud—like all my life, alas! nowadays.” She realised that this man, a gentleman by birth, was eager to extricate himself from the low position into which he had, by force of adverse circumstances, fallen. He was a cosmopolitan of cosmopolitans, a quiet, slow-speaking, slightly built, high-browed, genial-souled man, with his slight, dark moustache, shrewd dark eyes, and a mouth that had humour smiling at the corners; a man of middle height, his dark hair showing the first sign of changing early to grey, and a countenance bitten and scarred by all the winds and suns of the round globe; a wise and quiet man, able to keep his own counsel, able to get his own end with few words, and yet unable to shape his own destiny; a
Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com

More Related Content

PDF
The New Visual Neurosciences 1st Edition John S Werner Leo M Chalupa
ejowcjvyy1820
 
PDF
The Cognitive Neurosciences 3rd Edition Michael S. Gazzaniga (Editor)
aydashkrogh
 
PDF
Principles of Visual Attention Linking Mind and Brain 1st Edition Claus Bundesen
haenelpavi
 
PDF
The Asymmetrical Brain Hugdahl Kenneth Davidson Richard J
konkovvinckx
 
PDF
Theories Of Visual Perception 3rd Edition Ian E Gordon
soggiabinne
 
PDF
The Visual Brain In Action 2nd Edition A David Milner Melvyn A Goodale
miletoxainab98
 
PDF
Process Approaches to Consciousness in Psychology Neuroscience and Philosophy...
ovqkhmh6764
 
PDF
The Visual Neurosciences Volumes 1 2 1st Edition Leo M. Chalupa
guyetalioum
 
The New Visual Neurosciences 1st Edition John S Werner Leo M Chalupa
ejowcjvyy1820
 
The Cognitive Neurosciences 3rd Edition Michael S. Gazzaniga (Editor)
aydashkrogh
 
Principles of Visual Attention Linking Mind and Brain 1st Edition Claus Bundesen
haenelpavi
 
The Asymmetrical Brain Hugdahl Kenneth Davidson Richard J
konkovvinckx
 
Theories Of Visual Perception 3rd Edition Ian E Gordon
soggiabinne
 
The Visual Brain In Action 2nd Edition A David Milner Melvyn A Goodale
miletoxainab98
 
Process Approaches to Consciousness in Psychology Neuroscience and Philosophy...
ovqkhmh6764
 
The Visual Neurosciences Volumes 1 2 1st Edition Leo M. Chalupa
guyetalioum
 

Similar to Visual Perception Key Readings 1st Edition Steven Yantis (20)

PDF
Space Objects Minds And Brains Essays In Cognitive Psychology 1st Edition Lyn...
iifqyjha1504
 
PDF
The Visual Neurosciences Volumes 1 2 1st Edition Leo M. Chalupa
sikesstanci
 
PDF
Neural Basis Of Consciousness Naoyuki Osaka Eds
wffubpju100
 
PDF
Process Approaches to Consciousness in Psychology Neuroscience and Philosophy...
budhuradot
 
PDF
Brain Function Volume Iv Brain Function And Learning Reprint 2020 Donald B Li...
zamatanestoe
 
PDF
Cognitive Neuroscience Of Aging Linking Cognitive And Cerebral Aging 1st Edit...
embuskeesje
 
PDF
Cognitive Neuroscience The Biology Of The Mind 4th Edition Michael Gazzaniga
etalichile
 
PDF
The Visual Neurosciences Volumes 1 2 1st Edition Leo M. Chalupa
maheldailis
 
PDF
Computation Cognition And Pylyshyn Dedrick Don Trick Lana
gontholekena
 
PDF
Dementia Studies A Social Science Perspective Anthea Innes
snahavia
 
PDF
Mind Ascribed An Elaboration And Defence Of Interpretivism Bruno Mlder
anocheraible
 
PDF
Process Approaches to Consciousness in Psychology Neuroscience and Philosophy...
jayanejeuris
 
PDF
Working With Piaget Essays In Honour Of Brbel Inhelder Anastasia Tryphon
owadwya0505
 
PDF
The Thinking Mind A Festschrift For Ken Manktelow Niall Galbraith
gaetazunnoge
 
PDF
Under The Spell Of Landau When Theoretical Physics Was Shaping Destinies Mikh...
larkechenue
 
PDF
On Being Moved From Mirror Neurons To Empathy 1st Edition Stein Brten Editor
toxconpreet
 
PDF
The Handbook Of Industrial Work Organizational Psychology Volume 2 Organizati...
amnahmq1145
 
PDF
Explanation And Integration In Mind And Brain Science 1st Edition David M Kaplan
osjmxyy8041
 
PDF
The Visual Neurosciences Chalupa Lm Werner Js Eds
buallipuchol
 
PDF
Space And Sense Essays In Cognitive Psychology 1st Edition Susanna Millar
kiarkahullur
 
Space Objects Minds And Brains Essays In Cognitive Psychology 1st Edition Lyn...
iifqyjha1504
 
The Visual Neurosciences Volumes 1 2 1st Edition Leo M. Chalupa
sikesstanci
 
Neural Basis Of Consciousness Naoyuki Osaka Eds
wffubpju100
 
Process Approaches to Consciousness in Psychology Neuroscience and Philosophy...
budhuradot
 
Brain Function Volume Iv Brain Function And Learning Reprint 2020 Donald B Li...
zamatanestoe
 
Cognitive Neuroscience Of Aging Linking Cognitive And Cerebral Aging 1st Edit...
embuskeesje
 
Cognitive Neuroscience The Biology Of The Mind 4th Edition Michael Gazzaniga
etalichile
 
The Visual Neurosciences Volumes 1 2 1st Edition Leo M. Chalupa
maheldailis
 
Computation Cognition And Pylyshyn Dedrick Don Trick Lana
gontholekena
 
Dementia Studies A Social Science Perspective Anthea Innes
snahavia
 
Mind Ascribed An Elaboration And Defence Of Interpretivism Bruno Mlder
anocheraible
 
Process Approaches to Consciousness in Psychology Neuroscience and Philosophy...
jayanejeuris
 
Working With Piaget Essays In Honour Of Brbel Inhelder Anastasia Tryphon
owadwya0505
 
The Thinking Mind A Festschrift For Ken Manktelow Niall Galbraith
gaetazunnoge
 
Under The Spell Of Landau When Theoretical Physics Was Shaping Destinies Mikh...
larkechenue
 
On Being Moved From Mirror Neurons To Empathy 1st Edition Stein Brten Editor
toxconpreet
 
The Handbook Of Industrial Work Organizational Psychology Volume 2 Organizati...
amnahmq1145
 
Explanation And Integration In Mind And Brain Science 1st Edition David M Kaplan
osjmxyy8041
 
The Visual Neurosciences Chalupa Lm Werner Js Eds
buallipuchol
 
Space And Sense Essays In Cognitive Psychology 1st Edition Susanna Millar
kiarkahullur
 
Ad

Recently uploaded (20)

PDF
Supply Chain Operations Speaking Notes -ICLT Program
labyankof
 
PDF
GENETICS IN BIOLOGY IN SECONDARY LEVEL FORM 3
ElishamichaelSamwel
 
PDF
LDMMIA Reiki Yoga Finals Review Spring Summer
LDMMia Reiki Lectures
 
PPTX
History, Philosophy and sociology of education (1).pptx
tmdth1
 
PDF
Empowerment Technology for Senior High School Guide
solthereseamericandr
 
PDF
Indian roads congress 037 - 2012 Flexible pavement
offersjackpot
 
PDF
Complications of Minimal Access Surgery at WLH
mohitsuren827
 
PDF
Weekly quiz Compilation Jan -July 25.pdf
Nitin Suresh
 
PPTX
Orientation - ARALprogram of Deped to the Parents.pptx
JeromeTamayoSantiago1
 
PPTX
Chinmaya Tiranga Azadi Quiz (Class 7-8 )
Nitin Suresh
 
PPTX
CHAPTER IV. MAN AND BIOSPHERE AND ITS TOTALITY.pptx
greciamaycalambro
 
PPTX
Introduction to Building Materials
Mrunali Vasava
 
PDF
SOIL: Factor, Horizon, Process, Classification, Degradation, Conservation
Sipra Biswas
 
PPTX
Introduction-to-Literarature-and-Literary-Studies-week-Prelim-coverage.pptx
EricaRamos80
 
PDF
1_English_Language_Set_2.pdf probationary
emailjagmeetsingh9
 
PDF
Paper A Mock Exam 9_ Attempt review.pdf.
SameeraNaveed2
 
DOC
Soft-furnishing-By-Architect-A.F.M.Mohiuddin-Akhand.doc
ArchitectAFMMohiuddi
 
PPTX
Final Presentation General Medicine 03-08-2024.pptx
ZaheerAhmad228692
 
PDF
RTP_AR_KS1_Tutor's Guide_English [FOR REPRODUCTION].pdf
RobertMentino1
 
PDF
Chinmaya Tiranga quiz Grand Finale.pdf
Nitin Suresh
 
Supply Chain Operations Speaking Notes -ICLT Program
labyankof
 
GENETICS IN BIOLOGY IN SECONDARY LEVEL FORM 3
ElishamichaelSamwel
 
LDMMIA Reiki Yoga Finals Review Spring Summer
LDMMia Reiki Lectures
 
History, Philosophy and sociology of education (1).pptx
tmdth1
 
Empowerment Technology for Senior High School Guide
solthereseamericandr
 
Indian roads congress 037 - 2012 Flexible pavement
offersjackpot
 
Complications of Minimal Access Surgery at WLH
mohitsuren827
 
Weekly quiz Compilation Jan -July 25.pdf
Nitin Suresh
 
Orientation - ARALprogram of Deped to the Parents.pptx
JeromeTamayoSantiago1
 
Chinmaya Tiranga Azadi Quiz (Class 7-8 )
Nitin Suresh
 
CHAPTER IV. MAN AND BIOSPHERE AND ITS TOTALITY.pptx
greciamaycalambro
 
Introduction to Building Materials
Mrunali Vasava
 
SOIL: Factor, Horizon, Process, Classification, Degradation, Conservation
Sipra Biswas
 
Introduction-to-Literarature-and-Literary-Studies-week-Prelim-coverage.pptx
EricaRamos80
 
1_English_Language_Set_2.pdf probationary
emailjagmeetsingh9
 
Paper A Mock Exam 9_ Attempt review.pdf.
SameeraNaveed2
 
Soft-furnishing-By-Architect-A.F.M.Mohiuddin-Akhand.doc
ArchitectAFMMohiuddi
 
Final Presentation General Medicine 03-08-2024.pptx
ZaheerAhmad228692
 
RTP_AR_KS1_Tutor's Guide_English [FOR REPRODUCTION].pdf
RobertMentino1
 
Chinmaya Tiranga quiz Grand Finale.pdf
Nitin Suresh
 
Ad

Visual Perception Key Readings 1st Edition Steven Yantis

  • 1. Visual Perception Key Readings 1st Edition Steven Yantis download https://ebookbell.com/product/visual-perception-key-readings-1st- edition-steven-yantis-43686586 Explore and download more ebooks at ebookbell.com
  • 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Visual Perception Of Emotional And Conversational Facial Expressions 1st Edition Kathrin Kaulard https://ebookbell.com/product/visual-perception-of-emotional-and- conversational-facial-expressions-1st-edition-kathrin-kaulard-51627100 Visual Perception Part 1 Fundamentals Of Vision Low And Midlevel Processes In Perception 1st Edition Susana Martinezconde https://ebookbell.com/product/visual-perception-part-1-fundamentals- of-vision-low-and-midlevel-processes-in-perception-1st-edition-susana- martinezconde-2192184 Visual Perception An Introduction 2nd Ed Michael T Swanston https://ebookbell.com/product/visual-perception-an-introduction-2nd- ed-michael-t-swanston-37316980 Visual Perception And Control Of Underwater Robots 1st Edition Yu https://ebookbell.com/product/visual-perception-and-control-of- underwater-robots-1st-edition-yu-38373170
  • 3. Visual Perception Through Video Imagery Michel Dhome https://ebookbell.com/product/visual-perception-through-video-imagery- michel-dhome-4313878 Visual Perception Fundamentals Of Awareness Multisensory Integration And Highorder Perception S Martinezconde https://ebookbell.com/product/visual-perception-fundamentals-of- awareness-multisensory-integration-and-highorder-perception-s- martinezconde-4423928 Visual Perception From A Computer Graphics Perspective William Thompson https://ebookbell.com/product/visual-perception-from-a-computer- graphics-perspective-william-thompson-4748526 Visual Perception Physiology Psychology And Ecology 4th Vicki Bruce https://ebookbell.com/product/visual-perception-physiology-psychology- and-ecology-4th-vicki-bruce-5151800 Visual Perception For Humanoid Robots Environmental Recognition And Localization From Sensor Signals To Reliable 6d Poses 1st Ed David Israel Gonzlez Aguirre https://ebookbell.com/product/visual-perception-for-humanoid-robots- environmental-recognition-and-localization-from-sensor-signals-to- reliable-6d-poses-1st-ed-david-israel-gonzlez-aguirre-7324776
  • 5. KEY READIN GS sNGst GG Nt kOe ‘Visual Perception Steven Yantis —
  • 6. LANCHESTER LIBRARY IANIOMIIAMUM 3 8001 00299 0111 lob PERCEPTION
  • 7. Key Readings in Cognition The aim of this series is to make available to senior undergraduate and graduate students key articles in each area of cognition in an attractive, user-friendly format. Many professors want to encourage their students to engage directly with research in their fields, yet this can often be daunting for students coming to detailed study of a topic for the first time. Moreover, declining library budgets mean that articles are not always readily available, and course packs can be expensive and time-consuming to produce. Key Readings in Cognition aims to address this need by providing compre- hensive volumes, each one of which will be edited by a senior and active researcher in the field. Articles will be carefully chosen to illustrate the way the field has developed historically as well as current issues and research directions. Each volume will have a similar structure to include: ¢ An overview chapter, as well as introductions to sections and articles ¢ Questions for class discussion e Annotated bibliographies e Full author and subject indexes Titles in Preparation: The Psychology of Human Memory Edited by Henry L. Roédiger, Ill and Kathleen McDermott Cognitive Neuroscience Edited by Marie Banich and Neal Cohen The Psychology of Language Edited by Michael Tanenhaus
  • 8. VloUHL PERCEPTION Essential Headings Edited by Steven Yantis Johns Hopkins University Baltimore, MD LOG oe & »y Gs ~n Nn Ay ~” Mw
  • 9. USA Publishing Office: » PSYCHOLOGY PRESS A member of the Taylor & Francis Group 325 Chestnut Street Philadelphia, PA 19106 Tel: (215) 625-8900 Fax: (215) 625-2940 Distribution Center: PSYCHOLOGY PRESS A member of the Taylor & Francis Group 7625 Empire Drive Florence, KY 41042 Tel: 1-800-634-7064 Fax: 1-800-248-4724 PSYCHOLOGY PRESS A member of the Taylor & Francis Group 27 Church Road Hove E. Sussex, BN3 2FA Tel: +44 (0) 1273 207411 Fax: +44 (0) 1273 205612 VISUAL PERCEPTION: Essential Readings Copyright © 2001 Taylor & Francis. All rights reserved. Printed in the United States of America. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written permission of the publisher. 12534. 567-89. 0 Printed by Sheridan Books - Braun-Brumfield, Ann Arbor, MI, 2001. A CIP catalog record for this book is available from the British Library. € 6 The paper in this publication meets the requirements of the ANSI Standard Z39.48-1984 (Permanence of Paper) Vasily Kandinsky, Composition 8, July 1923, Oil on canvas, Solomon R. Guggenheim Museum, New York. Photograph by David Heald © The Solomon R. Guggenheim Foundation, New York, FN 37.262 Library of Congress Cataloging-in-Publication Data Visual perception : essential readings / edited by Steven Yantis. p. cm. -- (Key readings in cognition) Includes bibliographical references and index. ISBN 0-86377-597-7 (case : alk. paper) -- ISBN 0-86377-598-5 (paper : alk. paper) 1. Visual perception. I. Yantis, Steven. II. Series. QP491 . V577 2000 152.14--dce21 CIP 00-042311 ISBN: 0-86377-597-7 (case) ISBN: 0-86377-598-5 (paper) overtly University
  • 10. Contents About the Editor xi Acknowledgements _ xiii Visual Perception: An Overview 1 PART 1 Theoretical Perspectives 19 READING1 Introduction to Reading 1 21 Concerning the Perceptions in General 24 H, von Helmholtz READING 2 Introduction to Reading 2. 45 A Decision-Making Theory of Visual Detection 48 W. P. Tanner, Jr. and J. A. Swets READING 3 Introduction to Reading 3. 57 Single Units and Sensation: A Neuron Doctrine for Perceptual Psychology? 60 H. B, Barlow READING 4 Introduction to Reading 4 81 The Theory of Information Pickup and its Consequences 84 J, J. Gibson
  • 11. vi m_ Visual Perception READING 5 Introduction to Reading 5 101. The Philosophy and the Approach 104 D. Marr i Part 1: Discussion Questions and Suggested Readings 124 PART 2 Early Vision 125 READING 6 Introduction to Reading 6 127 An Opponent-Process Theory of Color Vision 129 L. M. Hurvich and D, Jameson READING 7 Introduction to Reading 7 145 Receptive Fields and Functional Architecture of Monkey Striate Cortex 147 D. H. Hubel and T. N. Wiesel READING 8 Introduction to Reading 8 169 On the Existence of Neurones in the Human Visual System Selectively Sensitive to the Orientation and Size of Retinal Images 172 C, Blakemore and F. W. Campbell READING 9 Introduction to Reading 9 191 A Direct Demonstration of Functional Specialization in Human Visual Cortex 193 S, Zeki, J. D. G. Watson, C. J. Lueck, K. J, Friston, C. Kennard, and R, S. J. Frackowiak READING 10 Introduction to Reading 10 203 Neuronal Correlates of a Perceptual Decision 205 W. T. Newsome, K. H. Britten, and J. A. Movshon Part 2: Discussion Questions and Suggested Readings 209
  • 12. Contents m vii PART 3 Perceptual Organization and Constancy 211 READING 11 Introduction to Readings 11 and 12 213 Laws of Organization in Perceptual Forms 216 M. Wertheimer READING 12 Figure and Ground 9225 E, Rubin READING 13 Introduction to Reading 13 231 The Moon Illusion 233 L. Kaufman and |, Rock READING 14 Introduction to Readings 14 and 15 243 Brightness Constancy and the Nature of Achromatic Colors 246 H. Wallach READING 15 Grouping Based on Phenomenal Similarity of Achromatic Color 256 |,Rock, R. Nijhawan, S. Palmer, and L. Tudor ° Part 3: Discussion Questions and Suggested Readings 266 PART 4 Object and Spatial Vision 269 READING 16 Introduction to Reading 16 271 A Case of Visual Agnosia with a Contribution to Theory 274 H, Lissauer READING 17 Introduction to Reading 17 293 Object Vision and Spatial Vision: Two Cortical Pathways 296 M. Mishkin, L. G. Ungerleider, and K. A. Macko
  • 13. viii m Visual Perception READING 18 Introduction to Reading 18 303 Phenomenal Coherence of Moving Visual Patterns 306 E, H. Adelson and J. A. Movshon READING 19 Introduction to Reading 19 311 Mental Rotation of Three-Dimensional Objects 313 R, N. Shepard and J. Metzler READING 20 Introduction to Reading 20 317 Recognition-by-Components: A Theory of Human Image Understanding 320 |, Biederman Part 4: Discussion Questions and Suggested Readings 341 PART 5 Visual Attention and Awareness 343 READING 21 Introduction to Reading 21 345 A Feature-Integration Theory of Attention 347 A.M. Treisman and G. Gelade READING 22 Introduction to Readings 22 and 23 359 Selective Attention Gates Visual Processing in the Extrastriate Cortex 361 J. Moran and R. Desimone READING 23 Voluntary Attention Modulates fMRI Activity in Human MT-MST 365 K, M. O'Craven, B. R. Rosen, K. K. Kwong, A. Treisman, and R, L. Savoy READING 24 Introduction to Readings 24 and 25 375 Visual Capacity in the Hemianopic Field Following a Restricted Occipital Ablation 378 L. Weiskrantz, E. K. Warrington, M. D, Sanders, and J. Marshall
  • 14. Contents m ix READING 25 The Role of Temporal Cortical Areas in Perceptual Organization 393 D, L. Scheinberg and N. K. Logothetis Part 5: Discussion Questions and Suggested Readings 402 Appendix: Reading Journal Articles in Cognitive Psychology 405 H. L. Roediger, Ill and D. A. Gallo Author Index 417 Subject Index 425 The following color plates follow page 210 Color Plate 9.1 for Reading 9 Color Plate 9.2 for Reading 9 Color Plate 9.3 for Reading 9 Color Plate 9.4 for Reading 9 Color Plate 23.5 for Reading 23 Color Plate 23.6 for Reading 23
  • 15. — “=e, a hive me ia nal cpa umes es nsilesins gOlandQa20 4 AD te fctiesadti >: egebaah bateaiglieBinsano a Re if: 20k ancnn aapga — s ~- *5 Secu awl > Aas a) on SF xa onto, eases 38 eSh >xebat Wwirbtties, on to Ailing £0 Lally Ranggnits aeoagetinge . “igen Sear pie = Pal = weary Lake we oa jones poe ional eae ae r £5nities 101 APS.stat yu), 1 etd Mection ane Awlwraness 7. ~— i" wire 1 ee . ; — a <a ; Poa pecoinoereat vache k 93 7 —— = : ators inteciation Maury of Agron. ja , -_ ; oO ee ent , Gitex ~~. 7 py divvk gt : ‘ io "WEES Netliny: oa dag BD] ge ae er ptetFakes.ina Pend «ete S a ; ae tay et”. eee - i; “on 9 a cin it ae ao as | - hares eae ll ica ~ 2° oe vere 2 <a. 7a si i ye teal a artes Wi Prides eh Orcas, Wii Sens Ration mY.
  • 16. Hhout the Editor eo Steven Yantis is a Professor of Psychology and Cognitive Sciences at The Johns Hopkins University. He received a B.S. in Psychology from the University of Washington in 1978 and a PhD in Experimental Psychology from the University of Michigan in 1985. He is the recipient of the Troland Research Award from the National Academy of Sciences and the Early Career Award from the American Psychological Association. Professor Yantis’s research is concerned with the psychological and neural mechanisms of visual attention and perceptual organization. x1
  • 17. o ~ i — ns A sat sicamcmmacmmmcatnt ee A A ALLO D is : ’ _ 7 4 Ke ‘ s 7 a ogee «gs be Yooh so wo 8 NY ge ah inertygolodnae thaa 5 (Sven sti gent acto a a _ rot mi vig hein mtSA truOTBACORA a, 1 en a renee brat ooh YoBrigg, orf ofoH .eBer rn é : cent awh ees aidott tafe naan Ws ytd whan ’ : : ihe reovuen) riaang vaneten conc Mieka 7% . a e an eae Mile < ar reali tenraty ioervaings cea ipo : ‘| ae rae ; y "a * = ‘ a e A yn | see = » ; re e > : ; | ‘ ¥ ; ’ » 3 ' 7
  • 18. Acknowledgements The Authors and Publishers are grateful to the fol- lowing for permission to reproduce the articles in this book: Reading 1: H. von Helmholtz (1925). Physiologi- cal Optics. (Vol III, $26. Concerning the percep- tions in general. pp. 1-36). Translated from the Third German Edition and Edited by J.P.C. Southall. New York: Optical Society of America. (Original work published 1896). Reading 2: H. B. Barlow, Single units and sensa- tion: A neuron doctrine for perceptual psychology? Perception, 1, 371-394. Copyright © 1972 by Pion Limited, London. Reprinted with permission. Reading 3: W. P. Tanner, Jr. and J. A. Swets, A decision-making theory of visual detection. Psy- chological Review, 61, 401-409. Copyright ©1954 by the American Psychological Association. Re- printed with permission. Reading 4: J. J. Gibson, The ecological approach to visual perception. Chapter 14: The theory of information pickup and its consequences (pp. 238- 263). Boston: Houghton Miflin Co., 1979. Re- printed with permission. Reading 5: D. Marr, Vision. Chapter 1: The phi- losophy and the approach (pp. 8-38). San Fran- cisco: W. H. Freeman and Co., 1982. Reprinted with permission. Reading 6: L. M. Hurvich and D. Jameson, An opponent-process theory of color vision. Psycho- logical Review, 64, 384-404. Copyright © 1957 by the American Psychological Association. Re- printed with permission. Reading 7: D. H. Hubel and T. N. Wiesel, Recep- tive fields and functional architecture of monkey striate cortex. Journal of Physiology, 195, 215-243. Copyright © 1968 by Cambridge University Press. Reprinted with permission. Reading 8: C. Blakemore and F. W. Campbell, On the existence of neurones in the human visual sys- tem selectively sensitive to the orientation and size of retinal images. Journal of Physiology, 203, 237- 260. Copyright © 1969 by Cambridge University Press. Reprinted with permission. Reading 9: S. Zeki, J. D. G. Watson, C. J. Lueck, K. J. Friston, C. Kennard, & R. S. J. Frackowiak, A direct demonstration of functional specialization in human visual cortex. Journal of Neuroscience, 11, 641-649. Copyright © 1991 by the Society for Neuroscience. Reprinted with permission. Reading 10: W. T. Newsome, K. H. Britten, & J. A. Movshon, Neuronal correlates of a perceptual decision. Nature, 341, 52-54. Copyright © 1989 by Macmillan Magazines Ltd. Reprinted with per- mission. Reading 11: M. Wertheimer, Untersuchungen zur Lehre von der Gestalt, II [Laws of organization in perceptual forms]. Psycholoche Forschung, 4, 301- 350. Excerpts translated and reprinted in W. D. Ellis (Ed.), A source book of Gestalt psychology (pp. 71-88). New York: Routledge, 1999. Reprinted with permission. Xili
  • 19. xiv m_ Visual Perception Reading 12: E. Rubin, Visuaell wahrgenommene Figuren [Figure and Ground]. Copenhagen: Gyldendalske, 1921. [excerpts translated and re- printed in D. C. Beardslee & M. Wertheimer (Eds.), Readings in Perception (pp. 194-203). Princeton, NJ: D. Van Norstrand Co., Inc.]. Reading 13: L. Kaufman and I. Rock, The Moon Illusion. Scientific American, 207, 120-132. Copy- right © 1962 by Scientific American, Inc. Re- printed with permission. All rights reserved. Reading 14: H. Wallach, Brightness constancy and the nature of achromatic colors. Journal of Experi- mental Psychology, 38, 310-324. Copyright © 1948 by the American Psychological Association. Reprinted with permission. Reading 15: I. Rock, R. Nijhawan, S. Palmer, & L. Tudor, Grouping based on phenomenal simi- larity of achromatic color. Perception, 21, 779-789. Copyright © 1992 by Pion Limited, London. Re- printed with permission. Reading 16: H. Lissauer (1890), Ein Fall von Seelenblindheit nebst einem Beitrag zur Theorie derselben. Archiv fiir Psychiatrie, 21, 222-270. [translated and reprinted in 1988 as “A case of vi- sual agnosia with a contribution to theory.” Cog- nitive Neuropsychology, 5, 157-192; commentary by T. Shallice & M. Jackson. Lissauer on agnosia. Cognitive Neuropsychology, 5, 153-156. Copy- right © 1988 by Taylor and Francis. Reprinted with permission. Reading 17: M. Mishkin, L. G. Ungerleider, & K. A. Macko, Object vision and spatial vision: Two cortical pathways. Trends in Neurosciences, 6, 414-417. Copyright © 1982 by Elsevier Science. Reprinted with permission. Reading 18: E. H. Adelson and J. A. Movshon, Phenomenal coherence of moving visual patterns. Nature, 300, 523-525. Copyright © 1982 by Macmillan Magazines, Ltd. Reprinted with per- mission. Reading 19: R. N. Shepard and J. Metzler, Men- tal rotation of three-dimensional objects. Science, 171, 701-703. Copyright © 1971 by the American Association for the Advancement of Science: Re- printed with permission. Reading 20: I. Biederman, Recognition-by-com- ponents: A theory of human image understanding. Computer Vision, Graphics, and Image Process- ing, 32, 29-73. Copyright © 1985 by Academic Press. Reprinted with permission. Reading 21: A. M. Treisman and G. Gelade, A feature-integration theory of attention. Cognitive Psychology, 12, 97-136. Copyright ©1980 by Academic Press. Reprinted with permission. Reading 22: J. Moran and R. Desimone, Selec- tive attention gates visual processing in the extrastriate cortex. Science, 229, 782-784. Copyright © 1985 by American Association for the Advancement of Science. Reprinted with permission. Reading 23: K. M. O’Craven, B. R. Rosen, K. K. Kwong, A. Treisman, & R. L. Savoy, Voluntary attention modulates fMRI activity in human MT- MST. Neuron, 18, 591-598. Copyright © 1997 by Cell Press. Reprinted with permission. Reading 24: L. Weiskrantz, E. K. Warrington, M. D. Sanders, & J. Marshall, Visual capacity in the hemianopic field following a restricted occipital ablation. Brain, 97, 709-728. Copyright © 1974 by Oxford University Press. Reprinted with permission. Reading 25: D. L. Sheinberg and N. K. Logothetis, The role of temporal cortical areas in perceptual organization. Proc. National Academy of Sciences USA, 94, 3408-3413. Copyright © 1997 by The National Academy of Sciences of the USA. Re- printed with permission. Appendix: From Research Methods in Psychol- ogy (with InfoTrac), 6" edition, by D.G. Elmes, B.H. Kantowitz, and H.L. Roediger, III © 1999. Reprinted with permission of Wadsworth, a divi- sion of Thomson Learning. Fax 800-730-2215.
  • 20. V Visual Perception: finOverview Steven Yantis isual perception allows organisms to obtain knowledge of their surroundings by sensing light reflected from surfaces. Its ultimate purpose is to allow one to know what objects are present so as to behave appropriately and in accordance with one’s current behavioral goals. Although recognizing objects is subjectively effortless and nearly instantaneous, the steps that it requires are amazingly complex. Understanding visual perception requires an understanding of a diverse and multifaceted body of knowledge, including physics and optics, neuroanatomy and neurophysiology, psychology, and computation. This volume contains 25 influential chapters and articles from the vision science litera- ture that reflect the theoretical and empirical themes that have dominated research in visual perception over the last century. They are representative of a huge body of work that is truly interdisciplinary. The fertile crosstalk between psychology and neuroscience in par- ticular is made possible by the close correspondence between neural activity on the one hand and certain aspects of perceptual behavior on the other. This means that we can begin to grasp the neural implementation of vision more readily than that of other cognitive functions like memory, language, or planning and decision making where the link between neurophysiology.and behavioris-less.well.understood, What makes this interplay particu- larly exciting right now are technological advances that have brought new scientific tools to bear in the cognitive neuroscience of vision (e.g., technologies that permit us to collect images of brain activity in behaving human observers), so that new discoveries about how the human brain implements vision are now reported routinely in the scientific literature and the press. In this introductory chapter, I will outline of some of these theoretical and empirical themes and how they have been influenced by the tools that are available to practicing scientists. Of course, in this brief space it is not possible to adequately cover this topic in all its complexity and richness. At the end of this chapter, a list of several other sources that provide deeper treatments is provided. I started with the assertion that the purpose of vision is object recognition. Of course, one could quibble with this assertion, but it is a good starting point nevertheless. If we wish to understand how the visual system can take as its input a dynamic pattern of light and color at the retina and produce as its output a label that identifies or categorizes an object, we must first break the problem down into a smaller number of slightly more manageable parts. The selected readings have been grouped according to the following questions:
  • 21. 2 Visual Perception * The problem of early vision: How are elementary sensory attributes (e.g. color, motion) encoded? ¢ The problem of perceptual constancy: How are elementary attributes “corrected” for irrelevant viewing conditions? ¢ The problem of perceptual organization: How are sensory attributes conjoined into rep- resentations of surfaces and objects? * The problem of object recognition: How are surface representations used to categorize or identify objects? * The problem of attention: How does the observer select relevant objects and ignore irrel- evant objects? ° The problem of visual awareness: How is consciousness achieved? All of these problems are under active investigation; none has been solved. Of course, more progress has been made in some areas than others: we know a very great deal (but not everything) about how elementary sensory properties are represented in the visual system; we know very little about awareness. Although object recognition is arguably the most significant of these problems, there is little consensus about how this is achieved. This is not surprising, of course, given its complexity and that it relies to some degree on success in all of the other areas. The first section in this book is titled Theoretical Approaches. The choice of a theoreti- cal orientation is in some sense orthogonal to the substantive issues sketched above. Of course, one’s choice of a theoretical approach is utterly crucial in determining research strategies and identifying relevant questions. Indeed, according to certain theoretical tradi- tions, the questions outlined are simply not the important ones. Yet most scientists do not spend much time mulling possible theoretical approaches. In most sciences one main ap- proach tends to be dominant at any given time, and most scientists implicitly accept that approach. In contemporary vision research, the dominant theoretical approach can be traced to Helmholtz’s constructivism, which is the view that we construct internal representations of the objects in the visible scene that are most likely to have given rise to the pattern of light currently impinging on our retinas. This requires us to take into account the early sensory data, of course, but it also allows for the influence of past experiences (i.e., memory), expectation about what is likely in the current context, and what aspects of the scene are important given current behavioral goals (i.e., attention). In other words, the representation of objects in a scene is a joint result of both bottom-up information from the eyes, and top- down information from memory and attentional mechanisms. Almost all of the readings in this volume start from this common theoretical platform. (Gibson’s idea that perception is in an important sense direct constitutes the most notable alternative view, but his ideas have been extremely influential even among adherents of the constructivist perspective.) Two Principles of Visual Perception Before turning to the visual themes themselves, there are two central principles that char- acterize vision and that will help to orient the discussion. The first principle is that of functional specialization. In the section on methodology that concludes this chapter, the evidence for functional specialization emerged in the late 19" century. There are many ways in which functional specialization can be illustrated. Perhaps the first point at which specialization occurs in vision is at the very earliest point it could occur: in the photoreceptors themselves. There are four different types of photoreceptors grouped into two classes: the rods, and three types of cones. In addition to their shapes, there are at least three ways in which the rods and cones differ, and each of these has implications for their functions. First, the rods are poorly suited to represent
  • 22. An Overview m 3 color information, while the three types of cones are ideally suited for this task (more on this later). Second, the rods are much more sensitive to low levels of illumination, and they are what. we must rely on for night vision; the cones are used under ordinary daylight viewing conditions. Third, the rods and cones are distributed throughout the retinal mosaic differently: the cones are packed densely in the center of gaze—the fovea—with no rods; both rods and cones are interspersed outside the fovea with density gradually decreasing into the periphery. The high density of the cones in central vision allows a very fine spatial sampling of the pattern of light in central vision, which in turn allows us to see fine spatial detail there. Functional specialization does not end with the rods and cones. The optic nerve, a bundle of fibers that carries visual information from the eye to the brain, consists of at least two distinct subpopulations of fibers that project to distinct regions in the brain. M fibers tend to favor information that varies temporally, such as motion or flicker, while P fibers tend to carry information about static properties such as color, orientation, or depth. This functional distinction continues as the information from the eyes makes its way to higher centers in the visual cortex, and it is intimately related to the distinction between the “what” and “where” pathways suggested by Mishkin et al. (Reading 17). As suggested by Zeki et al. (Reading 9), certain areas of cortex contain a predominance of cells that appear to be function- ally specialized for a single visual attribute (e.g., area V4 for color, area V5 for motion). A second principle of visual perception is that of distributed coding of perceptual at- tributes. There are (roughly speaking) two ways that any given attribute (e.g., color, edge orientation) can be represented in cortical tissue. According to a local coding approach, any given attribute or combination of attributes is represented by a single specialized cell. In the limit, we might arrive at a cell that responds if and only if your grandmother’s face appears somewhere in the visual field (a so-called “grandmother cell”). The advantage of this approach is that all one would need to do is monitor the cells that were most active at each location and these would reveal just what combination of features was present there. One major disadvantage of local coding is that it would require an implausibly large num- ber of specialized cells to adequately capture the richness of the visual world. An alternative approach, and this is the one the visual system tends to use, is that of distributed coding. Here, there exist cells that will respond to a range of feature values to different degrees. For example, a cell might respond best to an edge oriented vertically, and less well (in a graded fashion) as the orientation is displaced from the optimal one. Such a cell could be said to be tuned to orientation, and the location and width of the tuning function characterizes its sensory properties. A given edge would tend to drive a popula- tion of cells with different preferred orientations. The edge represented by the pattern of activity over that population of cells. This is precisely how orientation representation works, and it is also the basis for the trichromatic theory of color vision, where there are only three different types of color-sensitive cells with highly overlapping wavelength preference func- tions (called spectral sensitivity functions) that can with great precision encode a vast number of different colors by virtue of their relative levels of activity for any given wavelength. These two foundational and related principles of functional specialization and distrib- uted coding will reappear throughout the readings in this volume. In the following sections, some of the progress that has been made in each of the areas outlined earlier is reviewed. Palmer (1999) provides a detailed and highly recommended treatment of these and many other problems in vision. Early Vision Early vision refers to the part of vision that creates the initial representation of elementary sensory properties such as color, the orientation of edges, motion, and depth. These repre-
  • 23. 4 sm Visual Perception sentations tend to be local and veridical (in the sense that they closely mimic the properties that are present in the retinal image itself). This is not to say that these representations are simple, however. Before proceeding, it will be useful to define the concept of a receptive field. This is often a difficult concept to grasp, but it turns out to be absolutely crucial to any real under- standing of how the brain represents sensory information. A receptive field is a property of an individual sensory neuron. The receptive field of a visual neuron is that portion of the retina that, when stimulated, causes a change in the firing rate of the neuron in question. Imagine you are a visual neurophysiologist and you have placed an electrode next to the optic nerve, so you can “listen in” on the activity of a single optic nerve fiber. From time to time you may detect a stray signal from the cell. When you then present a stimulus on the retina (e.g., a small spot of red light, or a moving vertical bar, or even a toilet brush), you may eventually find a location that will cause the cell to increase its firing rate when illu- minated. If you repeat this experiment many times and “add up” all the locations that cause the cell to either increase or decrease its firing rate, you will have defined that cell’s recep- tive field. It is, in essence, the portion of the retina that the cell in question monitors. One also must specify the properties of the stimulus that effectively drives the cell, for example its color, direction of motion, and shape. The extent to which any given cell responds to a visual stimulus depends jointly on where on the retina the stimulus appears, and what the stimulus is. As one measures the receptive field properties of cells at higher levels of the visual system, one finds that more and more complex configurations of light are required to drive the cells. In the retina, cells will respond to small circular spots of light in a very specific location; in V1, cells respond best to small bars of light in a particular location and with a particular orientation. In V4, many cells prefer a corner with a particular color and orientation. In the inferotemporal cortex, many cells seems to prefer highly complex and seemingly arbitrary shapes (Tanaka, 1996). The following is a brief review of what we know about early vision. For present pur- poses, the discussion will be restricted to orientation, spatial scale or size, depth, color, and motion; these are the sensory properties that have received the most attention. The first three of these are aspects of spatial vision and contribute to representations of surface shape, which is crucial for object identification. As pointed out by Palmer (1999), knowing the shape of an object tells you a lot about what the object is, what it is for, even what its color is likely to be, but knowing the color or motion of an object tells you almost nothing about what the object is or about what its other attributes are likely to be. Color and mo- tion, of course, are themselves of great importance in enriching our visual experience and in helping us deal with the inherently dynamic world in which we live. Space The investigation of spatial properties like orientation, depth, and spatial scale began with Hubel and Wiesel’s pioneering investigations (Reading 7) in which they discovered (among many other things) that in primary visual cortex, most cells prefer an elongated, oriented, and specifically located region of stimulation to fire optimally. The representation in V1 was not of spots, but of something more useful for representing shape: oriented edges. It is useful to step back a moment and think about why an explicit representation of oriented edges is so important. The retinal image contains simply a pattern of light and dark distributed in two dimensions (we can ignore color, depth, and motion for now) and sensed by the mosaic of photoreceptors. One can think of this as a 2D array of numbers, each of which corresponds to the brightness of a small location on the retina. The shape, location, and orientation of objects, shadows, and surfaces is not explicitly given in the array of brightness values. In order to recover these properties, some computation must be performed on the array of brightnesses (e.g., taking the difference between adjacent num-
  • 24. An Overview m 5 bers to ask whether the brightnesses there differ, as they would if an edge was present in that location). That is just the sort of computation that retinal ganglion cells (RGCs) and cortical cells carry out. RGCs specify whether a given location has more (or less) brightness than the surrounding locations in all directions. V1 cells specify whether a given location contains an edge (or part of an edge) with a particular orientation and aparticular polarity (e.g., dark to light or vice-versa). It does so by combining the outputs of a set of RGCs that happen to be aligned properly so as to provide a sample of the retina in the corresponding location and orientation. This V1 representation has essentially thrown away the brightnesses in the retinal image. What is left, however, is much more useful: a representation of localized edges. This is useful precisely because a shape is specified by the locations and orienta- tions of its boundaries. The discovery that cortex was in the business of representing ori- ented edge fragments was therefore a landmark in the effort to understand object recognition. Scenes contain (at least) two other sorts of elementary spatial information. One of these is information about the size (or spatial scale) of features. The visual system encodes infor- mation about spatial scale with cells whose receptive fields vary in size so as to be selective for different spatial frequencies. This turns out to be important because of the following mathematical fact: any spatial pattern can be represented as the sum of an appropriately chosen set of spatial frequency patterns called gratings. (This is a nonintuitive idea that is discussed in more detail in the introduction to Reading 8, Blakemore and Campbell, and in the reading itself.) This means that by representing spatial frequencies, the visual system provides some of the building blocks that are likely to be useful for representing shapes and objects. The last kind of spatial information that is contained in scenes is surface depth and orientation. Because objects are typically three-dimensional, it is often not sufficient to represent simply the shape of the object’s outer boundaries (in fact, it is quite difficult to recognize all but the simplest of objects based only on the object’s silhouette). Therefore, it is necessary to know something of the location of the surfaces that bound the object and their distances and orientations relative to the viewer. There are many redundant sources of information in the retinal image, including pictoral cues like linear perspective, overlap, or relative size, relative motjon cues (objects a different depths will have different motion trajectories as the observer moves in the environment), and, perhaps most importantly, stereoscopic depth cues. Stereopsis is the process of combining the two slightly different views of the visual scene that are available at the two eyes (because they view the world from approximately 6 cm apart) to recover information about the relative depth of features in the scene. This requires a process of matching corresponding features in the two eyes (this is the correspondence problem) and estimating the extent to which they fall on corre- sponding or noncorresponding points on the two retinae. By definition, the image feature at the current point of fixation falls on corresponding points in the two eyes (i.e., the two foveas). Any other image feature that falls on corresponding points on the two eyes (that is, points that are equidistant from the foveas in the two eyes) are said to fall on the horopter, an imaginary surface that is defined by all points at the same distance from the observer as the current point of fixation. Any image feature that falls on noncorresponding points in the two eyes exhibits binocular disparity; whenever binocular disparity is nonzero, the image feature in question is at a depth other than the depth of the current point of fixation (or, equivalently, it is either closer than or further away than the horopter). The magnitude of binocular disparity for any given image feature is proportional to that feature’s distance from the horopter. It is known that stereoscopic depth information (specifically, the magnitude of binocular disparity) is available early in cortex. Thus depth information can in principle be combined with information about orientation and spatial scale to provide a rich representation of the spatial layout of edges and surfaces in the scene.
  • 25. 6 @ Visual Perception Color The first major steps in coming to understand color vision were taken early in the 19" century by Thomas Young, and then, independently in the mid 19" century by Helmholtz. The story of the trichromatic theory of color vision and the debate about the alternative opponent-process theory advocated by Herring is contained in the introduction to Reading 6, Hurvich and Jamison. We now know in detail the spectral sensitivities of the three classes of cones and how the cone responses contribute to our ability to match mixtures of colors to one another. This is a theoretical success story of the first magnitude. We also know, based on the work of Hurvich and Jamison and of neurophysiologists since, that the oppo- nent color system that receives input from the cones subserves judgements about color appearance. These two mechanisms can together account for a wide range of color percep- tion phenomena, including color adaptation (the fact that after looking at a red patch fora few minutes, one sees green when subsequently viewing a white paper); simultaneous color contrast (the fact that a gray patch surrounded by bright red will appear slightly green, or a gray patch surrounded by white will appear darker than the same gray patch surrounded by black); Mach Bands, first described in the 19 century by Ernst Mach (the fact that the apparent brightness of the regions surrounding a black-white edge is nonuni- form: the dark side of the edge looks somewhat darker near the edge than do regions far- ther away, while the light side of the edge looks somewhat lighter near the edge than do regions further away); and many forms of color blindness (which can often be attributed to a genetic defect in which one or more of the three cone classes is missing). Motion The first and by far most common kind of motion experience arises when (a) a stationary observer views a moving object or (b) a moving observer views stationary objects (or some combination of these). The simplest case is for a stationary observer with eyes still, look- ing at a stationary scene containing a single smoothly moving object. For this case, simple models of motion representation such as that suggested by Reichert and discussed in the introduction to Reading 18 (Adelson and Movshon) are suitable. Needless to say, however, this kind of simple motion is extremely rare. If nothing else, people tend to move their eyes about 3 times a second, and this causes abrupt displacements of the visual scene even if it contains no motion at all. Furthermore, people tend to move through their environment, and yet they have no trouble experiencing the stable parts of the scene as such and tracking objects that are moving relative to them (e.g., the problem faced by an outfielder trying to catch a fly ball; see McBeath et al., 1995). These forms of motion require more complex systems for coordinating self-motion and retinal image motion. There are several additional motion phenomena, some fairly common, others laboratory inventions, all of which require explanation. These include the motion aftereffect (the fact that after viewing a motion stimulus for several minutes—for example, a waterfall—one sees motion in the opposite direction — in this case, upward—when the view is shifted to a static scene), autokinetic motion (the fact that a small stationary spot of light in a com- pletely dark room will appear to move randomly after a short while), and many forms of apparent motion (motion experiences when viewing a sequence of locations that are stimu- lated in turn but discretely, as in a motion picture). Perceptual Constancy and Organization The early sensory representation of a scene is inadequate for object recognition in at least two major respects. First, it contains representations of image properties and not represen-
  • 26. An Overview m 7 tations of scene properties. Second, the representation is local in scope but any representa- tion of objects necessarily requires a more global perspective to conjoin parts into wholes. Perceptual Constancy First is the problem of perceptual constancy, which can be stated as follows: Most retinal measurements of scenic attributes are affected not only by properties of the object in ques- tion, but also by irrelevant and transient factors such as the current illumination, the dis- tance between the object and the observer, and the object’s orientation relative to the ob- server. To be useful, vision needs to represent object properties and to discount the irrelevant environmental conditions that are not intrinsic to the objects. In other words, how can we correctly perceive objects despite irrelevant variation in viewing conditions? To give just one example of this problem, consider color vision for a moment. Informa- tion about color is carried by the wavelength of light: the spectrum of colors in a rainbow depicts the portion of electromagnetic radiation with wavelengths from about 400 (blue) to 700 (red) nanometers. The color of a surface depends on the distribution of wavelengths in the light reflected from that surface from a light source such as the sun or a lightbulb. For example, a red apple will tend to reflect long wavelengths of light and absorb short wave- lengths of light. So far, so good. We simply need a way to encode different wavelengths, and we can recover the surface color of the apple. Unfortunately, there is a problem. The spectral composition of the illuminant (e.g., the sun) affects the distribution of light reflected from a surface. A white light such as sunlight has about equal energy at all visible wavelengths, but some sources, such as a flourescent lightbulb, have quite imbalanced spectral distributions. If a red apple is illuminated by a source that does not have much energy in the red portion of the spectrum, the distribution of light that itreflects into the eye will be very different than ifitis illuminated by a source, like the sun, with a fairly balanced spectral distribution. This suggests that our perception of the color of various surfaces will depend both on the (desired) surface properties of the object and on the (irrelevant) properties of the illuminant. An apple could appear to change color when we carry it from our yard into the kitchen where it is illuminated by incandescent light. Similar problems arise jn other stimulus domains. Visual information about object size depends on the object’s true size and the object’s distance from the observer, because the retinal image size of the object varies directly with distance. How can we come to know the relevant property (object size) and discount the irrelevant property (distance)? Visual in- formation about object shape depends on the orientation of the object relative to our line of sight (so that a square will project a parallelogram or a diamond onto the retina depending on its slant and tilt). How can we come to know the relevant property (shape) and discount the irrelevant property (orientation)? Each of these is a problem in perceptual constancy. We are, in fact, able to correctly know the color, size, and shape of objects despite variation in irrelevant factors such as illumi- nation, distance, and orientation. The question is how this is achieved. It turns out that the visual system has evolved different solutions in each domain, specialized for the particular properties in question. Although it is implied that the mechanisms of perceptual constancy are not part of early vision by placing them in separately labeled sections, most of percep- tual constancy is indeed achieved by what most vision scientists would call early vision. Color constancy, for example, begins as early as area V4, and certain aspects of color constancy, its precursors so to speak, arise even earlier (see Zeki, 1993, for a discussion). Perceptual Organization The second main problem of the early sensory representation is that it is local in scope. This is a problem because most real scenes contain multiple objects that are arranged in 3D
  • 27. 8 Visual Perception space and that both occlude one another and occlude parts of themselves. To take an eXx- treme example, consider the problem of looking at your garden through the partially open slats of a Venetian blind. Although the tree and foliage in the garden are broken up into horizontal strips by the blind, we are able to attend to and recognize the objects in the garden with little difficulty. Any understanding of this ability requires us to explain how the visible part of a branch in one horizontal strip are correctly conjoined with the corre- sponding part of the branch in the next horizontal strip, and more puzzlingly how the occluded parts of the branch are “filled in” to produce a representation of a single, coher- ent, solid branch despite extreme fragmentation in the retinal image. Many of the Gestalt principles of perceptual organization are discussed in Reading 11 by Wertheimer. These principles characterize how the visual system puts elements in the retinal image together so as to form coherent object representations. These principles in- clude similarity (the texture and color and depth of the branch are similar at all points along it, and furthermore those properties are different for nearby image regions; thus similarity of surface properties is a good clue that those regions all belong to the same object), proximity (the visible parts of a reasonably compact object tend to be in nearby regions of space, so all other things being equal, two image regions will tend to be near one another if they are part of the same object), and common motion (if the branch is blowing in the wind, all its parts will move together, and other non-branch objects, such as the fence behind the branch, will tend not to move with it; thus common motion in two regions is a good clue that they are part of the same object). The problem of perceptual completion or “filling in” has also been investigated by per- ceptual psychologists. There are two sides to this problem. One is that of what Michotte et al. (1991) called amodal completion. The partly occluded branch behind the blind is expe- rienced as complete and continuous despite the fact that parts of the branch are completely absent from the retinal image. This is called amodal completion, because there is no infor- mation coming from a sensory modality to support perceptual completion, and in fact there is a sense that one “knows” that the branch is complete without having an explicit visual experience of the branch in the occluded regions. Amodal completion depends on two factors: (1) the edges of the occluded objects must be “relatable” such that they could plausibly be considered part of a single common contour (this essentially requires that there are no implausibly large bends in the contour; see Kellman & Shipley, 1991), and (2) the boundary between the occluding surface and the occluded surface must “belong” to the former and not the latter (this essentially requires that the occluding surface be perceived as in front of the occluded surface; see Nakayama et al., 1995). The other form of perceptual completion is the perception of illusory contours, a phe- nomenon first described by Kanizsa (e.g., 1979). Illusory contours arise most famously in the context of the Kanizsa triangle at right. Here the seg- mented disks serve as inducers for the illusory shape. The shape is illusory because there is no physical contrast to & a define the edges of the triangle where the page is white, yet most observers report that they see a dim but clearly ” discernable edge there. Furthermore, the brightness of the figure is often reported as being greater than the brightness ae of the background. Clearly the relatability of the bound- ne aries of the segments (i.e., that they are nearly collinear) supports the perceptual conclusion that a curved white tri- os angle is partly occluding three black disks (which, by the way, are perceived as complete disks by virtue of amodal completion!). The visual system “fills in’ where the con- G8 trast of the perfectly white triangle with the perfectly white
  • 28. An Overview m 9 background is zero, and makes it slightly greater than zero. Von der Heydt et al. (1984) have provided evidence about the neural basis of illusory contours in V2. Object and Spatial Vision Recognizing objects visually is a difficult problem for the following reason: the data upon which we must base our identification decision is a retinal projection that can vary dra- matically depending on our viewpoint relative to the object (not to mention the lighting, where we are looking, whether we or the object are moving, and so forth; we will ignore these complications because we have enough trouble as it is). Even something as simple as a cube can project a nearly infinite variety of retinal images depending on its orientation and its distance from us. So there is a highly underdetermined many-to-one mapping of retinal images onto the object’s true shape. Obviously the visual system has found a way to solve this problem with staggering efficiency and accuracy. The question is: how? There are two major routes one can take in approaching this problem. They can be broadly classified as view-based_.or part-based models. View-based (also called image-based) mod- els assume that internal representations of whole objects as they would appear to the ob- server are stored as such in memory. One approach to dealing with the many-to-one prob- lem discussed in the last paragraph is to store in memory, for every object one might wish eventually to recognize, all the possible retinal images that object could produce (or at least the images one is likely to encounter under normal circumstances, which reduces the number of possibilities from infinite to merely unimaginably large). This is the idea behind simple template matching accounts of object recognition. There are, of course, many prob- lems with this approach. First, it would require a very large number of stored templates: more than one could plausibly suppose are stored in the brain. New instances of a familiar category (a strange but recognizable chair, for instance) could not be recognized by this mechanism. Finally, there is no evidence for the sort of unguided search through a huge database of templates that would be required by this mechanism. So this simple brute- force account of object recognition will not work. A more sophisticated version of template matching, however, does have considerable merit. According to this ide&, there exist internal representations of objects depicted in one or, more often, several canonical views (that is, views that permit one to represent most of the critical parts of the object from a view that is often encountered in real life). The reason multiple views are necessary is because any one view usually will not contain information about the part of the object facing away from the viewpoint, and of course we know per- fectly well what the back of a chair looks like even when we can’t see it (or at least we can make a very good guess). When an object is to be recognized, it is first normalized, that is, it is transformed so as to match the size, location, and viewpoint that is likely to be stored for that object, and this normalized representation is then compared to stored object repre- sentations. This mechanism avoids some of the problems of the simple template matching models by allowing for normalization so that only a small number of views of each object need be stored. A very different approach to object recognition is offered by feature- or part-based mod- els of object recognition. According to these theories, object representations consist of lists of features (e.g., oriented lines, angles, the presence or absence of symmetry) or parts (e.g., volumetric primitives such as cylinders), together with some specification of the relations among the parts. These accounts deal with the many-to-one problem by virtue of the fact that they are inherently more abstract than view-based theories: they transform the retinal image into a set of symbolic descriptions that do not depend on rotation, size, and so forth. One major subclass of part-based accounts are structural description theories. According
  • 29. 10 ® Visual Perception to these theories (advocated by Marr and Nishihara, 1978, among others), stored object representations consist of 3D models of objects made up of a set of volumetric part primi- tives such as cylinders, cones, cubes, etc. that are arranged according to the structure of the object itself. The representation is not viewpoint dependent. They are called descriptions because they are not depictive (as in the multiple view plus transformation models). They are abstract and consist of a list of the parts together with a description of the structural relations among the parts (e.g., “cone on top of cylinder”). These models completely by- pass the problem of too many representations by using fairly abstract representations that are insensitive to such things as absolute size or viewpoint. As long as the parts and their relations can be recognized, a match to a stored representation is possible. Biederman (Reading 20) describes a detailed model of this type, and makes the case for it on both theoretical and empirical grounds. These alternative accounts of object recognition are fundamentally different in certain respects, and one might think that if one is the “correct” theory, the other must be wrong. However, it is possible, even likely, that both kinds of theories are correct to some extent (in a way analogous to the trichromatic and opponent-process theories of color vision), and that the challenge will be to specify the conditions under which each class of theories best accounts for recognition performance. Attention and Awareness To many people, attention and awareness are one in the same. We are aware of that to which we attend. Ever since Helmholtz and James first discussed these concepts, however, they have been treated differently by the scientific community. A look at the history of investigation of the two phenomena reveals a long and continuous line of scientific inves- tigation into attention. In contrast, awareness (and its related term, consciousness), while of sustained interest in psychology for the past century, has until recently been ignored by neuroscience. In the last 10 years, however, that imbalance has changed, and there has been a surge of interest in the neural substrates of visual awareness. Attention Attention is a term with a high potential for confusion, because it can mean different things to different people. One often hears the comment at scientific meetings that the term is meaningless because it has too many meanings. Perhaps the most common use of the term, and the meaning intended here, is that of perceptual selectivity, described below. Other meanings include vigilance, or the ability to focus for extended periods in anticipation of a signal (as when the operator of a power plant monitors for a particular warning light), and multi-tasking, or the ability to switch between two overlapping tasks, like driving and carrying on a conversation. Attention as perceptual selectivity refers to the problem first brought to prominence by William James at the end of the 19th century: we are confronted with far too much sensory information to grasp coherently at one time. We must select what is relevant and ignore what is irrelevant (which will tend to be almost everything). That means that until you started reading this sentence, you were unaware of the pressure of the chair you are sitting on against your skin, but now, with only a shift in attention to that source of sensory input, you are aware of it. Or of the faint sound of a vehicle out the window. And so forth. These kinds of attentional shifts are purely top-down and goal-driven (that is, you presumably set up, at least informally, a perceptual goal to give priority to certain tactile or auditory inputs and to ignore other inputs that might have been of higher priority previously, and this goal
  • 30. An Overview m II serves somehow to modulate the relative strengths of the sensory inputs available to you). Of course, the stimulus itself may well be capable of modifying your attentive state—an unexpectedly loud handclap will capture your attention in a purely stimulus-driven fash- ion. In both instances, however, the result is that certain sensory inputs are selected and enter awareness, and others are excluded. As implied by the last sentence, the gatekeeping act of selection and its result (to permit sensory information to enter awareness) are dis- tinct; hence attention and awareness are distinct concepts. But why is there a need for attention? What is the limitation that requires us to select some objects and ignore others? Broadbent (1958), in his seminal book that brought the study of attention into the modern era, suggested simply that people are information chan- nels in the sense that we transmit sensory information into a behavioral response, and information theory states that any information channel must have an upper bound on its capacity. A more concrete answer was suggested by Treisman and Gelade (Reading 21), who were inspired by the finding of functional specialization by neurophysiologists in the 1960s and 70s. They pointed out that if visual objects are represented in terms of separate “feature maps,” one for color, another for orientation, and so forth, then at some point there must come a stage of reintegrating the separate features into a unified and coherent repre- sentation of the object we actually see. They proposed that the process of feature integra- tion was carried out by attending to one location at a time (and ignoring all other loca- tions), so that only the attended features there were strongly represented at that moment. This constitutes attentional selection, and results in feature integration. This was a groundbreaking idea, and it inspired a generation of experimental psycholo- gists to explore its consequences. The problem of feature integration came to be called the binding problem and it continues to be a subject of considerable interest 20 years later (see Roskies et al., 1999, and the introduction to Reading 21). Many of the details of Feature Integration Theory have been updated as a result of subsequent empirical work, but the basic idea remains nearly intact. Investigation of attention over the last decade or so has included a great deal of new work on the neural basis of attention inspired by Moran and Desimone (Reading 22), to- gether with advances in understanding how attention is controlled. This latter problem is a major challenge now, and begins to approach the problem of consciousness because it requires one to think about how goals are specified, and how acts of the will are imple- mented. Descartes proposed that communication between the Soul (where acts of the will originate) and the Body is mediated by the pineal gland. Today this idea is viewed as a quaint anachronism, and other approaches are being considered, including the notion that a working memory store that is probably located in the dorsolateral prefrontal cortex con- tains representations of goals that can cause feedback signals to sensory regions that in turn can modulate the strength of neural representations there. This remains an active area of investigation that is really just beginning. Awareness If attention is perceptual selection, then what is consciousness? Consciousness is generally broken down into smaller problems, some of which will be harder to solve than others. Philosophers (e.g., Chalmers, 1996) have identified the “hard problem” of consciousness: the problem of qualia. How do we experience the redness of red? As Newton said, there is nothing in the light that is what we would call red; there is only a wavelength that happens to have the ability to evoke a pattern of neural response in the three classes of retinal cones. How does that pattern get translated into a subjective experience like red? Nobody has any idea (some would say this is a fundamentally intractable problem), and so this will be left to the philosophers for now. But the problem of awareness, that is, of
  • 31. 12 m Visual Perception how certain kinds of neural activity correspond to something we experience and can report verbally or otherwise, seems more tractable (see Crick and Koch, 1998, for a discussion). The question can be stated as follows: some of the neural activity in the brain correlates well with our visual experience, that is, with what we are seeing at any given moment, and much of the neural activity in the brain is unrelated to what we are seeing. What is the difference between the parts of the brain (cells, layers, areas) that are and are not correlated with visual awareness. What is it about the type of cell (pyramidal?), the neurotransmitters it uses (serotonin? dopamine?), its connectivity pattern, its location in cortex (layer 4?; area V1?), some combination of these, or something else that endows it with the power to cause visual awareness? This is a new area of investigation; little is known. Investigators have had to be clever to devise experiments that will be revealing in this regard. Sheinberg and Logothetis (Read- ing 25) is a good recent example. Methods in Vision Science I conclude with a discussion of how different methods have contributed to advances in vision science. As in all sciences, theoretical and empirical advances go hand-in-hand with technological developments. The development of the telescope permitted astronomers to see celestial objects previously unimagined. The microscope provided biologists with a tool to investigate cellular phenomena at the foundation of that science. So it is with any science. Visual perception has been no exception. Psychophysics and Behavior The first methodological developments in perception research were behavioral. In his 1860 treatise Elements of Psychophysics, Fechner described a collection of techniques for mea- suring with precision the relation between physical stimulus magnitude (e.g., the bright- ness of a light) and perceptual experience by collecting behavioral responses from observ- ers in perceptual experiments (e.g., stating whether or not a dim flash was seen). These methods have withstood the test of time and continue to be used (in somewhat modified form) to this day (see Readings 6 and 8 for good examples). Psychophysical experiments provide the very bedrock of vision science, for they provide the precise measurements that are needed of the perceptual phenomena that require explanation. The criterion by which the usefulness and value of any psychophysical technique can be judged is whether that technique provides a measurement of experience that is as direct as possible and free of subjectivity and bias. Observers often have implicit theories of how vision works, and if these are permitted to contaminate psychophysical measurements, they are likely to lead to improper inferences. The development of signal detection theory (Reading 3 by Tanner and Swets) was inspired by precisely this concern. Further developments during the 20th century gave rise to psychophysical methods that allowed for more precise control of the stimulus than those available to Fechner and his contemporaries. For example, the development of the tachistoscope provided a means of presenting one or more visual stimuli of precisely known brightness and duration to an observer. Such a device was essential for investigating temporal phenomena such as visual masking, where two objects are presented in rapid succession and the effect of one on the visibility of the other is measured. Many other devices were invented to serve a huge vari- ety of special purposes in the investigation of visual function. In the 1960s and 70s, digital computers entered the laboratories of vision scientists and eventually replaced the tachistoscope and other special-purpose devices. The reason is simple: general-purpose computers permit almost infinite flexibility in creating and pre-
  • 32. An Overview m 13 senting stimuli, and largely (but not completely) did away with the necessity to invent and build a new apparatus for each new experimental question, greatly accelerating research efforts. Neuropsychology All of the developments described in the previous paragraphs contributed to characterizing how variation in the properties of physical stimuli caused changes in visibility or discriminability as measured by behavioral methods. In parallel with these, a multitude of methods to investigate the neural basis of visual perception emerged. The earliest modern approach to brain-behavior relations was the proposal by Franz Joseph Gall that the brain consists of a number of functionally distinct compartments each of which was responsible for a particular kind of behavior (including such faculties as parental love, hope, and de- structiveness). He argued that when a given faculty was exercised, the corresponding part of the brain grew larger (just as a muscle, when exercised, grows larger), and that this in turn was reflected in bumps on the skull, which grew to accommodate the brain tissue beneath them. Gall attempted to assign function to different brain areas by correlating personality traits to bumps on the head, a field of investigation known as phrenology. Pierre Flourens tested these notions by removing parts of the brains of experimental animals, and he concluded that the brain is not functionally compartmentalized at all, but is instead functionally undifferentiated. According to Flourens, any part of the brain can carry out any function as well as any other. Although Flourens’ conclusion has the distinct merit that it opposes the highly implau- sible theory of Gall, it was soon shown to be wrong. In 1861, Pierre Paul Broca (1824— 1880) reported his observations of a patient who had suffered damage to his left frontal lobe (discovered after autopsy upon the patient’s death) that had prevented him from speak- ing, while sparing most other cognitive functions, including the ability to understand speech. Broca asserted that the frontal lobes contained the faculty of speech production. This idea of functional localization was met with much skepticism at the time (a continued reaction against Gall), but it was corroborated with subsequent observations. For example, in 1876, Carl Wernicke described a patient who had difficulty understanding speech but not in pro- ducing it, a pattern complementary to that reported by Broca. Wernicke found that this patient had a lesion in the posterior temporal lobe, now called Wernicke’s area. This double dissociation provided strong evidence that the perception and production of speech are carried out by functionally distinct brain regions. Among the first evidence for functional specialization in vision was reported by the French ophthalmologist Louis Verrey who, in 1888, described his observations of a patient who had suffered a stroke in her left visual cortex. Among other less dramatic effects, the patient lost her ability to see color in the right visual field—all the objects there were seen in shades of gray. Color vision was normal in the left visual field. The acquired loss of color vision due to cortical damage (in this case, stroke) is known as achromatopsia. The existence of this syndrome strongly suggests that there is a local brain region that is neces- sary for the experience of color. It is important to keep in mind that other parts of the visual system are also necessary for normal color vision (for example, the cone photoreceptors in the retina are certainly necessary for color vision!), but this brain region appears to be specialized for color because damage there does not much affect other visual abilities, such as form or motion perception. Inspired by the findings of the neurologists, Korbinian Brodmann analyzed the anatomy of the human brain around the turn of the century using the method of cytoarchitectonics, which relies on differences in the form and structure of cortical neurons. He identified 52 distinct areas and suggested that these areas may well be functionally distinct as well. Brodmann’s classification scheme continues in wide use to- day.
  • 33. 14m Visual Perception Functional specialization in the brain is now as close to a fact as anything in science can be. Brain-damaged patients have been described with specific deficits in motion percep- tion, color vision, the perception of faces or objects, and in the ability to attend to specific regions of space, among many other syndromes, suggesting that each of these functions may rely selectively on the damaged region of the brain. Lissauer (Reading 16) provided an insightful early analysis of a patient with cortical damage that caused not blindness, but an inability to recognize objects that could nevertheless be seen, a condition known as visual agnosia. The blindsight patient of Weiskrantz et al. (Reading 24) revealed a surpris- ing dissociation between explicit visual awareness and measured visual performance. Subtle deficits such as these have turned out to be extremely informative in developing accounts of how the normal visual system works. Neurophysiology Vision is an emergent property of neural activity in the brain. Acomplete understanding of visual perception, therefore, will require an understanding of how neurons subserve vi- sion. Of course, there is much to learn by analyzing visual perception through purely be- havioral means using the methods of visual psychophysics and cognitive neuropsychol- ogy. But despite its almost infinite complexity, the brain is a concrete medium in which visual behavior is implemented and it can therefore provide important clues about the principles of visual perception. The development of the microscope provided physiologists with a crucial tool that led to the idea, first promoted by Santiago Ramon y Cajal, that the brain consists not of continu- ous and undifferentiated cellular tissue but of billions of separate nerve cells that commu- nicate with one another at specialized junctions called synapses. Johannes Miller, Hermann von Helmholtz, and others learned that axons carry electrical signals from the cell body to the synaptic gap, which can in turn activate the next cell. Neurons are exquisitely complex biological devices whose functional properties are a result of a multitude of molecular, chemical, and electrical forces acting in concert. Mod- ern neuroscience is dedicated to uncovering how the interactions among these forces gives rise to patterns of activity in single neurons. However, for purposes of understanding how neural activity produces visual function, one can view a neuron as a device that produces a discrete signal (an action potential that can be measured with a microelectrode) in response to light (in the case of the photoreceptors) or to the release of neurotransmitters at the synapse (in the case of all other neurons). Cellular neurophysiologists place electrodes in or near the axon of a neuron and record action potentials as they occur. The rate with which action potentials are produced by the cell provide a compact measure of the activity of that cell, and this in turn can reveal how that cell participates in any given perceptual act. For example, as suggested by Barlow (Reading 2), one can characterize a cell’s function by determining what sorts of visual stimuli cause that cell to fire most effectively. The neuro- physiological studies of Hubel and Weisel (Reading 7), Newsome et al. (Reading 10), Moran and Desimone (Reading 22), and Scheinberg and Logothetis (Reading 25) each provides an excellent example of the insights to be gained from single-cell recording com- bined with carefully designed stimuli and behavioral tasks. Functional Neuroimaging Until the 1970s, direct measurements of brain function in behaving organisms was only possible in two ways. First, one could analyze patterns of impaired performance in brain- damaged patients, relying on “experiments of nature” such as a stroke or closed-head in- jury to reveal what the intact system must be like in order that it could be damaged in such a way as to have produced the observed functional deficit (e.g., damage causing cortical
  • 34. An Overview m 15 achromatopsia but not affecting the perception of shape or motion suggested that there exists a distinct “color center” in the brain). The main disadvantage here is that true experi- ments are not an option (one had to seek out patients that happened to have had a stroke in a focal region that happened to correspond to a functional area). Second, one could per- form experiments using animals, recording from single cells or producing focal lesions in cats or monkeys. The disadvantages of the single-cell approach is that although single cell responses are an extremely rich source of data, the evidence is quite local so that global cortical circuits are difficult to study without great effort. The difficulty with lesions is that the interpretation of a pattern of impaired performance is not always unambiguous. Fur- thermore, there are limitations in what tasks monkeys can be trained to do (for example, language tasks are not available). Finally, the anatomy and function of monkey and cat brain differ from those of human brain in both known and unknown ways. Over the last several decades, neuroscientists perfected the use of the electroencephalo- gram (or EEG) to measure electrical field potentials at the scalp that are caused by the activity of millions of neurons in cortex several millimeters below the electrode’s scalp location. This technique provides a highly detailed record of the timecourse of brain activ- ity, and has been used very effectively (e.g., Hillyard et al., 1996). The data one obtains from this technique are similar in many respects to single cell recording in monkeys, en- joying the same very high temporal resolution of that technique. It has several advantages over single cell recording: it can be carried out in humans, and it provides measurements over the entire surface of the scalp, potentially overcoming the inherently local measure- ment that emerges from single cell recording. However, the event-related potential suffers from its comparatively low and sometimes uncertain spatial resolution. Magnetoencephalography (or MEG) has recently emerged as a new technology that may one day replace the EEG. It operates on principles that are analogous to the EEG, but instead of measuring electrical potentials, it measures extremely small magnetic field in- homogeneities produces by the electrical activity of neurons in cortex. This technique ar- guably has better spatial resolution than EEG, and similar temporal resolution. It also can provide a map of activity throughout the brain. One of its drawbacks is that it is currently a fairly expensive technology that is available at only a few sites. This is perhaps the new- est of the brain-imaging technologies and therefore the least is known about it. Finally, two techniques that have emerged over the last two decades have generated great interest although their full potential as scientific tools to investigate vision has not yet been reached. Both methods are based on changes in blood flow or in the composition of oxygenated or deoxygenated hemoglobin within the brain that is caused by changes in neural activity. In rough outline, when neurons produce action potentials, they require oxygen and other nutrients that are carried by the blood. They signal their need for blood, and this in turn causes changes in blood flow in those regions. Thus both of these methods are indirect, although surprisingly precise, measures of the location of neural activity. The first method is positron emission tomography or PET, which was used by Zeki et al. (Reading 9). It requires an injection of a radioactively labeled agent (usually oxygen in water) that is carried by the blood to the brain. The PET scanner can detect the location and concentration of the radioactive agent, and thereby pinpoint the region of neural activity. A major advantage of PET over EEG is its improved spatial precision. The disadvantages of PET are the invasive use of a radioactive substance, which is potentially harmful, and the very low temporal resolution (on the order of several minutes). The second of the two blood-based methods is functional magnetic resonance imaging, or fMRI, used by O’Craven et al. (Reading 23). It exploits the fact that deoxygenated hemoglobin is paramagnetic and can be detected by an appropriately programmed MRI scanner (see Haxby et al., 1998, fora tutorial review). This is relevant because the concen- tration of deoxygenated hemoglobin changes systematically where there is neural activity. fMRI has better spatial and much better temporal resolution than PET (on the order of a
  • 35. 16 # Visual Perception few seconds) and it does not require an invasive injection. It also has better spatial resolution than EEG. However, {MRI has much poorer temporal resolution than EEG or MEG, and it is currently quite expensive to carry out a study. On the other hand, most modern medical cen- ters own MRI scanners for clinical purposes, and so they are often more readily available than MEG scanners. } The techniques for studying vision, ranging from behavior to brain imaging, are comple- mentary rather than competing. Each represents a tradeoff between cost, spatial and temporal resolution, invasiveness, and other factors. Each requires a unique approach to designing ex- periments, and the range of questions that can be addressed differs markedly across the meth- ods. The only generalization that can be made is that vision science can only benefit from the development of new methods and new technologies that will allow new questions to be ad- dressed and old questions to be re-examined in new ways. About this Volume The articles selected for this volume are among the most influential writings on visual percep- tion published in the last 125 years. These 25 articles were selected from an initial list of more than 200 classic articles culled from many sources. In the end, the selection reflects a need to provide reasonably complete coverage of the major themes and methods in visual perception, and a desire to include articles that are accessible by non-experts. Also included are some articles from the last decade to reflect some recent trends in the field (e.g., the use of func- tional magnetic resonance imaging to measure human brain activity in vision). Obviously many important topics and many influential works have been omitted. It would not be diffi- cult to compile three more volumes of readings of equivalent quality and impact. Each of the five sections concludes with a set of questions designed to stimulate discussion of the articles contained in that section. In addition, several additional articles are listed that could easily have been included in this volume if only there had been room. These, together with the references following each of the article introductions, constitute a larger set of essen- tial readings in visual perception that readers are urged to seek out on their own. In order to include as many articles as are included here, it was necessary to reduce the length of some articles through editing. In each case where this was done, the omitted mate- rial included experiments that, while they contributed importantly to the theoretical point that was being made, were not essential to the reader’s understanding of that point. I benefitted greatly from the advice of many colleagues during the process of selection. Stephen Palmer provided an exceptionally detailed and scholarly evaluation of an early list of proposed articles for which I am grateful. Richard Abrams, Jim Enns, Bruce Goldstein, Patrick Green, Glyn Humphreys, Eileen Kowler, William Newsome, Harvey Schiffman, and Nick Wade offered valuable feedback and ideas for readings I hadn’t considered. My colleagues at Johns Hopkins — Ed Connor, Howard Egeth, Mike McCloskey, Brenda Rapp, and Trish Van Zandt — were helpful and patient in our conversations over the months I was engaged in this project. To all I offer my thanks. REFERENCES Broadbent, D. E. (1958). Perception and communication. London: Pergamon Press. Chalmers, D. J.(1996). The conscious mind: In search of a fundamental theory. New York: Oxford University Press. Crick, F., & Koch, C. (1998). Consciousness and neuroscience. Cerebral Cortex, 8, 97-107. Fechner, G. T. (1860). Elemente des psychophysik. Leipzig: Breitkopf and Hartel. (Translated by H. E. Adler, Elements of psychophysics. New York: Holt, Rinehart, & Winston, 1966) Haxby, J. V., Courtney, S. M., & Clark, V. P. (1998). Func- tional magnetic resonance imaging and the study of atten- tion. In R. Parasuraman (Ed.), The attentive brain (pp. 123— 142). Cambridge, MA: MIT Press. Hillyard, S.A., Anllo-Vento, L., Clark, V. P., Heinze, H. J., Luck, S. J., & Mangun, G. R. (1996). Neuroimaging ap-
  • 36. proaches to the study of visual attention: A tutorial. In A. F. Kramer, M. G. H. Coles, & G. D. Logan (Eds.). Converg- ing operations in the study of visual selective attention (pp. 107-138). Washington, DC: American Psychological As- sociation. Kanizsa, G. (1979). Organization in vision: Essays on Ge- stalt perception. New York: Praeger. Kellman, P. J., & Shipley, T. F. (1991). A theory of visual interpolation in object perception. Cognitive Psychology, 23, 141-221. Marr, D., & Nishihara, H. K. (1978). Representation and rec- ognition of the spatial organization of three-dimensional shapes. Proceedings of the Royal Society ofLondon B, 200, 269-294, McBeath, M. K., Shaffer, D. M., & Kaiser, M. K. (1995). How baseball outfielders determine where to run to catch fly balls. Science 268, 569-573. Michotte, A., Thinés, G., & Crabbé, G. (1991). Amodal suggested Readings | An Overview m 17 completion of perceptual structures (E. Miles & T.R. Miles, Trans.). In G. Thinés, A. Costall, & G. Butterworth (Eds.), Michotte’s experimental phenomenology ofperception (pp. 140-167). Hillsdale, NJ: Erlbaum. (Original work published 1964) Nakayama, K., He., Z. J., & Shimojo, S. (1995). Visual sur- face representation: A critical link between lower-level and higher-level vision. In S. M. Kosslyn & D. N. Osherson (Eds.), An invitation to cognitive science. Vol. 2: Visual Cognition (pp. 1-70). Cambridge, MA: MIT Press. Roskies, A. L., et al. (1999) The binding problem. Neuron, 24, 7-125. Tanaka, K. (1996). Inferotemporal cortex and object vision. Annual Review of Neuroscience, 19, 109-139. von der Heydt, R., Peterhans, E., & Baumgartner, G. (1984). Illusory contours and cortical neuron responses. Science, 224, 1260-1261. Zeki, S. (1993). A vision of the brain. Oxford: Blackwell. The following books provide excellent overviews of vision science from various thematic and methodological perspectives. Farah, M. J. (2000). The cognitive neuroscience ofvision. Malden, MA: Blackwell Publish- ers, Inc. Gordon, LE. (1997). Theories of visual perception (2nd Ed.). New York: John Wiley & Sons. Gregory, R. L. (1970). The intelligent eye. New York: McGraw-Hill. Hubel, D. H. (1988). Eye, brain, and vision. New York: W. H. Freeman & Co. Marr, D. (1982). Vision. San Francisco: W. H. Freeman & Co. Palmer, S. E. (1999). Vision science: Photons to phenomenology. Cambridge, MA: MIT Press. Rock, I. (1984). Perception., New York: Scientific American Books. Zeki, S. (1993). A vision of the brain. Oxford: Blackwell.
  • 37. ii om woer Pb ie : ott hh colt Sica “pe hi Iepbs ; ; Pope ss irs Lee Wes ie « . | A) north Aa Aa 'sirheae, Vinee noe aa Leadbatliny wire Rantidlby Torgeen Gee praca tee a siete wea _ ae ’ ye Cap ‘nda — don leper? § MS.Ee ure = nil J S antipier “A co. a tie wag Lvond raver inroe 7" ‘ ely spy ‘ i : : wate a VAP IA ST aes cane 10 He ze bus wan 6 MOP ieting D4 vb, FE andy WEUICpre agers nate) ay Be a> “ASotlheLapeantitagea>’ ors. Ep fA SEQUEN BY penta Ansa) gate =| at OO Lake Be gtate ite: cote hate ‘ ny = eee seta bagi ae pe in a A a apart 5 wine i~ tae mr WPA 2), per teh ue ie saese (ety Leslidenwtiiadl Jy 1) cere Gr hacen aimete Bd ae ee : ern will gaonn dite “a eeedGhD §nares An GO?atwia CMe ots £78 : sain ee ed 4 ey Ae ; iMitatety BAT: Tinie aia Lh hgh rckalt @ ynvaly & Atv I) 1 ss tenet a LfetHD : ; ‘ ya Tapert dire Vala tytareomtcag det ape intheiad ig ;Dwll et ta rent SSS year. These 2S egies peg ‘nigh is Tae = <t Thou 2h a) ied Sneen ans NY IOs rh ify (crete « ratty “amg? of Catia ibeines weg Te oitertiadt Coppa) ee vewmsiere Se TKR “ailsaes, hyts : Dad & Pale 4A* feat sas Pa sas inal iaantaalt! ¢Ati. cues sits ardenHORA ny icp u LK : oy eal Udy whigre rad Dior Peeve: Green CMW padyaadabed, --y 8 4 “3£0) sotaensG Sty alah Aas E- a es was featths Warha aot One bei te. ; c Ped Lary gt Bt. std Leas ae 0G 4 aan Moho wetupneia SAB rye ra af With tc. 2 ely 59 Yh AMY poet pac it TIM LAA yapsaresios “) Topi PAL sph ethos torn pom ~. ie? wp many ox UG LU oere, 1 vee Legeds oP ics 2 igMange)5 Ugieio® BTavs vai Wyn MA} eh, Wy Rey ia dass) thee yy : veres Conds} - xe, Spas he vader s anhkey peters om cht pg | ical ice of tne) ioaget drink,the ema Sey, 1 SB WPRMIEAL ii Weil asterwarly ¢ Bak rea rt Wyss : 2 Rn, Pita dere Ji Ba) Og (in ‘ j + niles ee wah (Seer nestled oe ty Wate © — we ae s oy a | ancadeaesl. ge * Lotuye 3 ‘ pe Vers, ihr eta eekosDac Rap “be é : _ a id ait ise. ube iacen f
  • 40. Introduction toReading | oq Hermann von Helmholtz (1821-1894) was responsible for half a dozen truly seminal contributions to science, any one of which might have been worthy of a Nobel Prize. This includes the trichromatic theory of color vision (which he discovered independently of Thomas Young; see the introduction to Reading 7 by Hurvich and Jamison for more details), the place theory of pitch perception (which was corroborated a century later by the Nobel Prize-winning work of Georg von Békesy), and the first measurement of the conduction velocity of a neural signal (about 100 meters/sec). Among his most influential contributions is the monumental Treatise on Physiological Optics, in which he analyzes the functioning of the visual system at many levels in exquisite detail. This selection is‘taken from Volume Ill of that work, subtitled The Perceptions of Vision. It contains one of Helmholtz’s most enduring theoretical contributions: his articulation of the constructivist perspective (what he terms the empirical theory) in perception. According to this view, perception is not merely an “internal picture” of the retinal image; instead, it is a joint result of sensory input on the one hand and what the observer knows and expects on the other. Each of these provide evidence about what is actually present in the world, and a form of perceptual thought that Helmholtz called unconscious inference gives rise to veridical experiences despite the limitations and distortions of our senses. As Helmholtz puts it, “such objects are always imagined as being present in the field of vision as would have to be there in order to produce the same impression on the nervous mechanism, the eyes being used under ordinary normal conditions” (p. 25). The alternative to the empirical theory Helmholtz calls the intuition theory, according to which perception does not depend at all on experience or expectation, but is wholly innate Pil
  • 41. 22 Visual Perception and stimulus-driven, He argues persuasively at various points in this selection against this position. He notes, for example, that one’s ability to navigate about a room in twilight is dramatically improved if one has had previous experience viewing the layout of the room under good illumination. One’s memory of the layout, combined with the indistinct impressions of the dimly illuminated furniture, combine to produce an accurate perceptual experience. A standard example of unconscious inference is its manifestation in size constancy. The problem solved by size constancy is this: the retinal image size of an object depends directly on the distance between it and the observer: the closer the observer, the larger the retinal image size. Indeed, any particular retinal image is consistent with an infinite number of possible size/ distance combinations (small/near or large/far). Yet observers routinely make correct judgements about the true size of objects. This fact of size constancy implies that size judgements cannot be based only on the retinal image, but must also take into account the perceived distance to the object (see Reading 14 by Kaufman and Rock for more on size constancy). Helmholtz’s explanation for this and many other phenomena is that the observer unconsciously combines the retinal image size with an estimate of the distance to the object, based perhaps on stereoscopic or other depth cues. Helmholtz discusses some of the ideas behind the notion of unconscious inference (or, as translated here, unconscious conclusion) that were later to have such a significant impact on the field and continue to hold sway today. He uses as an initial example the effects accompa- nying mechanical stimulation of the retina (e.g., a finger press on the outer edge of the eyeball). The pressure activates a small set of photoreceptors, producing what are called “phosphenes.” The perceptual experience is not of an event occurring at the point of stimulation, however, but of light coming from the direction of the bridge of the nose; such a light would, if present, have stimulated the part of the retina that is being mechanically stimulated. The brain “concludes” that there must be light in the scene in a location that would stimulate the active part of the retina. In addition to laying out the principle of unconscious inference, Helmholtz brings to the argument examples from his own observations of a wide range of perceptual phenomena, anticipating by more than a century some issue of great contemporary interest. These include the “ohantom limb” syndrome in which an amputee experi- ences sensations (often pain) that appear to come from the limb that is no longer there (é.g., Ramachandran, 1993): the perception of biological motion (e.g., Johansson, 1973); and filling in of the blind spot (the location on the retina where the optic nerve leaves the eye and where, therefore, there are no photoreceptors; e.g., Ramachandran, 1993). He dismisses the notion of isomorphic brain representations, and what Dennett (1991) referred to as the “Cartesian theater” a century later: “Now |ask, what similarity can be imagined between the process in the brain that is concomitant with the idea of a table and the table itself? Is the form of the table to be supposed to be outlined by electric Currents? . .. such an electrical reproduction of the table in the brain would be simply another bodily object to be perceived, but no idea of the table” (p. 35). It should be emphasized that in Helmholtz’s view, unconscious inference occurs not only when there are several distinct sources of information that must be combined in an algorithmic way (as in the size constancy example); it plays a role in virtually every percept. A distinction is drawn between the sensation, that pure sensory signal that is not affected by experience or
  • 42. expectation, and the perception, which is an “idea” of the objects the scene must contain. Sensations themselves are difficult to experience because they do not serve any useful purpose, other than to provide evidence for perceptual conclusions (“In the ordinary affairs of life the sensations have no other importance for us. Subjective sensations are of interest chiefly for scientific investiga- tions only,” p. 27). One must make a deliberate effort to experience sensations themselves. J. S, Bruner (1957) proposed a theory in this tradition that also explicitly implicates perceptual hypotheses in vision; Bruner’s contribution was part of what came to be termed the “New Look” in perception. REFERENCES Bruner, J. S. (1957). On perceptual readiness. Psychological Review, 64, 123-152. Dennett, D. (1991). Consciousness explained. Boston: Little, Brown. Gregory, R. (1980). Perceptions as hypotheses. Philosophi- cal Transactions of the Royal Society of London B, 290, 181-197. Johansson (1973). Visual perception of biological motion and Introduction to Reading 1 m 23 For example, when observers viewed briefly-flashed playing cards that occasionally were colored incorrectly (e.g., a black ten of diamonds), they would sometimes report the color as purple, as iftheir knowledge of the correct color (a perceptual hypothesis of red) was somehow combined with the sensory evidence (black) to yield a sort of perceptual compromise. Other modern constructivists include Irvin Rock (e.g., 1983) and Richard Gregory (e.g., 1974). The constructivist position is rarely discussed in the contemporary literature, however. This is because it is now part of the set of commonly held foundational assumptions of almost all vision scientists. a model for its analysis. Perception & Psychophysics, 14, 201-211. Ramachandran, V. S. (1993) Filling the gaps in perception: II. Scotomas and phantom limbs. Current Directions in Psychological Science, 2, 56-65. Rock, I. (1983). The logic of perception. Cambridge, MA: MIT Press.
  • 43. PEA D EN G4 Concerning thePerceptions inGeneral H. von Helmholtz he sensations aroused by light in the nervous mechanism of vision enable us to form concep- tions as to the existence, form and position of ex- ternal objects. These ideas are called visual per- ceptions. In this third subdivision of Physiological Optics we must try to analyze the scientific re- sults which we have obtained concerning the con- ditions which give rise to visual perceptions. Perceptions of external objects being therefore of the nature of ideas, and ideas themselves being invariably activities of our psychic energy, percep- tions also can only be the result of psychic energy. Accordingly, strictly speaking, the theory of per- ceptions belongs properly in the domain of psy- chology. This is particularly true with respect to the mode of the mental activities in the case of the perceptions and with respect to the determination of their laws. Yet even here there is a wide field of investigation in both physics and physiology, in- asmuch as we have to determine, scientifically as far as possible, what special properties of the physi- cal stimulus and of the physiological stimulation are responsible for the formation of this or that particular idea as to the nature of the external ob- jects perceived. In this part of the subject, there- fore, we shall have to investigate the special prop- erties of the retinal images, muscular sensations, etc., that are concerned in the perception of a defi- nite position of the observed object, not only as to its direction but as to its distance; how the percep- tion of the form of a body of three dimensions depends on certain peculiarities of the images; and under what circumstances it will appear single or double as seen by both eyes, etc. Thus, our main 24 purpose will be simply to investigate the material of sensation whereby we are enabled to form ideas, in those relations that are important for the per- ceptions obtained from them. This problem can be solved entirely by scientific methods. At the same time, we cannot avoid referring to psychic activities and the laws that govern them, as far as they are concerned with the perception of the senses. But the discovery and description of these psychic activities will not be regarded as an es- sential part of our present task, because then we might run the risk of losing our hold of established facts and of not adhering steadily to a method founded on clear, well-recognized principals. Thus, for the present at least, I' think the psychological domain of the physiology of the senses should be kept separate from pure psychology, whose prov- ince really is to establish as far as possible the laws and nature of the processes of the mind. Still we cannot altogether avoid speaking of the mental processes that are active in the sense-per- ceptions, if we wish to see clearly the connection between the phenomena and to arrange the facts in their proper relation to one another. And hence, to prevent any misconception of the plan I have in mind, I intend to devote the latter part of this chap- ter to a discussion of the conclusions which Ithink can be inferred with respect to these mental pro- cesses. And yet we know by experience that people 'In this volume (contrary to the usage adopted in the two previous volumes of the English translation), the editor has deemed it best to retain the more intimate language of the original text, and let the author speak throughout in the first person. (J.P.C.S.)
  • 44. very seldom come to any agreement as to abstract questions of this nature. The keenest thinkers, phi- losophers like Kant for instance, have long ago analysed these relations correctly and demon- strated them, and yet there is no permanent and general agreement about them among educated people. And, therefore, in the subsequent chapters devoted specially to the theory of the visual per- ceptions, I shall endeavour to avoid all reference to opinions as to mental activity, as involving ques- tions that always have been, and perhaps always will be, subjects of debate between the various metaphysical schools; so as not to distract the reader’s attention from those facts about which an agreement may possibly be reached, by wrangling over abstract propositions that are not necessarily involved in the problem before us. Here I shall merely indicate at the outset cer- tain general characteristics of the mental processes that are active in the sense-perceptions, because they will be constantly encountered in connection with the various subjects to be considered. With- out some previous explanation of their general sig- nificance and wide range of activity, the reader might be apt in some special case to regard them as paradoxical and incredible. The general rule determining the ideas of vi- sion that are formed whenever an impression is made on the eye, with or without the aid of optical instruments, is that such objects are always imag- ined as being present in the field ofvision gs would have to be there in order to produce the same im- pression on the nervous mechanism, the eyes be- ing used under ordinary normal conditions. To employ an illustration which has been mentioned before, suppose that the eyeball is mechanically stimulated at the outer corner of the eye. Then we imagine that we see an appearance of light in front of us somewhere in the direction of the bridge of the nose. Under ordinary conditions of vision, when our eyes are stimulated by light coming from outside, if the region of the retina in the outer cor- ner of the eye is to be stimulated, the light actu- ally has to enter the eye from the direction of the bridge of the nose. Thus, in accordance with the above rule, in a case of this kind we substitute a luminous object at the place mentioned in the field of view, although as a matter of fact the mechani- cal stimulus does not act on the eye from in front of the field of view nor from the nasal side of the Concerning the Perceptions in General m 25 eye, but, on the contrary, is exerted on the outer surface of the eyeball and more from behind. The general validity of the above rule will be shown by many other instances that will appear in the following pages. In the statement of this rule mention is made of the ordinary conditions of vision, when the visual organ is stimulated by light from outside; this out- side light, coming from the opaque objects in its path that were the last to be encountered, and hav- ing reached the eye along rectilinear paths through an uninterrupted layer of air. This is what is meant here by the normal use of the organ of vision, and the justification for using this term is that this mode of stimulation occurs in such an enormous major- ity of cases that all other instances where the paths of the rays of light are altered by reflections or refractions, or in which the stimulations are not produced by external light, may be regarded as rare exceptions. This is because the retina in the fundus of the firm eyeball is almost completely protected from the actions of all other stimuli and is not easily accessible to anything but external light. When a person is in the habit of using an optical instrument and has become accustomed to it, for example, if he is used to wearing spectacles, to a certain extent he learns to interpret the visual images under these changed conditions. Incidentally, the rule given above corresponds to a general characteristic of all sense-perceptions, and not simply to the sense of sight alone. For example, the stimulation of the tactile nerves in the enormous majority of cases is the result of in- fluences that affect the terminal extensions of these nerves in the surface of the skin. It is only under exceptional circumstances that the nerve-stems can be stimulated by more powerful agencies. In ac- cordance with the above rule, therefore, all stimu- lations of cutaneous nerves, even when they af- fect the stem or the nerve-centre itself, are perceived as occurring in the corresponding pe- ripheral surface of the skin. The most remarkable and astonishing cases of illusions of this sort are those in which the peripheral area of this particu- lar portion of the skin is actually no longer in ex- istence, as, for example, in case of a person whose leg has been amputated. For a long time after the operation the patient frequently imagines he has vivid sensations in the foot that has been severed. He feels exactly the places that ache on one toe or
  • 45. 26 Visual Perception the other. Of course, in a case of this sort the stimu- lation can affect only what is left of the stem of the nerve whose fibres formerly terminated in the amputated toes. Usually, it is the end of the nerve in the scar that is stimulated by external pressure or by contraction of the scar tissue. Sometimes at night the sensations in the missing extremity get to be so vivid that the patient has to feel the place to be sure that his limb is actually gone. Thus it happens, that when the modes of stimu- lation of the organs of sense are unusual, incor- rect ideas of objects are apt to be formed; which used to be described, therefore, as illusions of the senses. Obviously, in these cases there is nothing wrong with the activity of the organ of sense and its corresponding nervous mechanism which pro- duces the illusion. Both of them have to act ac- cording to the laws that govern their activity once for all. It is rather simply an illusion in the judg- ment of the material presented to the senses, re- sulting in a false idea of it. The psychic activities that lead us to infer that there in front of us at a certain place there is a certain object of a certain character, are generally not conscious activities, but unconscious ones. In their result they are equivalent to a conclusion, to the extent that the observed action on our senses enables us to form an idea as to the possible cause of this action; although, as a matter of fact, it is invariably simply the nervous stimulations that are perceived directly, that is, the actions, but never the external objects themselves. But what seems to differentiate them from a conclusion, in the or- dinary sense of that word, is that a conclusion is an act of conscious thought. An astronomer, for example, comes to real conscious conclusions of this sort, when he computes the positions of the stars in space, their distances, etc., from the per- spective images he has had of them at various times and as they are seen from different parts of the orbit of the earth. His conclusions are based on a conscious knowledge of the laws of optics. In the ordinary acts of vision this knowledge of optics is lacking. Still it may be permissible to speak of the psychic acts of ordinary perception as unconscious conclusions, thereby making a distinction of some sort between them and the common so-called con- scious conclusions. And while it is true that there has been, and probably always will be, a measure of doubt as to the similarity of the psychic activity in the two cases, there can be no doubt as to the similarity between the results of such unconscious conclusions and those of conscious conclusions. These unconscious conclusions derived from sensation are equivalent in their consequences to the so-called conclusions to the so-called conclu- sions from analogy. Inasmuch as in an overwhelm- ing majority of cases, whenever the parts of the retina in the outer corner of the eye are stimulated, it has been found to be due to external light com- ing into the eye from the direction of the bridge of the nose, the inference we make is that it is so in every new case whenever this part of the retina is stimulated; just as we assert that every single in- dividual now living will die, because all previous experience has shown that all men who were for- merly alive have died. But, moreover, just because they are not free acts of conscious thought, these unconscious con- clusions from analogy are irresistible, and the ef- fect of them cannot be overcome by a better un- derstanding of the real relations. It may be ever so clear how we get an idea of a luminous phenom- enon in the field of vision when pressure is ex- erted on the eye; and yet we cannot get rid of the conviction that this appearance of light is actually there at the given place in the visual field; and we cannot seem to comprehend that there is a lumi- nous phenomenon at the place where the retina is stimulated. It is the same way in case of all the images that we see in optical instruments. On the other hand, there are numerous illustra- tions of fixed and inevitable associations of ideas due to frequent repetition, even when they have no natural connection, but are dependent merely on some conventional arrangement, as, for ex- ample, the connection between the written letters of a word and its sound and meaning. Still to many physiologists and psychologists the connection between the sensation and the conception of the object usually appears to be so rigid and obliga- tory that they are not much disposed to admit that, to a considerable extent at least, it depends on ac- quired experience, that is, on psychic activity. On the contrary, they have endeavoured to find some mechanical mode of origin for this connection through the agency of imaginary organic struc- tures. With regard to this question, all those expe- riences are of much significance which show how the judgment of the senses may be modified by experience and by training derived under various circumstances, and may be adapted to the new conditions. Thus, persons may learn in some mea- sure to utilize details of the sensation which oth-
  • 46. erwise would escape notice and not contribute to obtaining any idea of the object. On the other hand, too, this new habit may acquire such a hold that when the individual in question is back again in the old original normal state, he may be liable to illusions of the senses. Facts like these show the widespread influence that experience, training and habit have on our perceptions. But how far their influence really does extend, it would perhaps be impossible to say pre- cisely at present. Little enough is definitely known about infants and very young animals, and the in- terpretation of such observations as have been made on them is extremely doubtful. Besides, no one can say that infants are entirely without expe- rience and practice in tactile sensations and bodily movements. Accordingly, the rule givén above has been stated in a form which does not anticipate the decision of this question. It merely expresses what the result is. And so it can be accepted even by those who have entirely different opinions as to the way ideas originate concerning objects in the external world. Another general characteristic property of our sense-perceptions is, that we are not in the habit of observing our sensations accurately, except as they are useful in enabling us to recognize exter- nal objects. One the contrary, we are wont to dis- regard all those parts of the sensations that are of no importance so far as external objects are con- cerned. Thus in most cases some special agsistance and training are needed in order to observe these latter subjective sensations. It might seem that nothing could be easier than to be conscious of one’s own sensations; and yet experience shows that for the discovery of subjective sensations some special talent is needed, such as Purkinje mani- fested in the highest degree; or else it is the result of accident of theoretical speculation. For instance, the phenomena of the blind spot were discovered by Mariotte from theoretical considerations. Simi- larly, in the domain of hearing, I discovered the existence of those combination tones which I have called summation tones. In the great majority of cases, doubtless it was accident that revealed this or that subjective phenomenon to observers who happened to be particularly interested in such matters. It is only when subjective phenomena are so prominent as to interfere with the perception of things, that they attract everybody’s attention. Once the phenomena have been discovered, it is gener- ally easier for others to perceive them also, pro- Concerning the Perceptions in General m 27 vided the proper precautions are taken for observ- ing them, and the attention is concentrated on them. In many cases, however—for example, in the phe- nomena of the blind spot, or in the separation of the overtones and combination tones from the fun- damental tones of musical sounds, etc.—such an intense concentration of attention is required that, even with the help of convenient external appli- ances, many persons are unable to perform the experiments. Even the after-images of bright ob- jects are not perceived by most persons at first except under particularly favourable external con- ditions. It takes much more practice to see the fainter kinds of after-images. A common experi- ence, illustrative of this sort of thing, is for a per- son who has some ocular trouble that impairs his vision to become suddenly aware of the so-called mouches volantes in his visual field, although the causes of this phenomenon have been there in the vitreous humor all his life. Yet now he will be firmly persuaded that these corpuscles have de- veloped as the result of his ocular ailment, although the truth simply is that, owing to his ailment, the patient has been paying more attention to visual phenomena. No doubt, also, there are cases where one eye has gradually become blind, and yet the patient has continued to go about for an indefinite time without noticing it, until he happened one day to close the good eye without closing the other, and so noticed the blindness of that eye.’ When a person’s attention is directed for the first time to the double images in binocular vision, he is usually greatly astonished to think that he had never noticed them before, especially when he re- flects that the only objects he has ever seen single were those few that happened at the moment to be about as far from his eyes as the point of fixation. The great majority of objects, comprising all those that were farther or nearer than this point, were all seen double. Accordingly, the first thing we have to learn is to pay heed to our individual sensations. Ordinarily we do so merely in case of those sensations that enable us to find out about the world around us. In the ordinary affairs of life the sensations have no other importance for us. Subjective sensations are of interest chiefly for scientific investigations only. If they happen to be noticed in the ordinary activ- 2 Nearly everybody has a dominant eye, which governs the other eye; and in which the vision is superior to that in the other eye. But not many persons are aware of the fact. (J.P.C.S.)
  • 47. 28 m Visual Perception ity of the senses, they merely distract the atten- tion. Thus while we may attain an extraordinary degree of delicacy and precision in objective ob- servation, we not only fail to do so in subjective observations, but indeed we acquire the faculty in large measure of overlooking them and of form- ing our opinions of objects independently of them, even when they are so pronounced that they might easily be noticed. The most universal sign by which subjective visual phenomena can be identified appears to be by the way they accompany the movement of the eye over the field of view. Thus, the after-images, the mouches volantes, the blind spot, and the “lu- minous dust” of the dark field all participate in the motions of the eye, and coincide successively with the various stationary objects in the visual field. On the other hand, if the same phenomena recur again invariably at the same places in the visual field, they may be regarded as being objec- tive and as being connected with external bodies. This is the case with contrast phenomena produced by after-images. The same difficulty that we have in observing subjective sensations, that is, sensations aroused by internal causes, occurs also in trying to ana- lyze the compound sensations, invariably excited in the same connection by any simple object, and to resolve them into their separate components. In such cases experience shows us how to recognize a compound aggregate of sensations as being the sign of a simple object. Accustomed to consider the sensation-complex as a connected whole, gen- erally we are not able to perceive the separate parts of it without external help and support. Many il- lustrations of this kind will be seen in the follow- ing pages. For instance the perception of the ap- parent direction of an object from the eye depends on the combination of those sensations by which we estimate the adjustment of the eye, and on be- ing able to distinguish those parts of the retina where light falls from those parts where it does not fall. The perception of the solid form of an object of three dimensions is the result of the com- bination of two different perspective views in the two eyes. The gloss of a surface, which is appar- ently a simple effect, is due to differences of colouring or brightness in the images of it in the two eyes. These facts were ascertained by theory and may be verified by suitable experiments. But usually it is very difficult, if not impossible, to discover them by direct observation and analysis of the sensations alone. Even with sensations that are much more involved and always associated with frequently recurring complex objects, the oftener the same combination recurs, and the more used we have become to regarding the sensation as the normal sign of the real nature of the object, the more difficult it will be to analyze the sensa- tion by observation alone. By way of illustration, it is a familiar experience that the colours of a land- scape come out much more brilliantly and defi- nitely by looking at them with the head on one side or upside down than they do when the head is in the ordinary upright position. In the usual mode of observation all we try to do is to judge correctly the objects as such. We know that at a certain dis- tance green surfaces appear a little different in hue. We get in the habit of overlooking this difference, and learn to identify the altered green of distant meadows and trees with the corresponding colour of nearer objects. In the case of very distant ob- jects like distant ranges of mountains, little of the colour of the body is left to be seen, because it is mainly shrouded in the colour of the illuminated air. This vague blue-grey colour, bordered above by the clear blue of the sky or the red-yellow of the sunset glow, and below by the vivid green of meadows and forests, is very subject to variations by contrast. To us itis the vague and variable colour of distance. The difference in it may, perhaps, be more noticeable sometimes and with some illumi- nations than at other times. But we do not deter- mine its true nature, because it is not ascribed to any definite object. We are simply aware of its variable nature. But the instant we take an unusual position, and look at the landscape with the head under one arm, let us say, or between the legs, it all appears like a flat picture; partly on account of the strange position of the image in the eye, and partly because, as we shall see presently, the bin- ocular judgment of distance becomes less accu- rate. It may even happen that with the head upside down the clouds have the correct perspective, whereas the objects on the earth appear like a paint- ing on a vertical surface, as the clouds in the sky usually do. At the same time the colours lose their associations also with near or far objects, and con- front us now purely in their own peculiar differ- ences.° Then we have no difficulty in recognizing that the vague blue-grey of the far distance may * This explanation in given also by O. N. Rood, Silliman’s Journ., (2) xxxii. 1861. pp.184, 185.
  • 48. indeed be a fairly saturated violet, and that the green of the vegetation blends imperceptibly through blue-green and blue into this violet, etc. This whole difference seems to me to be due to the fact that the colours have ceased to be distinc- tive signs of objects for us, and are considered merely as being different sensations. Consequently, we take in better their peculiar distinctions with- out being distracted by other considerations. The connection between the sensations and ex- ternal objects may interfere very much with the perception of their simplest relations. A good il- lustration of this is the difficulty about perceiving the double images of binocular vision when they can be regarded as being images of one and the same external object. In the same way we may have similar experi- ences with other kinds of sensations. The sensa- tion of the timbre of a sound, as I have shown else- where,* consists of a series of sensations of its partial tones (fundamental and harmonics); but it is exceedingly difficult to analyze the compound sensation of the sound into these elementary com- ponents. The tactile sensation of wetness is com- posed of that of coldness and that of smoothness of surface. Consequently, on inadvertently touch- ing a cold piece of smooth metal, we often get the impression of having touched something wet. Many other illustrations of this sort might be ad- duced. They all indicate that we are exceedingly well trained in finding out by our sensations the objective nature of the objects around us, but that we are completely unskilled in observing the sen- sations per se and that the practice of associating them with things outside of us actually prevents us from being distinctly conscious of the pure sen- sations. This is true also not merely with respect to quali- tative differences of sensation, but it is likewise true with respect to the perception of space-rela- tions. For example, the spectacle of a person in the act of walking is a familiar sight. We think of this motion as a connected whole, possibly taking note of some of its most conspicuous singularities. But it requires minute attention and a special choice of the point of view to distinguish the up- ward and lateral movements of the body in a person’s gait. We have to pick out points or lines of reference in the background with which we can compare the position of his head. But look through an astronomical telescope at a crowd of people in motion far away. Their images are upside down, Concerning the Perceptions in General m 29 but what a curious jerking and swaying of the body is produced by those who are walking about! Then there is no trouble whatever in noticing the pecu- liar motions of the body and many other singularities of gait; and especially differences between individuals and the reasons for them, sim- ply because this is not the everyday sight to which we are accustomed. On the other hand, when the image is inverted in this way, it is not so easy to tell whether the gait is light or awkward, dignified or graceful, as it was when the image was erect. Consequently, it may often be rather hard to say how much of our apperceptions (Anschauungen) as derived by the sense of sight is due directly to sensation, and how much of them, on the other hand, is due to experience and training. The main point of controversy between various investiga- tors in this territory is connected also with this difficulty. Some are disposed to concede te the influence of experience as much scope as possible, and to derive from it especially all notion of space. This view may be called the empirical theory (empiristische Theorie). Others, of course, are obliged to admit the influence of experience in the case of certain classes of perceptions; still with respect to certain elementary apperceptions that occur uniformly in the case of all observers, they believe it is necessary to assume a system of in- nate apperceptions that are not based on experi- ence, especially with respect to space-relations. In contradistinction to the former view, this may per- haps be called the intuition theory (nativistische Theorie) of the sense-perceptions. In my opinion the following fundamental prin- ciples should be kept in mind in this discussion. Let us restrict the word idea (Vorstellung) to mean the image of visual objects as retained in the memory, without being accompanied by any present sense-impressions; and use the term ap- perception (Anschauung) to mean a perception (Wahrnehmung) when it is accompanied by the sense-impressions in question. The term immedi- ate perception (Perzeption) may then be employed to denote an apperception of this nature in which there is no element whatever that is not the result of direct sensations, that is, an apperception such 4 Helmholtz, Die Lehre von den Tonempfundungen. Braunschweig 1862. (See English translation by A. J. Ellis, entitled On the sensations of tone as a physiological basis for the theory of music. 3rd ed. London and New York, 1895.— JEP.G:S5)
  • 49. 30 @ Visual Perception as might be derived without any recollection of previous experience. Obviously, therefore, one and the same apperception may be accompanied by the corresponding sensations in very different mea- sure. Thus idea and immediate perception may be combined in the apperception in the most differ- ent proportions.° A person in a familiar room which is brightly lighted by the sun gets an apperception that is abun- dantly accompanied by very vivid sensations. In the same room in the evening twilight he will not be able to recognize any objects except the brighter ones, especially the windows. But whatever he does actually recognize will be so intermingled with his recollections of the furniture that he can still move about in the room with safety and lo- cate articles he is trying to find, even when they are only dimly visible. These images would be utterly insufficient to enable him to recognize the objects without some previous acquaintance with them. Finally, he may be in the same room in com- plete darkness, and still be able to find his way about in it without making mistakes, by virtue of the visual impressions formerly obtained. Thus, by continually reducing the material that appeals to the senses, the perceptual-image (Anschauungs- bild) can ultimately be traced back to the pure memory-image (Vorstellungsbild) and may gradu- ally pass into it. In proportion as there is less and less material appeal to the senses, a person’s move- ments will, of course, become more and more un- certain, and his apperception less and less accu- rate. Still there will be no peculiar abrupt transition, but sensation and memory will continually supple- ment each other, only in varying degrees. But even when we look around a room of this sort flooded with sunshine, a little reflection shows us that under these conditions too a large part of our perceptual-image may be due to factors of memory and experience. The fact that we are ac- customed to the perspective distortions of pictures of parallelopipeds and to the form of the shadows they cast has much to do with the estimation of the shape and dimensions of the room, as will be seen hereafter. Looking at the room with one eye shut, we think we see itjust as distinctly and defi- nitely as with both eyes. And yet we should get exactly the same view in case every point in the room were shifted arbitrarily to a different distance from the eye, provided they all remained on the same lines of sight. Thus in a case like this we are really consider- ing an extremely multiplex phenomenon of sense; but still we ascribe a perfectly definite explana- tion to it, and it is by no means easy to realize that the monocular image of such a familiar object nec- essarily means a much more meagre perception than would be obtained with both eyes. Thus too it is often hard to tell whether or not untrained observers inspecting stereoscopic views really notice the peculiar illusion produced by the instru- ment. We see, therefore, how in a case of this kind reminiscences of previous experiences act in con- junction with present sensations to produce a per- ceptual image (Anschauungsbild) which imposes itself on our faculty of perception with overwhelm- ing power, without our being conscious of how much of it is due to memory and how much to present perception. Still more remarkable is the influence of the comprehension of the sensations in certain cases, especially with dim illumination, in which a vi- sual impression may be misunderstood at first, by not knowing how to attribute the correct depth- dimensions; as when a distant light, for example, is taken for a near one, or vice versa. Suddenly it dawns on us what it is, and immediately, under the influence of the correct comprehension, the correct perceptual image also is developed in its full intensity. Then we are unable to revert to the previous imperfect apperception. This is very common especially with compli- cated stereoscopic drawings of forms of crystals and other objects which come out in perfect clear- ness of perception the moment we once succeed in getting the correct impression. Similar experiences have happened to every- body, proving that the elements in the sense-per- ceptions that are derived from experience are just as powerful as those that are derived from present sensations. All observers who have thoroughly investigated the theory of the sense-perceptions, even those who were disposed to allow experience as little scope as possible, have always admitted this. Hence, at all events it must be conceded that, “It is very difficult to find the precise English equivalents for these metaphysical terms, which will prove satisfactory to everybody. And it may not be quite possible to restrict the English word “idea,” for example, to the definition here given. It is doubtful whether the author himself is scrupulously careful throughout the remainder of this work to distinguish these shades of meaning always exactly. (J.P.C.S.)
  • 50. even in what appears to the adult as being direct apperception of the senses, possibly a number of single factors may be. involved which are really the product of experience; although at the time it is difficult to draw the line between them. Now in my opinion we are justified by our pre- vious experiences in stating that no indubitable present sensation can be abolished and overcome by an act of the intellect; and no matter how clearly we recognize that it has been produced in some anomalous way, still the illusion does not disap- pear by comprehending the process. The attention may be diverted from sensations, particularly if they are feeble and habitual; but in noting those relations in the external world, that are associated with these sensations, we are obliged to observe the sensations themselves. Thus we may be un- mindful of the temperature-sensation of our skin when it is not very keen, or of the contact-sensa- tions produced by our clothing, as long as we are occupied with entirely different matters. But just as soon as we stop to think whether it is warm or cold, we are not in the position to convert the feel- ing of warmth into that of coldness; maybe be- cause we know that it is due to strenuous exertion and not to the temperature of the surrounding air. In the same way the apparition of light when pres- sure is exerted on the eyeball cannot be made to vanish simply by comprehending better the nature of the-process, supposing the attention is directed to the field of vision and not, say, to the ear or the skin. ‘ On the other hand, it may also be that we are not in the position to isolate an impression of sen- sation, because it involves the composite sense- symbol of an external object. However, in this case the correct comprehension of the object shows that the sensation in question has been perceived and used by the consciousness. My conclusion is, that nothing in our sense-per- ceptions can be recognized as sensation which can be overcome in the perceptual image and converted into its opposite by factors that are demonstrably due to experience. Whatever, therefore, can be overcome by fac- tors of experience, we must consider as being it- self the product of experience and training. By observing this rule, we shall find that it is merely the qualities of the sensation that are to be consid- ered as real; pure sensation; the great majority of space-apperceptions, however, being the product of experience and training. Concerning the Perceptions in General m 31 Still it does not follow that apperceptions, which persist in spite of our better conscious insight and continue as illusions, might not be due to experi- ence and training. Our knowledge of the changes of colour produced in distant objects by the hazi- ness of the atmosphere, of perspective distortions, and of shadow is undoubtedly a matter of experi- ence. And yet in a good landscape picture we shall get the perfect visual impression of the distance and the solid form of the buildings in it, in spite of knowing that it is all depicted on canvas. Similarly, our knowledge of the composite sound of the vowels is certainly obtained from experience; and yet we get the auditory-impres- sion of the vowel sound by combining the indi- vidual tones of tuning forks (as I have demon- strated) and grasp the sound in its entirety, although in this instance we know that it is really compound. Here we still have to explain how experience counteracts experience, and how illusion can be produced by factors derived from experience, when it might seem as if experience could not teach any- thing except what was true. In this matter we must remember, as was intimated above, that the sensa- tions are interpreted just as they arise when they are stimulated in the normal way, and when the organ of sense is used normally. We are not simply passive to the impressions that are urged on us, but we observe, that is, we adjust our organs in those conditions that enable them to distinguish the impressions most accu- rately. Thus, in considering an involved object, we accommodate both eyes as well as we can, and turn them so as to focus steadily the precise point on which our attention is fixed, that is, so as to get an image of it in the fovea of each eye; and then we let our eyes traverse all the noteworthy points of the object one after another. If we are interested in the general shape of the object and are trying to get as good an idea as we can of its relative di- mensions, we assume a position such that, with- out having to turn the head, we can survey the whole surface, enabling us at the same time to view as symmetrically as possible those dimensions we wish to compare. Thus, in looking at an object, as, for example, a building with prominent horizon- tal and vertical lines, we like to stand opposite to it with the centres of rotation of the two eyes ina horizontal line. This position of the eyes can be controlled at any moment by separating the double images; which in the case mentioned here are in the same horizontal plane.
  • 51. 32 ww Visual Perception Unquestionably, our reason for choosing this definite mode of seeing is because in this way we can observe and compare most accurately; and, consequently, in this so-called normal use of the eyes we learn best how to compare our sensations with the reality. And so we obtain also the most correct and most accurate perceptions by this method. But if, from necessity or on purpose, we em- ploy a different mode of looking at objects, that is, if we view them merely indirectly or without focusing both eyes on them, or without surveying them all over, or if we hold the head in some un- usual position, then we shall not be able to have as accurate apperceptions as when the eyes are used in the normal fashion. Nor are we so well trained in interpreting what we see under such circum- stances as in the other case. Hence there is more scope for interpretation, although, as a rule, we are not clearly aware of this uncertainty in the ex- planation of our sense-perceptions. When we see an object in front of us, we are obliged to assign it to some definite place in space. We cannot think of it as having some dubious intermediate posi- tion between two different places in space. With- out any recollections coming to our aid, we are wont to interpret the phenomenon as it would have to be interpreted if we had received the same im- pression in the normal and most accurate mode of observation. Thus certain illusions enter into the perception, unless we concentrate our eyes on the objects under observation, or when the objects are in the peripheral part of the visual field, or if the head is held to one side, or if we do not focus the object with both eyes at once. Moreover, the agree- ment between the images on the two retinas is most constant and regular in looking at distant objects. The fact that the horizontal floor usually happens to be in the lower part of the visual field, appar- ently influences the comparison of the fields of the two eyes in a peculiar manner. Thus, our judg- ment as to the position of near objects is not en- tirely correct when we observe them with the look tilted decidedly up or down. The retinal images presented in this way are interpreted just as if they had been obtained by looking straight ahead. We run across many illustrations of this sort. Our train- ing in interpreting immediate perceptions is not equally good in all directions of the eyes, but sim- ply for those directions which enable us to have the most accurate and most consistent perceptions. We transfer the latter to all cases, as in the instances just cited. Now.it is quite possible that the similarity be- tween a visual impression of this kind and one of the possible impressions obtained by normal ob- servation may not be so overwhelming and strik- ing as to preclude many other comparisons and corresponding interpretations of that impression. In such cases the explanation of the impression varies. Without any change of the retinal images, the same observer may see in front of him various perceptual images in succession, in which case the variation is easy to recognize. Or else one observer may incline more toward one comparison and in- terpretation, and another toward another. This has been a source of much controversy in physiologi- cal optics, because each observer has been disposed to consider the apperception which he obtained by the most careful observation he could make as being the only valid one. But supposing that we have such confidence in the observers as to as- sume that their observations were careful and un- prejudiced, and that they knew how to make them, it would not be proper in such cases to adopt one of the conflicting interpretations of the visual phe- nomenon as being the only correct one. And yet that is what they are disposed to do who try to derive the origin of perceptual images mainly from innate factors. The truth rather is, that in a case of this sort various perceptual images may be devel- oped; and we should seek rather to discover what circumstances are responsible for the decision one way or the other. It is true we meet with a difficulty here that does not exist in the other parts of the natural sciences. In many instances we have simply the assertions of individual observers, without being in the posi- tion to verify them by our own observation. Many idiosyncrasies are manifested in this region, some of which are doubtless due to the structure of the eyes, others to the habitual way of using the eyes, and others still perhaps to previous impressions and apperceptions. Of course, nobody save the person who has peculiarities of this nature can observe their effects, and nobody else can give an opinion about them. On the other hand, observa- tion in this region is by no means so easy as might be supposed at first. Steady fixation of a point for a long time while observations are being made in indirect vision; controlling the attention; taking the mind away from the ordinary objective interpre-
  • 52. tation of sense-impression; estimation of differ- ence of colour and of difference of space in the visual field—all these things take much practice. And hence a number of facts in this region cannot be observed at all without having had previous long training in making observations in physiological optics. It cannot be done even by persons who are skilled in making other kinds of observations. Thus, with respect to many matters we have to depend on the observations of a very limited num- ber of individuals, and hence when the results found by somebody else are different, it is much harder in this subject than anywhere else to judge rightly whether secondary influences have not contributed in an observation of this sort. Accord- ingly, I must apprise the reader in advance that much of the material that is perhaps: new in the following chapters may possibly be due to indi- vidual peculiarities of my own eyes. Under such circumstances, there was no alternative for me except to observe as carefully as possible the facts as they appeared to my own eyes, and to try to ascertain their connection. Discrepancies that have been found by other observers have been noted. But how widespread this or the other mode of vi- sion may be, is something that has to be left to the future to determine. Incidentally, the more the visual impressions are unlike the normal ones, the greater will be the va- riety of interpretation as a rule. This is a natural consequence of the view which I hold, and is an essential characteristic of the activity of psychic influences. Heretofore practically nothing has been ascer- tained as to the nature of psychic processes. We have simply an array of facts. Therefore, it is not strange that no real explanation can be given of the origin of sense-perceptions. The empirical theory attempts to prove that at least no other forces are necessary for their origin beyond the known faculties of the mind, although these forces them- selves may remain entirely unexplained. Now gen- erally it is a useful rule in scientific investigation not to make any new hypothesis so long as known facts seem adequate for the explanation, and the necessity of new assumptions has not been dem- onstrated. That is why I have thought it incum- bent to prefer the empirical view essentially. Still less does the intuition theory attempt to give any explanation of the origin of our perceptual images; for it simply plunges right into the midst of the Concerning the Perceptions in General m 33 matter by assuming that certain perceptual images of space would be produced directly by an innate mechanism, provided certain nerve fibres were stimulated. The earlier forms of this theory im- plied some sort of self-observation of the retina; inasmuch as we were supposed to know by intu- ition about the form of this membrane and the positions of the separate nerve terminals in it. In its more recent development, especially as formu- lated by E. Hering, there is an hypothetical sub- jective visual space, wherein the sensations of the separate nerve fibres are supposed to be registered according to certain intuitive laws. Thus in this theory not only is Kant’s assertion adopted, that the general apperception of space is an original form of our imagination, but certain special ap- perceptions of space are assumed to be intuitive. The naturalistic view has been called also a spe- cial theory of identity, because in it the perfect fu- sion of the impressions on the corresponding places of the two retinas has to be postulated. On the other hand, the empirical theory is spoken of as a theory of projection,® because according to it the perceptual images of objects are projected in space by means of psychic processes. I should like to avoid this term, because both supporters and opponents of this view have often attached undue importance to the idea that this projection must take place parallel to the lines of direction; which was certainly not the correct description of the psychic process. And, even if this construction were admitted as being valid simply with respect to the physiological description of the process, the idea would be incorrect in very many instances. I am aware that in the present state of knowl- edge it is impossible to refute the intuition theory. The reasons why I prefer the opposite view are because in my opinion: 1. The intuition theory is an unnecessary hypoth- esis. 2. Its consequences thus far invariably apply to perceptual images of space which only in the fewest cases are in accordance with reality and with the correct visual images that are undoubt- edly present; as will be shown in detail later. The adherents of this theory are, therefore, obliged to make the very questionable assump- tion, that the space sensations, which accord- See remarks in Appendix I as to misunderstandings connected with the term “projection theory.’—K.
  • 53. 34 om Visual Perception ing to them are present originally, are continu- ally being improved and overruled by knowl- edge which we have accumulated by experi- ence. By analogy with all other experiences, however, we should have to expect that the sen- sations which have been overruled continued to be present in the apperception as a conscious illusion, if nothing else. But this is not the case. 3. It is not clear how the assumption of these origi- nal “space sensations” can help the explana- tion of our visual perceptions, when the adher- ents of this theory ultimately have to assume in by far the great majority of cases that these sen- sations must be overruled by the better under- standing which we get by experience. In that case it would seem to me much easier and sim- pler to grasp, that all apperceptions of space were obtained simply by experience, instead of supposing that the latter have to contend against intuitive perceptual images that are generally false. This is by way of justifying my point of view. A choice had to be made simply for the sake of get- ting at least some sort of superficial order amid the chaos of phenomena; and so I believed I had to adopt the view I have chosen. However, I trust it has not affected the correct observation and description of the facts. To prevent misunderstandings as to my mean- ing, and to make it clearer to the natural intelli- gence of those readers who have never thought much about their sense-perceptions, the follow- ing explanations will be added. Thus far the sensations have been described as being simply symbols for the relations in the ex- ternal world. They have been denied every kind of similarity or equivalence to the things they denote. Here we touch on the much disputed point as to how far our ideas agree in the main with their ob- jects; that is, whether they are true or false, as one might say. Some have asserted that there is such an agreement, and others have denied it. In favour of it, a pre-established harmony between nature and mind was assumed. Or it was maintained that there was an identity of nature and mind, by re- garding nature as the product of the activity of a general mind; the human mind being supposed to be an emanation from it. The intuition theory of Space-apperceptions is connected with these views to the fact that, by some innate mechanism and a certain pre-established harmony, it admits of the origin of perceptual images that are supposed to correspond with reality, although in a rather im- perfect.fashion. Or else the agreement between ideas and their objects was denied, the ideas being explained therefore as illusions. Consequently, it was neces- sary to deny also the possibility of all knowledge of any objects whatsoever. This was the attitude of certain so-called “sensational” philosophers in England in the eighteenth century. However, it is not my purpose here to undertake an analysis of the opinions of the various philosophical schools on this question. That would be much too exten- sive a task in this place. I shall confine myself therefore merely to inquiring what I think should be the attitude of an investigator toward these con- troversies. Our apperceptions and ideas are effects wrought on our nervous system and our consciousness by the objects that are thus apprehended and con- ceived. Each effect, as to its nature, quite neces- sarily depends both on the nature of what causes the effect and on that of the person on whom the effect is produced. To expect to obtain an idea which would reproduce the nature of the thing conceived, that is, which would be true in an ab- solute sense, would mean to expect an effect which would be perfectly independent of the nature of the thing on which the effect was produced; which would be an obvious absurdity. Our human ideas, therefore, and all ideas of any conceivable intelli- gent creature, must be images of objects whose mode is essentially codependent on the nature of the consciousness which has the idea, and is con- ditioned also by its idiosyncrasies. In my opinion, therefore, there can be no pos- sible sense in speaking of any other truth of our ideas except of a practical truth. Our ideas of things cannot be anything but symbols, natural signs for things which we learn how to use in order to regu- late our movements and actions. Having learned correctly how to read those symbols, we are en- abled by their help to adjust our actions so as to bring about the desired result; that is, so that the expected new sensations will arise. Not only is there in reality no other comparison at all between ideas and things—all the schools are agreed about this—but any other mode of comparison is entirely unthinkable and has no sense whatever. This latter consideration is the conclusive thing, and must be grasped in order to escape from the labyrinth of conflicting opinions. To ask whether the idea Ihave ~
  • 54. of a table, its form, strength, colour, weight, etc., is true per se, apart from any practical use I can make of this idea, and whether it corresponds with the real thing, or is false and due to an illusion, has just as much sense as to ask whether a certain musical note is red, yellow, or blue. Idea and the thing conceived evidently belong to two entirely different worlds, which no more admit of being compared with each other than colours and musi- cal tones or than the letters of a book and the sound of the word they denote. Were there any sort of similarity of correspon- dence between the idea in the head of a person A and the thing to which the idea belongs, another intelligent person B, conceiving both the thing it- self and A’s idea of it, according to the same laws, might be able to find some similarity between them or at least to suppose so; because the same sort of thing represented (conceived) in the same way would have to give the same kinds of images (ideas). Now I ask, what similarity can be imag- ined between the process in the brain that is con- comitant with the idea of a table and the table it- self? Is the form of the table to be supposed to be outlined by electric currents? And when the per- son with the idea has the idea that he is walking around the table, must the person then be outlined by electric currents? Perspective projections of the external world in the hemispheres of the brain (as they are supposed to be) are evidently not suffi- cient for representing the idea of a bodily object. And granted that a keen imagination is not fright- ened away by these and similar hypotheses, such an electrical reproduction of the table in the brain would be simply another bodily object to be per- ceived, but no idea of the table. However, it is not simply persons with materialistic opinions who try to refute the proposed statement, but also persons with idealistic views. And for the latter I should think the argument would be still more forcible. What possible similarity can there be between the idea, some modification of the incorporeal mind that has no extension in space, and the body of the table that occupies space? As far as I am aware, the idealistic philosophers have never once inves- tigated even a single hypothesis or imagination in order to show this connection. And by the very nature of this view it is something that cannot be investigated at all. In the next place as to the properties of objects in the external world, alittle reflection reveals that all properties attributable to them may be said to Concerning the Perceptions in General m 35 be simply effects exerted by them either on our senses or on other natural objects. Colour, sound, taste, smell, temperature, smoothness, and firm- ness are properties of the first sort, and denote ef- fects on our organs of sense. Smoothness and firm- ness denote the degree of resistance either to the gliding contact or pressure of the hand. But other natural bodies may be employed instead of the hand. And the same thing is true in testing other mechanical properties such as elasticity and weight. Chemical properties are described by cer- tain reactions, that is, by effects exerted by one natural body on others. It is the same way with any other physical property of a body, optical, elec- trical, or magnetic. In every case we have to do with the mutual relations between various bodies and with the effects depending on the forces that different bodies exert on each other. For all natu- ral forces are such as are exerted by one body on others. When we try to think of mere matter with- out force, it is void of properties likewise, except as to its different distribution in space and as to its motion. All properties of bodies in nature are mani- fested therefore simply by being so situated as to interact with other bodies of nature or with our organs of sense. But as such interaction may oc- cur at any time, particularly too as it may be pro- duced by us voluntarily at any moment, and as then we see invariably the peculiar sort of interaction occurring, we attribute to the objects a permanent capacity for such effects which is always ready to become effective. This permanent capacity is a so- called characteristic property. The result is that in point of fact the character- istic properties of natural objects, in spite of this name, do not denote something that is peculiar to the individual object by itself, but invariably im- ply some relation to a second object (including our organs of sense). The kind of effect must, of course, depend always on the peculiarities both of the body producing it and of the body on which it is produced. As to this there is never any doubt even for an instant, provided we have in mind those properties of bodies that are manifested when two bodies belonging to the external world react on each other, as in the case of chemical reactions. But in the case of properties depending on the mutual relations between things and our organs of sense, people have always been disposed to forget that here too we are concerned with the reaction toward a special reagent, namely, our own nervous system; and that colour, smell, and taste, and feel-
  • 55. Another Random Document on Scribd Without Any Related Topics
  • 56. favourite tricks being that of making feint to assist a lady to descend and take her bag from her hand. “I don’t suppose,” she sighed, “that I shall ever see or hear of my ornaments again. Yet I think that if the thief but knew the truth concerning me he would regret. Perhaps he is without means, just as I am. Probably he became a thief of sheer necessity, as I have heard many men have become. Criminal instinct is not always responsible for an evil life. Many persons try to live honestly, but fate is ever contrary. Indeed, is it not so with my own self?” She turned, and her eyes fell upon the sleeping child. She was all she had now to care for in the whole wide world. Recollections of her last visit to Paris haunted her—that visit when Carl had so very indiscreetly followed her there, and taken her about incognito in open cabs to see the sights. There had been no harm in it whatsoever, no more harm than if he had been her equerry, yet her enemies had, alas! hurled against her their bitter denunciations, and whispered their lies so glibly that they were believed as truth. Major Scheel, the attaché at the Embassy, had recognised them, and being Leitolf’s enemy, had spread the report. It had been a foolish caprice of hers to take train from Aix-les- Bains to Paris to see her old French nurse Marie, who had been almost as a mother to her. The poor old woman, a pensioned servant of the Archducal family, had, unfortunately, died a month ago, otherwise she would have had a faithful, good friend in Paris. Marie, who knew Count Leitolf well, could have refuted their allegations had she lived; but an attack of pneumonia had proved fatal, and she had been buried with a beautiful wreath bearing the simple words “From Claire” upon her coffin.
  • 57. As the sunset haze fell over Paris she still sat beside the sleeping child. If her enemies condemned her, then she would not defend herself. God, in whom she placed her fervent trust, should judge her. She had no fear of man’s prejudices or misjudgment. She placed her faith entirely in her Maker. To His will she bowed, for in His sight the pauper and the princess are equal. That evening she had a little soup sent to her room, and when Ignatia was again sleeping soundly she went forth upon the balcony leading from the corridor, and sitting there, amused herself by looking down upon the life and movement of the great salon below. To leave the hotel was impossible because of Ignatia, and she now began to regret that she had not brought the maid with her from Wartenstein. Time after time the misfortune of the loss of her jewels recurred to her. It had destroyed her independence, and it had negatived all her plans. Money was necessary, even though she were an Imperial Archduchess. She was incognito, and therefore had no credit. The gay, after-dinner scene of the hotel was presented below—the flirtations, the heated conversations, and the lazy, studied attitudes of the bloused English girl, who lolls about in cane lounge-chairs after dining, and discusses plays and literature. From her chair on the balcony above she looked down upon that strange, changeful world—the world of tourist Paris. Born and bred at Court as she had been, it was a new sensation to her to have her freedom. The life was entirely fresh to her, and would have been pleasant if there were not behind it all that tragedy of her marriage.
  • 58. Several days went by, and in order to kill time she took little Ignatia daily in a cab and drove in the Bois and around the boulevards, revisiting all the “sights” which Leitolf had shown her. Each morning she went out driving till the luncheon-hour, and having once lunched with old Marie upstairs at the Brasserie Universelle in the Avenue de l’Opéra, she went there daily. You probably know the place. Downstairs it is an ordinary brasserie with a few chairs out upon the pavement, but above is a smart restaurant peculiarly Parisian, where the hors d’oeuvres are the finest in Eurorie and the vin gris a speciality. The windows whereat one sits overlook the Avenue, and from eleven o’clock till three it is crowded. She went there for two reasons—because it was small, and because the life amused her. Little Ignatia would sit at her side, and the pair generally attracted the admiration of every one on account of their remarkably good looks. The habitués began inquiring of the waiters as to who was the beautiful lady in black, but the men only elevated their shoulders and exhibited their palms. “A German,” was all they could answer. “A great lady evidently.” That she attracted attention everywhere she was quite well aware, yet she was not in the least annoyed. As a royalty she was used to being gazed upon. Only when men smiled at her, as they did sometimes, she met them with a haughty stare. The superiority of her Imperial blood would on such occasions assert itself, much to the confusion of would-be gallants. Thus passed those spring days with Paris at her gayest and best. The woman who had renounced a crown lived amid all that bright life, lonely, silent, and unrecognised, her one
  • 59. anxiety being for the future of her little one, who was ever asking when Allen would return.
  • 60. Chapter Fifteen. The Shy Englishman. One afternoon about four o’clock, as the Princess, leading little Ignatia, who was daintily dressed in white, was crossing the great hall of the Hotel Terminus on her way out to drive in the Bois, a rather slim, dark-haired man, a little under forty, well-dressed in a blue-serge suit, by which it required no second glance to tell that he was an Englishman, rose shyly from a chair and bowed deeply before her. At that hour there were only two or three elderly persons in the great hall, all absorbed in newspapers. She glanced at the stranger quickly and drew back. At first she did not recognise him, but an instant later his features became somehow familiar, although she was puzzled to know where she had met him before. Where he had bowed to her was at a safe distance from the few other people in the hall; therefore, noticing her hesitation, the man exclaimed in English with a smile,— “I fear that your Imperial Highness does not recollect me, and I trust that by paying my respects I am not intruding. May I be permitted to introduce myself? My name is Bourne. We met once in Treysa. Do you not recollect?” In an instant the truth recurred to her, and she stood before him open-mouthed. “Why, of course!” she exclaimed. “Am I ever likely to forget? And yet I saw so little of your face on that occasion that I
  • 61. failed now to recognise you! I am most delighted to meet you again, Mr Bourne, and to thank you.” “Thanks are quite unnecessary, Princess,” he declared; whereupon in a low voice she explained that she was there incognito, under the title of the Baroness Deitel, and urged him not to refer to her true station lest some might overhear. “I know quite well that you are here incognito,” he said. “And this is little Ignatia, is it?” and he patted the child’s cheeks. Then he added, “Do you know I have had a very great difficulty in finding you. I have searched everywhere, and was only successful this morning, when I saw you driving in the Rue Rivoli and followed you here.” Was this man a secret agent from Treysa, she wondered. In any case, what did he want with her? She treated him with courtesy, but was at the same time suspicious of his motive. At heart she was annoyed that she had been recognised. And yet was she not very deeply indebted to him? “Well, Mr Bourne,” said the Princess, drawing herself up, and taking the child’s hand again to go out, “I am very pleased to embrace this opportunity of thanking you for the great service you rendered me. You must, however, pardon my failure to recognise you.” “It was only natural,” the man exclaimed quickly. “It is I who have to apologise, your Highness,” he whispered. “I have sought you because I have something of urgent importance to tell you. I beg of you to grant me an interview somewhere, where we are not seen and where we cannot be overheard.” She looked at him in surprise. The Englishman’s request was a strange one, yet from his manner she saw that he
  • 62. was in earnest. Why, she wondered, did he fear being seen with her? “Cannot you speak here?” she inquired. “Not in this room, among these people. Are there not any smaller salons upstairs? they would be empty at this hour. If I recollect aright, there is a small writing-room at the top of the stairs yonder. I would beg of your Highness to allow me to speak to you there.” “But what is this secret you have to tell me?” she inquired curiously. “It surely cannot be of such a nature that you may not explain it in an undertone here?” “I must not be seen with you, Princess,” he exclaimed quickly. “I run great risk in speaking with you here in public. I will explain all if you will only allow me to accompany you to that room.” She hesitated. So ingenious had been the plots formed against her that she had now grown suspicious of every one. Yet this man was after all a mystery, and mystery always attracted her, as it always attracts both women and men equally. So with some reluctance she turned upon her heel and ascended the stairs, he following her at a respectful distance. Their previous meeting had indeed been a strange one. Fond of horses from her girlhood, she had in Treysa made a point of driving daily in her high English dogcart, sometimes a single cob, and sometimes tandem. She was an excellent whip, one of the best in all Germany, and had even driven her husband’s coach on many occasions. On the summer’s
  • 63. afternoon in question, however, she was driving a cob in one of the main thoroughfares of Treysa, when of a sudden a motor car had darted past, and the animal, taking fright, had rushed away into the line of smart carriages approaching on the opposite side of the road. She saw her peril, but was helpless. The groom sprang out, but so hurriedly that he fell upon his head, severely injuring himself; while at that moment, when within an ace of disaster, a man in a grey flannel suit sprang out from nowhere and seized the bridle, without, however, at once stopping the horse, which reared, and turning, pinned the stranger against a tree with the end of one of the shafts. In an instant a dozen men, recognising who was driving, were upon the animal, and held it; but the next moment she saw that the man who had saved her had fallen terribly injured, the shaft having penetrated his chest, and he was lying unconscious. Descending, she gazed upon the white face, from the mouth of which blood was oozing; and having given directions for his immediate conveyance to the hospital and for report to be made to her as soon as possible, she returned to the palace in a cab, and telephoned herself to the Court surgeon, commanding him to do all in his power to aid the sufferer. Next day she asked permission of the surgeon that she might see the patient, to thank him and express her sympathy. But over the telephone came back the reply that the patient was not yet fit to see any one, and, moreover, had expressed a desire that nobody should come near him until he had quite recovered.
  • 64. In the fortnight that went by she inquired after him time after time, but all that she was able to gather was that his name was Guy Bourne, and that he was an English banker’s clerk from London, spending his summer holiday in Treysa. She sent him beautiful flowers from the royal hot-houses, and in reply received his thanks for her anxious inquiries. He told the doctor that he hoped the Princess would not visit him until he had quite recovered. And this wish of his she had of course respected. His gallant action had, without a doubt, saved her from a very serious accident, or she might even have lost her life. Gradually he recovered from his injuries, which were so severe that for several days his life was despaired of, and then when convalescent a curious thing happened. He one day got up, and without a word of thanks or farewell to doctors, staff, or to the Crown Princess herself, he went out, and from that moment all trace had been lost of him. Her Highness, when she heard of this, was amazed. It seemed to her as though for some unexplained reason he had no wish to receive her thanks; or else he was intent on concealing his real identity with some mysterious motive or other. She had given orders for inquiry to be made as to who the gallant Englishman was; but although the secret agents of the Government had made inquiry in London, their efforts had been futile. It happened over two years ago. The accident had slipped from her memory, though more than once she had wondered who might be the man who had risked his life to save hers, and had then escaped from Treysa rather than be presented to her.
  • 65. And now at the moment when she was in sore need of a friend he had suddenly recognised her, and come forward to reveal himself! Naturally she had not recognised in the dark, rather handsome face of the well-dressed Englishman the white, bloodless countenance of the insensible man with a brass- tipped cart-shaft through his chest. And he wanted to speak to her in secret? What had he, a perfect stranger, to tell her? The small writing-room at the top of the stairs was fortunately empty, and a moment later he followed her into it, and closed the door. Little Ignatia looked with big, wondering eyes at the stranger. The Princess seated herself in a chair, and invited the Englishman to take one. “Princess,” he said in a refined voice, “I desire most humbly to apologise for making myself known to you, but it is unfortunately necessary.” “Unfortunately?” she echoed. “Why unfortunately, Mr Bourne, when you risked your life for mine? At that moment you only saw a woman in grave peril; you were not aware of my station.” “That is perfectly true,” he said quietly. “When they told me at the hospital who you were, and when you sent me those lovely flowers and fruit, I was filled with—well, with shame.” “Why with shame?” she asked. “You surely had no need to be ashamed of your action? On the contrary, the King’s intention was to decorate you on account of your brave action, and had already given orders for a letter to be sent to your own King in London, asking his Majesty to allow you
  • 66. as a British subject to receive and wear the insignia of the Order of the Crown and Sword.” “And I escaped from Treysa just in time,” he laughed. Then he added, “To tell you the truth, Princess, it is very fortunate that I left before—well, before you could see me, and before his Majesty could confer the decoration.” “But why?” she asked. “I must confess that your action in escaping as you did entirely mystified me.” “You were annoyed that I was ungentlemanly enough to run away without thanking your Highness for all your solicitude on my behalf, and for sending the surgeon of the royal household to attend to my injuries. But, believe me, I am most deeply and sincerely grateful. It was not ingratitude which caused me to leave Treysa in secret as I did, but my flight was necessary.” “Necessary? I don’t understand you.” “Well, I had a motive in leaving without telling any one.” “Ah, a private motive!” she said—“something concerning your own private affairs, I suppose?” He nodded in the affirmative. How could he tell her the truth? His disinclination to explain the reason puzzled her sorely. That he was a gallant man who had saved a woman without thought of praise or of reward was proved beyond doubt, yet there was something curiously mysterious about him which attracted her. Other men would have at least been proud to receive the thanks and decoration of a reigning sovereign, while he had utterly ignored them. Was he an anarchist?
  • 67. “Princess,” he said at last, rising from his chair and flushing slightly, “the reason I have sought you to-day is not because of the past, but is on account of the present.” “The present! why?” “I—I hardly know what to say, Princess,” he said confusedly. “Two years ago I fled from you because you should not know the truth—because I was in fear. And now Fate brings me again in your path in a manner which condemns me.” “Mr Bourne, why don’t you speak more plainly? These enigmas I really cannot understand. You saved my life, or at least saved me from a very serious accident, and yet you escaped before I could thank you personally. To-day you have met me, and you tell me that you escaped because you feared to meet me.” “It is the truth, your Highness. I feared to meet you,” he said, “and, believe me, I should not have sought you to-day were it not of most urgent necessity.” “But why did you fear to meet me?” “I did not wish you to discover what I really am,” he said, his face flushing with shame. “Are you so very timid?” she asked with a light laugh. But in an instant she grew serious. She saw that she had approached some sore subject, and regretted. The Englishman was a strange person, to say the least, she thought. “I have nothing to say in self-defence, Princess,” he said very simply. “The trammels of our narrow world are so hypocritical, our laws so farcical and full of incongruities,
  • 68. and our civilisation so fraught with the snortings of Mother Grundy, that I can only tell you the truth and offer no defence. I know from the newspapers of your present perilous position, and of what is said against you. If you will permit me to say so, you have all my sympathy.” And he paused and looked straight into her face, while little Ignatia gazed at him in wonder. “I wonder if your Highness will forgive me if I tell you the truth?” he went on, as though speaking to himself. “Forgive you? Why, of course,” she laughed. “What is there to forgive?” “Very much, Princess,” he said gravely. “I—I’m ashamed to stand here before you and confess; yet I beg of you to forgive me, and to accept my declaration that the fault is not entirely my own.” “The fault of what?” she inquired, not understanding him. “I will speak plainly, because I know that your good nature and your self-avowed indebtedness to me—little as that indebtedness is—will not allow you to betray me,” he said in a low, earnest tone. “You will recollect that on your Highness’s arrival at the Gare de l’Est your dressing-bag was stolen, and within it were your jewels—your most precious possession at this critical moment of your life?” “Yes,” she said in a hard voice of surprise, her brows contracting, for she was not yet satisfied as to the stranger’s bona fides. “My bag was stolen.” “Princess,” he continued, “let me, in all humility, speak the truth. The reason of my escape from Treysa was because your police held a photograph of me, and I feared that I might be identified. I am a thief—one of an international
  • 69. gang. And—and I pray you to forgive me, and to preserve my secret,” he faltered, his cheeks again colouring. “Your jewels are intact, and in my possession. You can now realise quite plainly why—why I escaped from Treysa!” She held her breath, staring at him utterly stupefied. This man who had saved her, and so nearly lost his own life in the attempt, was a thief!
  • 70. Chapter Sixteen. Light Fingers. Her Highness was face to face with one of those clever international criminals whose coups were so constantly being reported in the Continental press. She looked straight into his countenance, a long, intense look, half of reproach, half of surprise, and then, in a firm voice, said,— “Mr Bourne, I owe you a very great debt. To-day I will endeavour to repay it. Your secret, and the secret of the theft, shall remain mine.” “And you will give no information to the police?” he exclaimed quickly—“you promise that?” “I promise,” she said. “I admire you for your frankness. But, tell me—it was not you who took my bag at the station?” “No. But it was one of us,” he explained. “When the bag containing the jewels was opened I found, very fortunately, several letters addressed to you—letters which you evidently brought with you from Treysa. Then I knew that the jewels were yours, and determined, if I could find you, to restore them to you with our apologies.” “Why?” she asked. “You surely do not get possession of jewels of that value every day?” “No, Princess. But the reason is, that although my companions are thieves, they are not entirely devoid of the respect due to a woman. They have read in the newspapers
  • 71. of your domestic unhappiness, and of your flight with the little Princess, and have decided that to rob a defenceless woman, as you are at this moment, is a cowardly act. Though we are thieves, we still have left some vestige of chivalry.” “And your intention is really to restore them to me?” she remarked, much puzzled at this unexpected turn of fortune. “Yes, had I not found those letters among them, I quite admit that, by this time, the stones would have been in Amsterdam and re-cut out of all recognition,” he said, rather shamefacedly. Then, taking from his pocket the three letters addressed to her—letters which she had carried away from Treysa with her as souvenirs—he handed them to her, saying,— “I beg of you to accept these back again. They are better in your Imperial Highness’s hands than my own.” Her countenance went a trifle pale as she took them, and a sudden serious thought flashed through her mind. “Your companions have, I presume, read what is contained in these?” “No, Princess; they have not. I read them, and seeing to whom they were addressed, at once took possession of them. I only showed my companions the addresses.” She breathed more freely. “Then, Mr Bourne, I am still more deeply in your debt,” she declared; “you realised that those letters contained a woman’s secret, and you withheld it from the others. How can I sufficiently thank you?”
  • 72. “By forgiving me,” he said. “Remember, I am a thief, and if you wished you could call the hotel manager and have me arrested.” “I could hardly treat in that way a man who has acted so nobly and gallantly as you have,” she remarked, with perfect frankness. “If those letters had fallen into other hands they might, have found their way back to the Court, and to the King.” “I understand perfectly,” he said, in a low voice. “I saw by the dates, and gathered from the tenor in which they were written that they concealed some hidden romance. To expose what was written there would have surely been a most cowardly act—meaner even than stealing a helpless, ill-judged woman’s jewels. No, Princess,” he went on; “I beg that although I stand before you a thief, to whom the inside of a gaol is no new experience, a man who lives by his wits and his agility and ingenuity in committing theft, you will not entirely condemn me. I still, I hope, retain a sense of honour.” “You speak like a gentleman,” she said. “Who were your parents?” “My father, Princess, was a landed proprietor in Norfolk. After college I went to Sandhurst, and then entered the British Army; but gambling proved my ruin, and I was dismissed in disgrace for the forgery of a bill in the name of a brother officer. As a consequence, my father left me nothing, as I was a second son; and for years I drifted about England, an actor in a small travelling company; but gradually I fell lower and lower, until one day in London I met a well-known card-sharper, who took me as his partner, and together we lived well in the elegant rooms to which we inveigled men and there cheated them. The inevitable came
  • 73. at last—arrest and imprisonment. I got three years, and after serving it, came abroad and joined Roddy Redmayne’s gang, with whom I am at present connected.” The career of the man before her was certainly a strangely adventurous one. He had not told her one tithe of the remarkable romance of his life. He had been a gentleman, and though now a jewel thief, he still adhered to the traditions of his family whenever a woman was concerned. He was acute, ingenious beyond degree, and a man of endless resource, yet he scorned to rob a woman who was poor. The Princess Claire, a quick reader of character, saw in him a man who was a criminal, not by choice, but by force of circumstance. He was now still suffering from that false step he had taken in imitating his brother officer’s signature and raising money upon the bill. However she might view his actions, the truth remained that he had saved her from a terrible accident. “Yours has been an unfortunate career, Mr Bourne,” she remarked. “Can you not abandon this very perilous profession of yours? Is there no way by which you can leave your companions and lead an honest life?” When she spoke she made others feel how completely the purely natural and the purely ideal can blend into each other, yet she was a woman breathing thoughtful breath, walking in all her natural loveliness with a heart as frail- strung, as passion-touched, as ever fluttered in a female bosom. “Ah, Princess!” he cried earnestly, “I beg of you not to reproach me; willingly I would leave it all. I would welcome work and an honest life; but, alas! nowadays it is too late.
  • 74. Besides, who would take me in any position of trust, with my black record behind me? Nobody.” And he shook his head. “In books one reads of reformed thieves, but there are none in real life. A thief, when once a thief, must remain so till the end of his days—of liberty.” “But is it not a great sacrifice to your companions to give up my jewellery?” she asked in a soft, very kindly voice. “They, of course, recognise its great value?” “Yes,” he smiled. “Roddy, our chief, is a good judge of stones—as good, probably, as the experts at Spink’s or Streeter’s. One has to be able to tell good stuff from rubbish when one deals in diamonds, as we do. Such a quantity of fake is worn now, and, as you may imagine, we don’t care to risk stealing paste.” “But how cleverly my bag was taken!” she said. “Who took it? He was an elderly man.” “Roddy Redmayne,” was Bourne’s reply. “The man who, if your Highness will consent to meet him, will hand it back to you intact.” “You knew, I suppose, that it contained jewels?” “We knew that it contained something of value. Roddy was advised of it by telegraph from Lucerne.” “From Lucerne? Then one of your companions was there?” “Yes, at your hotel. An attempt was made to get it while you were on the platform awaiting the train for Paris, but you kept too close a watch. Therefore, Roddy received a telegram to meet you upon your arrival in Paris, and he met you.”
  • 75. What he told her surprised her. She had been quite ignorant of any thief making an attempt to steal the bag at Lucerne, and she now saw how cleverly she had been watched and met. “And when am I to meet Mr Redmayne?” she asked. “At any place and hour your Imperial Highness will appoint,” was his reply. “But, of course, I need not add that you will first give your pledge of absolute secrecy—that you will say nothing to the police of the way your jewels have been returned to you.” “I have already given my promise. Mr Redmayne may rely upon my silence. Where shall we fix the meeting? Here?” “No, no,” he laughed—“not in the hotel. There is an agent of police always about the hall. Indeed, I run great risk of being recognised, for I fear that the fact of your having reported your loss to the police at the station has set Monsieur Hamard and his friends to watch for us. You see, they unfortunately possess our photographs. No. It must be outside—say at some small, quiet café at ten o’clock to- night, if it will not disturb your Highness too much.” “Disturb me?” she laughed. “I ought to be only too thankful to you both for restoring my jewels to me.” “And we, on our part, are heartily ashamed of having stolen them from you. Well, let us say at the Café Vachette, a little place on the left-hand side of the Rue de Seine. You cross the Pont des Arts, and find it immediately; or better, take a cab. Remember, the Vachette, in the Rue de Seine, at ten o’clock. You will find us both sitting at one of the little tables outside, and perhaps your Highness will wear a thick veil, for a pretty woman in that quarter is so quickly noticed.”
  • 76. She smiled at his final words, but promised to carry out his directions. Surely it was a situation unheard-of—an escaped princess making a rendezvous with two expert thieves in order to receive back her own property. “Then we shall be there awaiting you,” he said. “And now I fear that I’ve kept you far too long, Princess. Allow me to take my leave.” She gave him her hand, and thanked him warmly, saying— “Though your profession is a dishonourable one, Mr Bourne, you have, nevertheless, proved to me that you are at heart still a gentleman.” “I am gratified that your Imperial Highness should think so,” he replied, and bowing, withdrew, and stepped out of the hotel by the restaurant entrance at the rear. He knew that the agent of police was idling in the hall that led out into the Rue St. Lazare, and he had no desire to run any further risk of detection, especially while that bag with its precious contents remained in the shabby upstairs room in the Rue Lafayette. Her Highness took little Ignatia and drove in a cab along the Avenue des Champs Elysées, almost unable to realise the amazing truth of what her mysterious rescuer of two years ago had revealed to her. She now saw plainly the reason he had left Treysa in secret. He was wanted by the police, and feared that they would recognise him by the photograph sent from the Prefecture in Paris. And now, on a second occasion, he was serving her against his own interests, and without any thought of reward! With little Ignatia prattling at her side, she drove along, her mind filled with that strange interview and the curious appointment that she had made for that evening.
  • 77. Later that day, after dining in the restaurant, she put Ignatia to bed and sat with her till nine o’clock, when, leaving her asleep, she put on a jacket, hat, and thick veil— the one she had worn when she escaped from the palace— and locking the door, went out. In the Rue St. Lazare she entered a cab and drove across the Pont des Arts, alighting at the corner of the Rue de Seine, that long, straight thoroughfare that leads up to the Arcade of the Luxembourg, and walked along on the left- hand side in search of the Café Vachette. At that hour the street was almost deserted, for the night was chilly, with a boisterous wind, and the small tables outside the several uninviting cafés and brasseries were mostly deserted. Suddenly, however, as she approached a dingy little place where four tables stood out upon the pavement, two on either side of the doorway, a man’s figure rose, and with hat in hand, came forward to meet her. She saw that it was Bourne, and with scarcely a word, allowed herself to be conducted to the table where an elderly, grey-haired man had risen to meet her. “This is Mr Redmayne,” explained Bourne, “if I may be permitted to present him to you.” The Princess smiled behind her veil, and extended her hand. She recognised him in an instant as the gallant old gentleman in the bright red cravat, who, on pretence of assisting her to alight, had made off with her bag. She, an Imperial Archduchess, seated herself there between the pair of thieves.
  • 78. Chapter Seventeen. In which “The Mute” is Revealed. When, in order to save appearances, Bourne had ordered her a bock, Roddy Redmayne bent to her, and in a low whisper said,— “I beg, Princess, that you will first accept my most humble apologies for what I did the other day. As to your Highness’s secrecy, I place myself entirely in your hands.” “I have already forgiven both Mr Bourne and yourself,” was her quiet answer, lifting her veil and sipping the bock, in order that her hidden face should not puzzle the waiter too much. “Your friend has told me that, finding certain letters in the bag, you discovered that it belonged to me.” “Exactly, and we were all filled with regret,” said the old thief. “We have heard from the newspapers of your flight from Treysa, owing to your domestic unhappiness, and we decided that it would be a coward’s action to take a woman’s jewels in such circumstances. Therefore we resolved to try and discover you and to hand them back intact.” “I am very grateful,” was her reply. “But is it not a considerable sacrifice on your part? Had you disposed of them you would surely have obtained a good round sum?” The man smiled. “We will not speak of sacrifice, your Highness,” the old fellow said. “If you forgive us and accept back your property, it is all that we ask. I am ashamed, and yet at the
  • 79. same time gratified, that you, an Imperial Princess, should offer me your hand, knowing who and what I am.” “Whatever you may be, Mr Redmayne,” she said, “you have shown yourself my friend.” “And I am your friend; I’ll stand your friend, Princess, in whatever service you may command me,” declared the keen-eyed old man, who was acknowledged by the Continental police to be one of the cleverest criminals in the length and breadth of Europe. “We have discovered that you are alone here; but remember that you are not friendless. We are your friends, even though the world would call us by a very ugly name—a gang of thieves.” “I can only thank you,” she sighed. “You are extremely good to speak like this. It is true that misfortune has fallen upon me, and being friendless, it is reassuring to know that I have at least two persons in Paris ready to perform any service I require. Mr Bourne once rendered me a very great service, but refused to accept any reward.” And she added, laughing, “He has already explained the reason of his hurried departure from Treysa.” “Our departures are often hurried ones, your Highness,” he said. “Had we not discovered that the jewels were yours, we should in an hour have dispersed, one to England, one to Germany, and one to Amsterdam. But in order to discover you we remained here, and risked being recognised by the police, who know me, and are aware of my profession. To- morrow we leave Paris, for already Hamard’s agents, suspecting me of the theft, are searching everywhere to discover me.” “But you must not leave before I make you some reward,” she said. “Where are the jewels?”
  • 80. “In that closed cab. Can you see it away yonder?” and he pointed to the lights of a vehicle standing some distance up the street. “Kinder, one of our friends, has it with him. Shall we get into the cab and drive away? Then I will restore the bag to you, and if I may advise your Highness, I would deposit it in the Crédit Lyonnais to-morrow. It is not safe for a woman alone to carry about such articles of great value. There are certain people in Paris who would not hesitate to take your life for half the sum they represent.” “Thank you for your advice, Mr Redmayne,” she said. “I will most certainly take it.” “Will your Highness walk to the cab with me?” Bourne asked, after he had paid the waiter. “You are not afraid to trust yourself with us?” he added. “Not at all,” she laughed. “Are you not my friends?” And she rose and walked along the street to where the cab was in waiting. Within the vehicle was a man whom he introduced to her as Mr Kinder, and when all four were seated within, Bourne beside her and Redmayne opposite her, the elder man took the precious bag from Kinder’s hand and gave it to her, saying,— “We beg of your Highness to accept this, with our most humble apologies. You may open it and look within. You will not, I think, find anything missing,” he added. She took the dressing-bag, and opening it, found within it the cheap leather bag she had brought from Treysa. A glance inside showed her that the jewels were still there, although there were so many that she, of course, did not count them. For a few moments she remained in silence; then thanking the two for their generosity, she said,—
  • 81. “I cannot accept their return without giving you some reward, Mr Redmayne. I am, unfortunately, without very much money, but I desire you to accept these—if they are really worth your acceptance,” and taking from the bag a magnificent pair of diamond earrings she gave them into his hand. “You, no doubt, can turn them into money,” she added. The old fellow, usually so cool and imperturbable, became at once confused. “Really, Princess,” he declared, “we could not think of accepting these. You, perhaps, do not realise that they are worth at least seven hundred pounds.” “No; I have no idea of their value. I only command you to accept them as a slight acknowledgment of my heartfelt gratitude.” “But—” “There are no buts. Place them in your pocket, and say nothing further.” A silence again fell between them, while the cab rolled along the asphalte of the boulevard. Suddenly Bourne said,— “Princess, you cannot know what a weight of anxiety your generous gift has lifted off our minds. Roddy will not tell you, but it is right that you should know. The fact is that at this moment we are all three almost penniless—without the means of escape from Paris. The money we shall get for those diamonds will enable us to get away from here in safety.”
  • 82. She turned and peered into his face, lit by the uncertain light of the street lamps. In his countenance she saw a deep, earnest look. “Then the truth is that without money to provide means of escape you have even sacrificed your chances of liberty, in order to return my jewels to me!” she exclaimed, for the first time realising the true position. He made no response; his silence was an affirmative. Kinder, who had spoken no word, sat looking at her, entirely absorbed by her grace and beauty. “Well,” she exclaimed at last, “I wonder if you would all three do me another small favour?” “We shall be only too delighted,” was Bourne’s quick reply. “Only please understand, your Highness, that we accept these earrings out of pure necessity. If we were not so sorely in need of money, we should most certainly refuse.” “Do not let us mention them again,” she said quickly. “Listen. The fact is this. I have very little ready money, and do not wish at this moment to reveal my whereabouts by applying to my lawyers in Vienna or in Treysa. Therefore it will be best to sell some of my jewellery—say one thousand pounds’ worth. Could you arrange this for me?” “Certainly,” Roddy replied, “with the greatest pleasure. For that single row diamond necklet we could get from a thousand to twelve hundred pounds—if that amount is sufficient.” She reopened the bag, and after searching in the fickle light shed by the street lamps she at length pulled out the necklet in question—one of the least valuable of the heap of
  • 83. jewels that had been restored to her in so curious and romantic a manner. The old jewel thief took it, weighed it in his hand, and examined it critically under the feeble light. He had already valued it on the day when he had secured it. It was worth in the market about four thousand pounds, but in the secret channel where he would sell it he would not obtain more than twelve hundred for it, as, whatever he said, the purchaser would still believe it to be stolen property, and would therefore have the stones re-cut and reset. “You might try Père Perrin,” Guy remarked. “It would be quicker to take it to him than to send it to Amsterdam or Leyden.” “Or why not old Lestocard, in Brussels? He always gives decent prices, and is as safe as anybody,” suggested Kinder. “Is time of great importance to your Highness?” asked the head of the association, speaking with his decidedly Cockney twang. “A week or ten days—not longer,” she replied. “Then we will try Père Perrin to-morrow, and let you know the result. Of course, I shall not tell him whose property it is. He will believe that we have obtained it in the ordinary way of our profession. Perrin is an old Jew who lives over at Batignolles, and who asks no questions. The stuff he buys goes to Russia or to Italy.” “Very well. I leave it to you to do your best for me, Mr Redmayne,” was her reply. “I put my trust in you implicitly.” “Your Imperial Highness is one of the few persons—beyond our own friends here—who do. To most people Roddy
  • 84. Redmayne is a man not to be trusted, even as far as you can see him!” and he grinned, adding, “But here we are at the Pont d’Austerlitz. Harry and I will descend, and you, Bourne, will accompany the Princess to her hotel.” Then he shouted an order to the man to stop, and after again receiving her Highness’s warmest thanks, the expert thief and his companion alighted, and, bowing to her, disappeared. When the cab moved on again towards the Place de la Bastille, she turned to the Englishman beside her, saying— “I owe all this to you, Mr Bourne, and I assure you I feel most deeply grateful. One day I hope I may be of some service to you, if,” and she paused and looked at him—“well, if only to secure your withdrawal from a criminal life.” “Ah, Princess,” he sighed wistfully, “if I only could see my way clear to live honestly! But to do so requires money, money—and I have none. The gentlemanly dress which you see me wearing is only an imposture and a fraud—like all my life, alas! nowadays.” She realised that this man, a gentleman by birth, was eager to extricate himself from the low position into which he had, by force of adverse circumstances, fallen. He was a cosmopolitan of cosmopolitans, a quiet, slow-speaking, slightly built, high-browed, genial-souled man, with his slight, dark moustache, shrewd dark eyes, and a mouth that had humour smiling at the corners; a man of middle height, his dark hair showing the first sign of changing early to grey, and a countenance bitten and scarred by all the winds and suns of the round globe; a wise and quiet man, able to keep his own counsel, able to get his own end with few words, and yet unable to shape his own destiny; a
  • 85. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com