Learning, Insight, and Innovation in Animals in the Context of Evolution

Interdisciplinary, unpaid research opportunities are available. Various academic specialties are required. If interested, email me at
dr.freedom@hotmail.ca.
Learning, Insight, and Innovation in Animals in the Context of Evolution
By Oleg Nekrassovski
The present paper deals with the historical progress of research on the topic of learning,
insight and innovation in animals, in the context of evolution. According to Goodenough,
McGuire, and Jakob (2010), giving a general definition of learning has proved to be very
difficult. However, after showing problems with several potential definitions, they manage to
outline a reasonably unproblematic definition. So according to them, learning is a change in the
probability of a certain behaviour occurring, which takes place as a result of experience but
excludes the effects of maturation of the nervous system, sensory adaptation, or fatigue
(Goodenough et al., 2010). Such a definition of learning, in spite of being rigorous, is too broad
to be useful. So it would help to define different types of learning. For our purposes it will
suffice to consider only two categories of learning: associative learning and social learning.
Associative learning is a type of learning in which a paired presentation of two stimuli
results in the formation of some kind of a mental connection, in the animal’s mind, between
these two stimuli (Goodenough et al., 2010). Associative learning consists of two types of
learning: classical conditioning and operant conditioning (Goodenough et al., 2010).
Classical conditioning works as follows. First, an animal has a specific inborn response
(called an innate reflex) to a certain natural stimulus. Next, an artificial stimulus to which the
animal initially has no response, whether inborn or otherwise, is presented to that animal right
before presenting the natural stimulus (Goodenough et al., 2010). Repeated presentation of the
two stimuli, in the stated order, eventually makes the presence of the artificial stimulus alone,
sufficient for activating the innate reflex. The apparent psychological connection, between the

natural stimulus and the artificial stimulus, which has thus been established in the animal’s mind,
is called a conditioned reflex. The conditioned reflex can be lost, just as easily as it was formed,
if the artificial stimulus continues to be frequently presented without being followed by the
natural stimulus (Goodenough et al., 2010).
Operant conditioning, also called trial-and-error learning, typically involves placing a
hungry animal into a box (called the Skinner box, after its inventor B. F. Skinner) with a food-
releasing lever and various non-functional features on the inside. The animal starts exploring the
inside of the box in the hope of finding food, until it accidentally presses the food-releasing
lever, which results in the immediate delivery of a small sample of food (Goodenough et al.,
2010). Since this food does not satisfy the animal, it resumes its food search inside the box until
it presses the lever again. Over successive cycles the animal gets faster and faster at getting to the
lever and pressing it because it learns to associate the food reward with the pressing of the lever
(Goodenough et al., 2010).
Social learning is simply learning from others. It includes imitation as well as other types
of learning (Goodenough et al., 2010). Imitation is a type of learning in which everything done
by a demonstrator is copied exactly by an observer. Many socially learned behaviours disappear
as quickly as they appear. Some, however, do not disappear, and instead spread through the
social group. Such socially learned behaviours are called traditions (Goodenough et al., 2010).
Giving an objective definition of animal innovation also seems to be very difficult.
Reader and Laland (2003) tackle this problem by first giving a thorough account of various,
often conflicting, scholarly opinions on this subject. Next they give a detailed analysis of various
definitions of innovation given by other scholars. Consequently, they conclude that innovation is
a novel behavioural end product of individual or group creativity, as well as a process of creation

and introduction of novel behaviours. Along these lines they arrive at two complementary
operational definitions of innovation. Innovation as a product: “An innovation is a new or
modified learned behaviour not previously found in the population” (Reader &Laland, 2003, p.
14). Innovation as a process: “Innovation is a process that results in new or modified learned
behaviour and that introduces novel behavioural variants into a population’s repertoire” (Reader
&Laland, 2003, p. 14). Insight is a special type of animal cognition in which, in spite of the
absence of prior experience, understanding seems to occur spontaneously (Goodenough et al.,
2010).
The Great Beginnings
The first two chapters of Darwin’s (1874) “The Descent or Origin of Man,” which was
first published in 1871 (Dewsbury, 1984), discuss the evidence for, and the manner of, primarily
physical, human evolution from some ‘lower’ form of organisms. The three subsequent chapters,
on the other hand, focus on comparing the mental qualities of humans with, and describing their
evolution from, those of other animals. In the process of this discussion Darwin elucidates, what
are now known to be primarily innately-caused, primarily experientially-caused, or caused by a
combination of both, animal behaviours. He also compares these animal behaviours with those of
humans and discusses their evolution. Also, Darwin attempts and largely fails to see whether one
or both of innate and experiential behavioural causes are responsible for which particular
behaviour. In addition, during the course of the same discussion, Darwin presents multiple
anecdotal cases of remarkably rapid learning, insight, and behavioural innovation in animals, as
well as describing how human social behaviour could have evolved from that of ‘lower’ animals.
Darwin’s (1965) The Expression of the Emotions in Man and Animals, which was first
published in 1872 (Dewsbury, 1984), is roughly divided into four parts. The first part is divided

into three chapters and focuses on describing the general principles of expression. In these
chapters Darwin describes his three chief principles of expression and gives evidence to support
their validity. During the course of this discussion Darwin puts significant stress on the learning
mechanisms, especially habitual repetition and associative learning, as being responsible for
many types of emotional expressions in man and animals. Also, in this section, a significant
attempt is made to show how various types of emotional expressions, whether in people or in
animals, could have evolved, as well as what adaptive value these emotional expressions have
now or could have had in the past. The second part deals with expressions of emotions in
animals, many types of which, as well as many specific cases of which, it describes in detail. The
third part, which is by far the largest, gives a detailed description of the multitude of emotional
expressions in humans. The fourth part (the last chapter) consists of concluding remarks, as well
as a summary of what has come before. The concluding remarks are noteworthy for their
proposal that the way many emotions are expressed in humans can be a result of imitation and
other forms of social learning.
In spite of its significant scientific value, The Expression of the Emotions in Man and
Animals suffers from two major flaws. One, is its overwhelming reliance on anecdotal evidence.
While the other is Darwin’s strong belief in the Lamarckian theory of the inheritance of acquired
characteristics. As a result, throughout the book, Darwin frequently alludes to the idea that a
particular emotional expression was initially acquired through some form of learning or
innovation and then fixed, in the organism’s behavioural repertoire, through habit. It was then
passed on hereditarily to subsequent generations, thus making this mode of emotional expression
innate.

Darwin’s influence on, and the significance for, all subsequent studies of animal
behaviour cannot be overestimated. Robert Boakes (1984) first announces his intentions to
describe the works, of some of the British philosophers and scientists, that gave rise to animal
psychology. And then states that the man, who stood head and shoulders above the rest of these
scholars, was Charles Darwin; whose book The Descent of Man was a key starting point for early
animal psychology. In a similar vein, the great ethologist Konrad Lorenz (1965) states that, in his
book The Expression of the Emotions in Man and Animals, Darwin has unintentionally predicted
the main theoretical problems of modern ethology and has presented a research strategy which is
still in use in ethology.
Also, in his introduction to a collection of articles on social learning in animals, Galef
(1996b) notes that a controversy between Charles Darwin and Alfred Russell Wallace, regarding
the evolution of the human mind from that of an animal by purely natural processes, triggered
the first systematic studies of animal imitation. Galef (1996b) further states that the second of
Darwin’s interests, distinct from his desire to demonstrate the animal roots of the human mind,
forms the roots of later work on social learning in animals. And in order to provide support for
this assertion, Galef quotes a paragraph from Darwin’s journal, consisting of observations of
bumble bees and honey bees successively feeding from the same flowers in an ‘ingenious’ way,
and subsequent theoretical conclusions. After quoting that paragraph Galef (1996b) goes on to
explain that in the quoted paragraph Darwin foresees the second major approach employed by
20th
-century animal behaviourists to study social learning. This approach, as Darwin eloquently
describes and Galef (1996b) eloquently explains, involves explaining the spread of adaptive
behaviours through a population of free-living individuals by assuming that social learning of
some kind is taking place.

Finally, in his book on the history of ethology, Burkhardt (2005) writes that Darwin’s
ideas were very broad and deep, and consequently begot many descendants, evident in very
diverse modern methodologies and problems. Aside from ethologists and early British field
naturalists, behaviourist psychologists could also claim to be Darwin’s intellectual descendants
(Burkhardt, 2005).
Subsequent Progress
Romane’s (1882) Animal Intelligence is a detailed compendium of animal behaviour. It
starts off by describing the method of deducing the mechanisms underlying animal behaviour
through anthropomorphic analogy. After that it proceeds to describing the main aspects of the
behaviour of molluscs, insects, fish, reptiles, birds, and mammals, with the sections on insects
and mammals being most expansive and detailed. Most sections are further divided into
subsections on emotions, memory, habits, and intelligence. The subsections on memory usually
describe various types of learning, while those on intelligence –insight and innovation. Almost
all the ‘evidence’ presented to support various claims regarding the existence of any particular
behaviour, is in the form of anecdotes obtained from other sources.
Throughout the various discussions and analyses of animal behaviour that occur
throughout the book, a periodic reference is made to the theory of evolution, and, equally
periodically, an attempt is made to analyse or interpret the particular behaviour under
investigation in evolutionary terms, or to find in the ‘lower’ organisms the traces of a particular
behaviour which finds its greatest expression in the ‘higher’ organisms.
The influence of other of Darwin’s ideas and works, upon Romanes, is also evident
throughout the book. The subsections on intelligence, especially when taken together, clearly
attempt to demonstrate the mental continuum from the human mind down to the minds of the

‘lowest’ organisms; very similar to that done by Darwin (1874) in his “The Descent or Origin of
Man.” Also the subsections, “emotions,” are very similar, in type of content and style of
exposition, to Darwin’s (1965) The Expression of the Emotions in Man and Animals.
In retrospect, Romanes’ (1882) Animal Intelligence turned out to be a collection of a few
interesting descriptions of, and experiments on animal behaviour (Galef, 1996a). However, it
mostly consisted of wildly implausible anecdotes of animal behaviour, and a consistent,
anthropomorphic interpretation of everything on the subject (Galef, 1996a).
E. Thorndike’s (1965a) “Animal Intelligence,” which was first published in 1898
(Thorndike, 1965b, p. v), eloquently criticizes the method of deriving facts of animal behaviour
from various anecdotes about it; skilfully showing its complete subjectivity and strong proneness
to bias. Consequently, in place of the anecdotal method, Thorndike proposes a strict,
experimental method, and moves on to describing it in detail. Along these lines, Thorndike does
his best to destroy misconceptions about animal behaviour created by preceding writers such as
Romanes, through their use of anecdote, anthropomorphism, and introspection. He accomplishes
this by discrediting the hypothesis that animals are capable of insight, by first presenting an
already established rival hypothesis that even the most extraordinary performances of animals
are solely a result of accident, associative learning or imitation. Then, in order to support the
later hypothesis and refute the former, Thorndike quotes a description of supposedly insightful
performances of animals from Romanes’ (1882) Animal Intelligence and then uses his objective,
experimental method to replicate the conditions under which the described performances took
place, and tests them. The results of these experiments refute Romanes’ hypothesis that the
performances he describes are a result of insight, while giving strong support to ‘Thorndike’s
hypothesis’ that such performances are a result of accident, associative learning, or imitation.

After that, “Animal Intelligence” discusses Thorndike’s experimental studies of the
various aspects of behaviour in cats, dogs, and chicks. It presents considerable evidence in
favour of the presence of trial and error and imitative learning in animals, and the absence of
insight, social consciousness, and other mental abilities possessed by humankind.
Thorndike’s (1965a) “Animal Intelligence” is also noteworthy for an argument that the
study of associative learning processes in animals is critical to the study of evolution of the
animal mind, because at least the ‘higher animals’ depend on associative learning, and the
consequent habits, for adaptation to their environments, and hence physical survival and
reproduction, far more than on instinct. Thus, according to Thorndike, associative learning is a
critical adaptive mechanism.
Hence, Thorndike’s (1965a) “Animal Intelligence” stands as a prime example of the great
contributions made by Thorndike towards the development of comparative psychology and
related fields (Galef, 1996a). These contributions lay in his demonstration of the need to study
animal behaviour in a systematic, quantitative manner and in his invention of practical
experimental methods for the study of animal behaviour (Galef, 1996a).
Pavlov (1958b) in his semi-historical first lecture of “Lectures on the Work of the
Cerebral Hemispheres,” which were first published in 1926 (Pavlov, 1958a), describes how
highly subjective, speculative attempts, by late 19th
century psychologists, to understand animal
behaviour gave way, around 1900, “to the experimental analysis of the subject, and from the
objective external aspect” (Pavlov, 1958b, p. 176). Pavlov (1958b) further states that he and
Thorndike were the main, independent, pioneers of this new approach. He then proceeds to
elucidate the phenomenon of innate reflex while underlining its indispensible role in the life-
preserving, organismic function of adaptation to constantly changing environment. Next, he

describes various kinds of innate reflexes and their difference from instincts. Finally, he points
out how the stimuli that trigger any particular innate reflex can change with the change in the
organism’s environment, and how this modification of reflex-based behaviour is critical for the
survival of the organism.
Pavlov’s significance lay in that he introduced new methods of research into the study of
higher nervous activity and brain functions. He managed to work out methods for investigating
the workings of intact (i.e. living) as opposed to dissected organisms, which allowed him to carry
out objective, experimental studies of animal behaviour (Koshtoyants, 1958). Pavlov’s discovery
of the conditioned reflex and his elucidation of the interaction between, and the mutual influence
upon each other of animal’s higher nervous activity and its environment, was a great step
forward in the history of neuropsychology (Koshtoyants, 1958). Pavlov also raised and
experimentally solved a problem of exceptional significance for animal psychology. This
problem concerned conditioned reflexes in animals and focused on their functional mechanisms
(Koshtoyants, 1958). It is also important to note that during the course of his research Pavlov
managed to demonstrate that conditioned reflexes have an adaptive function and can appear and
disappear in the course of individual development as a direct result of environmental influences
upon the individual (Koshtoyants, 1958).
In his article “On aims and methods of ethology,” Tinbergen (1963) presents his four-part
approach to the study of animal behaviour. The first part studies causation, or physiological
mechanisms underlying behaviour, especially those involving the endocrine and nervous
systems. Pavlov, who studied the neural mechanisms that underlie behaviour, especially the
various aspects of the innate reflex phenomenon, is a clear example of a behavioural researcher
who focused on studies of animal behaviour in the vein of the first part of Tinbergen’s approach.

The second part of Tinbergen’s approach is called ontogeny, or more simply – development, and
consists primarily of the study of how the organism’s experiences and physical environment
interact with its genotype. Pavlov’s research, as can be seen from its discussion in the previous
paragraphs, has a direct connection to this part of Tinbergen’s approach. Ontogeny studies all
types of learning processes that the organism goes or may go through during the course of its
life. It also studies all the insights that an organism makes or may make during the course of its
life. Finally, ontogeny also studies all the innovations which an organism adopts or may adopt or
to which it contributes or may contribute during the course of its life.
The third part of Tinbergen’s approach focuses on survival value or functional (i.e.
adaptive) significance of the behaviour to the organism that exhibits it. This part of the approach
focuses on how behaviour contributes to an organism’s physical survival and reproductive
success in the particular environment in which it lives. The fourth part of Tinbergen’s approach
focuses on how behaviour has evolved. This part of the approach can be further split into two
parts. Part one aims at describing the course of behavioural evolution. This is accomplished by
comparing the behaviours of closely related species, in order to see whether these behaviours
share a common ancestral behaviour, as well as to see what degree of evolutionary divergence
these behaviours exhibit. Part two focuses on elucidating the dynamics of behavioural evolution.
This gets done in two ways. The first way is nothing less than today’s behavioural genetics. The
second way is the study of how natural selection influences behavioural evolution. This problem
gets attacked from two angles. The first line of attack involves elucidating the various selection
pressures that are responsible for maintaining a particular behaviour in its present state, as well
as those selection pressures that may have been responsible for the emergence of the presently

observed behaviour. The second line of attack involves experimental (i.e. artificial) creation and
application of selection pressures to behaviour over a series of generations.
Tinbergen’s, above described, paper gave a strong framework for the study of animal
behaviour (Drickamer & Snowdon, 1996). This framework proved to be one of the most
encompassing, and by far the most influential of those that were developed during this period. In
the decades that followed, most, if not all, textbooks in ethology got structured according to this
framework. And an analogous thing happened at several, international ethology conferences
where the plenary sessions also got structured according to Tinbergen’s original framework
(Drickamer & Snowdon, 1996).
Dawkins’ (1989) article “The future of ethology: How many legs are we standing on?”
argues that contemporary ethology has significantly degenerated since its former days. She refers
to ethology’s traditional four-part approach (Tinbergen, 1963), and points out that, in its
contemporary form, ethology is a ‘one-legged animal.’ What she has in mind is that
contemporary studies in animal behaviour focus on survival value or adaptive significance of
behaviour, while largely failing to consider its underlying physiological mechanisms, its
development, or how it compares to that of related species. Dawkins is especially concerned with
the neglect of investigations of physiological mechanisms. Consequently she goes on to describe
the contemporary state of affairs of three areas of ethology ((1)neuroethology, (2) sociobiology
and behavioural ecology, and (3) applied ethology) in which, she argues, a neglect of
behaviour’s underlying physiological mechanisms has been most damaging. Thus, if Dawkins is
correct, there could have been no important studies on animal learning, insight, or innovation in
the 1980s. Moreover, even though in this article, Dawkins tries hard to persuade other ethologists

to ‘place ethology back, firmly on all of its four legs,’ it seems her efforts, as we shall see next,
came to nothing.
In their article “Is innovation in bird song adaptive?” Slater and Lachlan (2003) focus on
bird song learning. In the article the authors first discuss how bird song learning, being an
accurate process of cultural transmission, results in local traditions, and how these traditions can
have a positive effect on the fitness of the birds that practice them. The authors also show that in
spite of being accurately transmitted from generation to generation, bird song is subject to
individual innovation. So, during the subsequent course of the article the authors try to verify,
through a careful consideration of all available evidence, two predictions. Their first prediction is
that the initial practice of a new song type enhances the inclusive fitness of its inventor. And
their second, is that the subsequent practice of a new song type (when it has become a local
tradition) can be beneficial to the fitness of any member of that particular population of birds to
which its inventor belongs.
Thus, it is clear that even though this article deals with learning and, by extension,
innovation; an overwhelming emphasis is on how this innovation is, or may be, adaptive. Also,
while the article does a tiny bit of work in comparing the songs of related species, absolutely
nothing is said about the physiological mechanisms that underlie bird song learning. All of this in
spite of the fact that traditionally, bird song learning has been one of the most productive areas of
ethology (Barlow, 1989).The research in this area has traditionally focussed on the analysis of
song acquisition during early development. And an attempt was then made to relate this analysis
to the song’s neurological substrate and its evolutionary problems (Barlow, 1989).

References
Barlow, G. W. (1989). Has sociobiology killed ethology or revitalized it? Perspectives in
Ethology, 8, 1-45.
Boakes, R. (1984). From Darwin to behaviourism: Psychology and the minds of animals.
Cambridge: Cambridge University Press.
Burkhardt, R. W. Jr. (2005). Patterns of behavior: Konrad Lorenz, Niko Tinbergen, and the
founding of ethology. Chicago: The University of Chicago Press.
Darwin, C. (1874). The Descent or Origin of Man. In C. Darwin, The Descent of Man; and
Selection in Relation to Sex. New York, NY: D. Appleton.
Darwin, C. (1965). The Expression of the Emotions in Man and Animals. Chicago, Ill.: The
University of Chicago Press.
Dawkins, M. S. (1989). The future of ethology: How many legs are we standing on?
Perspectives in Ethology, 8, 47-54.
Dewsbury, D. A. (1984). Comparative Psychology in the Twentieth Century. Stroudsburg,
Pennsylvania: Hutchinson Ross Publishing Company.
Drickamer, L. C., & Snowdon, C. T. (1996). Defining the Goals, Approaches, and Methods. In
L. D. Houck, & L. C. Drickamer (Eds.), Foundations of Animal Behavior: Classic Papers
with Commentaries (pp. 71-86). Chicago: The University of Chicago Press.
Galef, B. G. Jr. (1996a). The Making of a Science. In L. D. Houck, & L. C. Drickamer (Eds.),
Foundations of Animal Behavior: Classic Papers with Commentaries (pp. 5-12).
Chicago: The University of Chicago Press.

Galef, B. G. Jr. (1996b). Introduction. In C. M. Heyes, & B. G. Galef, Jr. (Eds.), Social Learning
in Animals: The Roots of Culture (pp. 3-15). San Diego, CA: Academic Press.
Goodenough, J., McGuire, B., & Jakob, E. (2010). Perspectives on Animal Behavior. Hoboken,
NJ: John Wiley & Sons, Inc.
Koshtoyants, Kh. S. (1958). Ivan Petrovich Pavlov and the Significance of His Works. In I.P.
Pavlov, Experimental Psychology and Other Essays (pp.21-55). London, England: Peter
Owen Limited.
Lorenz, K. (1965). Preface. In C. Darwin, The Expression of the Emotions in Man and Animals
(pp. ix-xiii). Chicago, Ill.: The University of Chicago Press.
Pavlov, I. P. (1958a). Experimental Psychology and Other Essays. London, England: Peter Owen
Limited.
Pavlov, I. P. (1958b). Lecture one of Lectures on the Work of the Cerebral Hemispheres. In I.P.
Pavlov, Experimental Psychology and Other Essays (pp. 171-187). London, England:
Peter Owen Limited.
Reader, S. M., & Laland, K. N. (2003). Animal innovation: an introduction. In S.M. Reader, &
K. N. Laland (Eds.), Animal Innovation (pp. 3-35). New York, NY: Oxford University
Press.
Romanes, G. J. (1882). Animal Intelligence. London, England: Kegan Paul, Trench, & Co.
Slater, P.J.B., & Lachlan, R.F. (2003). Is innovation in bird song adaptive? In S.M. Reader, &
K.N. Laland (Eds.), Animal Innovation (pp. 117-135). New York, NY: Oxford University
Press.

Thorndike, E. L. (1965a). Animal Intelligence; an Experimental Study of Associative Processes
in Animals. In E. L. Thorndike, Animal Intelligence: Experimental Studies (pp. 20-155).
New York, N.Y.: Hafner Publishing Company.
Thorndike, E. L. (1965b). Animal Intelligence: Experimental Studies. New York, N.Y.: Hafner
Publishing Company.
Tinbergen, N. (1963). On aims and methods of ethology. Zeitschrift für Tierpsychologie, 20,
410-433.

Learning, Insight, and Innovation in Animals in the Context of Evolution

More Related Content

What's hot (6)

Similar to Learning, Insight, and Innovation in Animals in the Context of Evolution (20)

More from Oleg Nekrassovski (20)

Learning, Insight, and Innovation in Animals in the Context of Evolution