Dissertation Implicit Explicit Memory

University of Plymouth
Matthew D Jones
3/26/2015

1
Contents
Ethical Statement Page 2
Acknowledgements Page 3
Abstract & Title Page 4
Introduction Page 5
Materials & Methods Page 12
Results Section Page 18
Discussion & Conclusion Page 23
References Page 29
Appendices Page 31

2
Ethical statement
Participants were informed about details of the experiment, but were not told
outstandingly that they would be tested for implicit and explicit memory performance
on relevant tasks. Participants were told they would be taking part in a simple picture
judgement and recognition task. Participants were selected using the psychology
pool of participant system whereby participants could sign up to the corresponding
studies. Here participants were informed briefly about the contents of the experiment
its duration and its value (monetary or points). Participants were required to give
signed and written consent upon arrival to their participation. Participants were then
given a brief, explaining the direction of the experiment and briefly its intentions.
Participants were not told exclusively the aim of the experiment to avoid any
confounding variables influencing the results. Participants were all informed they
could stop participating at any time throughout the study and would not be penalised
for doing so. Once the study had been completed participants were fully debriefed
and provided with an explanation of the experiment and its purpose; testing the effect
varying exposure duration has over implicit and explicit memory performance.
Participant’s personal information and the data recorded from the study are kept
completely confidential and will not be published to any public domain. Participants
were given forms with contact details of the experimenter and the project supervisor.
This allowed participants to make future contact to discuss any implications of the
research.
The data was collected using mat Lab, a program which allowed the replication of
Zago et al., (2005) study to materialise. All data was handled and retrieved by
Matthew David Jones. The data from each participant was input into Microsoft Excel
and subsequently IBM SPSS Statistics for further analysis.

3
Acknowledgements
I would like to take this moment to thank someone whom was crucial to this project.
Dr. Chris Berry you have been an amazing supervisor. Chris is somebody who is
always on hand to meet and would regularly keep in touch and go the extra mile
supporting me in this project. I hope I have done myself and you justice in this paper.

4
A behavioural account of how implicit and explicit
memory are affected by a variation of exposure
durations; do they derive from the same or different
neural mechanisms?
Matthew David Jones
10402968
University of Plymouth
Abstract
How does exposure duration of a stimulus affect implicit and explicit memory
systems? (Zago et al., 2005). To find out, we manipulated six exposure durations
(40, 150, 250, 350, 500 and 1900ms) and tested the performance of these durations
for both implicit and explicit memory. Our findings illustrate that priming initiates very
early on into the stimulus presentation and that longer exposure does not necessarily
consolidate a better representation of a stimulus. Elements of the “rise” and “fall”
phenomenon are present in our study and a clear threshold in priming magnitude is
observed (350ms).Similarities between explicit and implicit groups were witnessed
when both excelled in performance for a 1900ms exposure duration suggesting both
systems are closer related than initially anticipated.
(120)

5
Introduction
The explicit memory system provides us with everyday memory; it is a verbal
conscious type of memory and primarily is used to recall experience. Explicit
memories are stored in chronological order and are sequenced together to create an
organized “file” of personal past experiences, consequently making a personal
unique history timeline. Explicit memory begins to develop at around 6 months old
(Nelson, 1995; 1997; 2000); however explicit memory does not reach full
development and become embedded into the human brain until the age of 5. Explicit
memories are often performed when re-calling everyday things such as ones
favourite birthday or holiday with the family. Implicit memory is the unconscious
memory system which is used to perform everyday activities that we execute over
and over again. For example, playing a sport; participating in the sport over
prolonged periods of time results in an improvement of performance which in turn
results in the activity becoming second nature, and consequently does not require
conscious thinking.
Research has indicated between the ages of 0-5 years, is the most impressionable
stage for implicit memory, requiring a safe and protected environment for implicit
memory to blossom wholly. If we repeat certain experiences multiple times we realise
that it becomes part of the implicit memory. These can be both positive and negative
so the need for a protective environment at an early stage is un-paralleled. Implicit
memory is largely sensation based, is mediated by the Amygdala and it is very
reactionary based and is a response to certain triggers (Knight et al, 2009).
Priming has been a major stakeholder in showing the effects of implicit memory. One
major discussion point amongst psychologists is its contrary results to those
witnessed in recognition tasks, which is testing our explicit memory, a conscious

6
memory recall. The differing performance on the relevant tasks (priming vs.
recognition) do suggest that there is strong evidence pointing towards a double-
dissociation (Voss & Gonsalves, 2010). Participants exposed to a stimulus for a brief
period of time (250ms) yielded a higher rate of priming, whereas recognition
increased when participants were presented with a longer duration of stimulus
(2000ms), suggesting a relationship between encoding duration time and recognition
performance(Voss & Gonsalves, 2010).Despite a wealth of research that suggests
there is a double dissociation between priming and recognition, a strict criterion for
classifying dissociation (reversed associations) prevents research excluding a single-
process with multi-functions entirely (Dunn & Kirsner, 1988; Voss & Gonsalves,
2010).
Previous research has debated the source of both implicit (non-conscious) and
explicit (conscious) memory. The debate usually consists of disagreement with
regards to implicit and explicit deriving from the same single source, or a
multifunctioning memory mechanism, that can account for both types of memories
declarative and non-declarative. One such study that suggests implicit and explicit
memory came from two distinct separate neural substrates is the work by Crowder,
Robert G.; Mary Louise Serafine and Bruno H. Repp (1984). Crowder et al. (1984)
research employed priming procedures in the study to investigate possible effects of
isolating and testing solely implicit memory. Crowder presented the first half of the
participants with familiar American folk music whilst the other half were presented
with the same familiar American folk music’s melody but with an entirely new set of
lyrics played over the top of the melody. The first half of the participants, who were
exposed to familiar American folk music (with the genuine lyrics being played), had a
much higher chance and rate (0.92) of recognizing the music and labelling it as

7
“familiar” than the second group. The second group who were exposed to “old words”
combined with a “new melody” showed a lower rate of recognition (0.78). From this
research it has been suggested that there is an implicit association being made in the
brain between the lyrics of the song and the melody of the song. Here, the memories
of the participants (able to recognize the American folk music) are implicitly creating a
single memory combining two separate entities; which are represented by both the
lyrics and the melody together, which later cannot be separated from each other. As
the participants in the second half of the research were exposed to a melody with
new lyrics over the top of it, recognition of the original song decreased. This suggests
that the brain’s memory systems couldn’t combine the two separate entities (the
melody and the new lyrics) together to form a memory that can be recognized and
retrieved as they have never been exposed to that melody alongside the set of new
lyrics. This was unlike the participants from the first half of the experiment as
mentioned previously, who were able to perform a higher rate of recognition as the
brain’s memory system had implicitly merged both the lyrics and melody to the same
memory. In the results Crowder et al. (1984) reports that a component is recognized
better when presented alongside the original context the component was presented
in, than when a component is presented alongside a new stimulus (page 294). It is
not the level of familiarity one has with the two contexts, but whether they have been
paired in the initial perception (Crowder et al. 1984: p294).
When considering the nature of the mechanism that drives both implicit and explicit
memory it is important to firstly gain an understanding of how the mechanism(s)
performance differs over time. If both explicit and implicit memories performance
deviate over differing time exposures it would suggest two differing neural substrates.
It has been suggested that behavioural priming performance maxes out at around

8
250 Ms for a previously encountered stimulus and then performance begins to
decrease as the duration of prior exposure to a stimulus increase. This in psychology
is referred to as the “rise” and “fall” phenomenon. This display of a “rise and fall”
pattern in behavioural priming has been witnessed in varying stimuli exposure
duration (Zago et al., 2005). It has been found that duration to a stimulus as low as
40 Ms (and lower) has yielded significant influence in improving priming performance.
This brief period of exposure (40 Ms) has shown a better performance in behavioural
priming than when the stimulus presentation was of a lot longer duration (Bar and
Biederman, 1998; 1999). Other studies that have investigated the duration effect on
behavioural priming support the result from Zago et al., (2005). One example of the
“rise” and “fall” phenomenon that supports the results of Zago et al. (2005) was the
research conducted by Crabb and Dark (1999;2003, experiment 2). This research
yielded similar results and displayed a “rise” when the exposure duration increased to
200ms but “fell” when the duration was increased up too 300ms. Crabb and Dark
second study (2003) showed that a further “fall” was witnessed when the duration
rose to 600 Ms and fell furthermore when duration of a prior stimulus rose again to
1000 Ms. Other research that looks into whether implicit memory is affected by
exposure duration again found corresponding results with Zago et al. (2005). A
shorter presentation of priming led to an increase in performance of behavioural
priming whereas longer exposure to the stimulus caused weaker behavioural priming
performance and even caused negative priming to occur (Barbot & Kouider, 2012;
Faivre & Kouider. 2011; Huber & O’Reilly, 2003). The reasons for this “rise” and “fall”
effect with regards to priming is unclear but an attempted explanation has been put
forward by Zago et al. (2005). Zago et al. (2005) proposed the cross model of
“selection” and “sharpening”.

9
“Selection” (Wiggs & Martin 1998) uses the important features of a stimulus as the
representation of that previously viewed stimulus, whilst discarding any non-essential
information. Selection occurs more prominently with extensive exposure duration to a
stimulus, selection operates on a high-level information and semantic knowledge. As
this “selection” model is being used, a small overlap between the features within the
representation of the stimulus and the target stimulus actually decrease, causing a
decline in priming. “Sharpening” is the second aspect of the combined model
proposed by Zago et al. (2005) to explain the “rise” and “fall” of priming. “Sharpening”
however suggests that neurons in the inferior temporal cortex represent only global
properties at 130 Ms but then the coding becomes more specific from 130 Ms
onwards and is stimulus specific at around 240 Ms (Tamura & Tanaka, 2001).
Despite this, the “sharpening” aspect of the cross model, is not widely accepted and
credited as the explanation for the “rise” and “fall” of priming.
Still, there does remain a large amount of empirical evidence that can be interpreted
to represent a single multi-functioning mechanism in the brain used for memory
across all functions. One example would be damage caused in the brain to the
medial temporal lobes (MLT); this was shown to hinder both implicit and explicit
memory. Now, although the explicit memory function and performance was hindered
more extensively than the implicit function, the fact that two “opposite” functions
(conscious and sub-conscious) are impacted by the same physiological component
does suggest that there are some overlaps between the two (Berry et al. 2014). One
interpretation of the research by Berry et al., (2014) is that because the MTL in the
brain region was damaged and consequently did affect both the abilities in explicit
and implicit memory tasks, then it would be logical to suggest that the two behave
similarly and would likely to be from a single multi-functioning system. If this was the

10
case and both explicit and implicit memory were both driven by a single system, it
would be expected that the performance over differing stimulus exposure durations
would be also similar. However, the fact that these two entities are both divided by
consciousness, they should rather act in mirror parallels of each other.
The primary concern when looking at stimulus exposure duration on explicit and
implicit memories is the repetition-related response. Repetition-related response
according to Zago et al., (2005) by definition is “associated with the level of
experience an individual has with a particular stimulus “. Despite this clear definition,
the manner in which it holds the relationship between stimulus duration and cortical
representation is somewhat absent. This was the main impetus behind the start of
the experiment and consequently stimulus duration was manipulated. The aim of the
current research was to investigate the effects of study exposure duration on
repetition priming and recognition memory. Participants were split into two conditions
and would be taking part on the conscious explicit memory task or the non-conscious
implicit memory task. The two conditions were both split into two identical sub-
conditions, which consisted of 6 sub conditions. Here the exposure duration was
manipulated between six differing durations .The research in question used a method
very similar to that witnessed in Zago et al. (2005, p 1656) however there were a few
adaptations that took place. The difference in the present investigation was that: One
of the groups completed a procedure to measure the performance of implicit memory
by using the semantic judgement task (known as the priming condition). The second
study group completed an explicit memory task known as the recognition task.
However, despite this interest on the varying performance of implicit memory, little
investment has been made on examining how explicit memory and recognition is
influenced by a change in stimuli exposure duration. Assuming that implicit and

11
explicit memory both derive from two distinct separate neural entity’s, that both
function in the opposite manner (consciously and sub-consciously) and considering
the results of previous research, it is logical to hypothesize that; as the exposure
duration to a stimulus increases, so does the performance of behavioural priming (up
to a threshold of 250 Ms). Once the threshold has been breached performance in
implicit memory tasks decreases .In contrast to implicit memory performance, the
second hypothesis assumes that as exposure duration to a stimulus increases the
performance on the explicit memory task (recognition) will also increase, this is
because subjects will have more time to consciously observe and “learn” the words. If
exposure duration to a stimulus does increase the performance of conscious
recognition and recollection and then exposure duration beyond 250 Ms hinders
behavioural priming performance, then it would suggest that there are two distinct
neural mechanisms at work with regards to explicit and implicit memory.

12
Materials and Methods
Participants
A total of 20 participants participated in our experiment. The mean age of the
participants in the implicit memory condition was 19.6 years and for the explicit
memory group the mean age was 23 years. In total there were 7 males and 13
females. These participants were recruited by using an opportunity sample from
Plymouth universities psychology pool website, consequently making all participants
psychology students. The participants were split into two separate groups, equally
divided so there were 11 participants in each condition. All participants had given full
written consent to partake in this study prior to the research. The Plymouth University
Ethics Committee granted ethical clearance for this experiment.
Stimuli and Apparatus
The stimuli used in this study were 425 pictures of familiar objects such as elephants,
tools, glasses, vehicles and food (Fig. 1). The dimensions of the images presented to
the participants on a computer screen with a plain white background were recorded
at 6cm height x 6cm width. These 425 images were randomly allocated to the
following conditions: 60 pictures x 6 repeat conditions (40, 150, 250, 350, 500 and
1900ms), 60 pictures x 1 new condition (to appear in the semantic judgement task
once only). As this experiment was an adapted replication of Zago et al. (2005), the
same 10 masks seen in Zago et al’s experiment were used here. These masks
consisted of very random images creating what is referred to as a “nonsense” image
that is not relatable to the object stimuli.
The images were appearing on the computer screen that was measured at 22”.
Participants were required to respond to judgements during the experiment by

13
pushing the relative button on the keyboard: “N” for natural, “M” for manufactured,
specific for the priming condition. Also, the numbers keys: 1, 2, 3 and 4 were required
for a confidence rating in the recognition condition.
Fig. 1
Here are some of the stimuli that participants saw. For the recognition condition
participants were required to make a judgement on their confidence rating’s as to
whether the image had been previously presented or not. In the priming condition, a
participant simply was required to make a judgement on whether the image on the
monitor was a manufactured object or is a naturally occurring object.

14
Design and Procedure
There were six different stimuli durations that each participant completed regardless
of condition. These six durations were 40, 150, 250, 350, 500 and 1900 Ms. Each
trial had a duration set at 2 seconds or 2000 Ms. In total there were 13 different
conditions for the object presentation. Both recognition and the priming condition had
the same 6 different exposure conditions and the same repeat condition that was set
at 500 Ms. At this point, a judgement was required from the participant again
dependant on which condition the participant was selected for. The total trial duration
as aforementioned is 2 seconds. The two-second period consisted of object
exposure duration that varied (40, 150, 250, 350, 500 and 1900 Ms) and the mask
duration that also varied. Hypothetically if the object were presented for 500 Ms then
the mask would be presented for 1500 Ms. The mask also symbolized the start and
end of that specific trial.
Priming Group
The implicit memory task that participants were required to complete in this condition
was a semantic priming task. The implicit memory task was measuring priming
performance and the independent variable was the stimulus duration. Images that
were presented once or only twice were intermixed and amongst other images that
had been previously repeated for multiple trials. The participants had a blank screen
with nothing but a single fixation point (‘+’), which was presented in Arial 28pt on the
screen. This was to be presented at the centre of the screen for 500ms duration.
Where the fixation point was situated, was the location that the images appear during
the continuous stream in the study phase. Each object was presented to the
participant on the monitor in front of them for a randomly selected duration (40, 150,

15
250, 350, 500 & 1900ms). Post the image being presented, the mask immediately
followed and represented the start and end of a trial. The masks were presented for
differing durations. One of the 10 masks was randomly selected on each trial to be
presented for the remainder of the 2000ms trial. Hypothetically, if the participant was
shown an object that was displayed for 500ms, then the mask would be present on
the screen for 1500ms, confirming that the total trial duration is at 2000ms. The
streams of images were consisting of either naturally occurring objects or
manufactured ones. The job of the participant was to realise and make a judgement
as fast as possible as to which of the two categories the object in question falls into
(as seen in figure 3 in the appendix). As aforementioned each trial lasts for a total
duration of 2000ms. After an item was presented for its specific exposure duration
(40, 150, 250, 350, 500 & 1900ms), a mask (nonsense image) was presented for the
remaining duration of the 2000ms trial. The trials were presented consecutively one
after the other, creating a constant stream of images appearing. Every block of 130
trials that the participant completed, signalled the start of 15 second break to allow
the participant to again prepare themselves.
Recognition Group
The explicit memory task that participants were required to complete in this condition
was a simple recognition task; the independent variable again was the stimulus
exposure duration. The participants had a blank screen with nothing but a single
fixation point (‘+’) which was presented in Arial 28pt on the screen. This was to be
presented at the centre of the screen for 500 Ms duration. Where the fixation point
was situated, was the location that the images appear during the continuous stream
in the study phase. The study phase consisted of 360 images. Again, the same
intervals between stimulus trials were set; each trial totalled up to duration of

16
2000ms. The total duration of each independent trial was 2000 Ms, the 2000 Ms was
made up of the stimulus duration which could be set at: 40, 150, 250, 350, 450, 500
or 1900 Ms, the remaining length of time would be made up by the mask (nonsense
image), completing the 2000 Ms elapsed time. The difference between the
recognition group and the priming group is the way that the test phase was
conducted. The participants in the recognition group were asked again to make a
judgement but instead were rating their confidence against whether the image
presented in front of them had been seen before in the study phase. The participants
were presented with a choice of four keys to hit (as seen in figure 4 in appendix). The
choices were high and low confidence ratings that the image wasn’t presented
before. The other two options were high and low confidence ratings on whether the
image had appeared before. During the test phase for the recognition group, the
participants were presented with a total of 150 items, 90 of which were old items
whilst 60 of them new items previously unseen to the participant.
The whole procedure was running for an approximate 25-30 minutes for both
conditions. A smaller number of stimuli were used in the recognition condition to keep
the duration of both priming and recognition conditions the same. Participants were
fully briefed prior and debriefed post the research. The experiment began with the
participant’s being able to partake in practice trials prior to the study commencing.
These practice trials gave the participant time and a chance to understand what was
expected of them through the duration of the study. Figure 2 shows the computer
screen image that the participants were presented with at the start of the study. Here
the participants are told that they are taking part in a categorization task. They were
informed that a series of images were to be presented in succession, alongside these
images, they were presented with a judgement; whether the image on the screen

17
was manufactured or was a natural object. Participants were informed that they
should aim to be as quick and concise as possible throughout the research.

18
Results
Our main findings are that: (i) with the increased magnitude of exposure duration to a
particular stimulus an increase in recall ability was profound, it should be said that the
increase in exposure duration facilitated the consolidation of a stimulus in conscious
recognition and (ii) that an increase in the magnitude of exposure duration in the
priming task displayed a variety of performance patterns; Most notably a decline in
performance on the 150,250 and 500ms conditions whereas the 40, 350 and 1900ms
conditions showed an increase.
Behavioural Results
On average the difference in mean response time (RT) in the priming task varied
greatly between the six conditions (40,150,250,350,500 & 1900ms). A repeated
measure ANOVA was conducted and specified that exposure duration (as witnessed
in the experiment) produced a significant effect on the behavioural priming task F (1,
9) = 12.95, p = 0.006, however it was not significant across the whole experimental
design and conditions F (5, 45) = 2.81, p = 0.27 (Fig.10). On the recognition task a
repeated measure ANOVA was also conducted and specified that there was a
significant effect for exposure duration on the recognition performance F (1, 9) =
33.55, p = 0.00** and was also significant across all conditions F (5, 45) = 3.80, p =
0.006.
A one sample t-test was conducted on the exposure durations for both memory
types, implicit and explicit. For the implicit memory group, the 6 exposure duration
conditions produced the following statistics; 40ms condition (M = 37.80, SD = 39.18),
150ms condition (M = 29.22, SD = 28.76), 250ms condition (M = 31.18, SD = 32.62),

19
350ms condition (M = 46.10, SD = 38.59), 500ms condition (M = 22.34, SD = 25.66)
and the 1900ms condition (M = 43.03, SD = 39.05).
The explicit memory group data also underwent a one sample t-test for the same 6
exposure duration conditions and produced the following statistics; 40ms condition
(M = 0.31, SD = 0.20), 150ms condition (M = 0.34, SD = 0.25), 250ms condition (M =
0.43, SD = 0.25), 350ms condition (M = 0.41, SD = 0.23), 500ms condition (M = 0.42,
SD = 0.26) and the 1900ms condition (M = 0.52, SD = 0.29).
Effect of exposure duration on priming performance
To be able to evaluate the effect that previous exposure to a particular stimulus has
on the performance of behavioural priming, priming values were calculated by
subtracting the response times (RT) participants provided in the repeat condition
from the RT’s for the new condition. The results obtained accept the null hypothesis.
The briefest exposure duration of 40ms produced the third largest mean difference in
reaction time between items presented “new” against items that were “repeated
(37.82). However, unlike previous research, the performance of priming was
hindered by an increase of exposure duration to a stimulus up to 150 and 250ms,
when this exposure duration was increased to 350ms; the largest difference in RT
between new and repeated stimuli was produced (46.09). However, when the
exposure duration increased up to 500ms, the RT between new and repeated stimuli
dropped (29.21) to the lowest score. However, when the stimulus exposure duration
reached the longest period of 1900ms, the difference was again large and
behavioural priming performance was increased (43.04). Furthermore, T-tests
showed that 1900ms of previous exposure to a stimulus caused a greater magnitude
of priming than all the other conditions. Despite that, exposure at 40ms caused a

20
greater magnitude of priming than 150, 250 and 500ms of exposure. This suggests
only brief exposure is necessary for behavioural priming to develop and perform
effectively, however the results also indicate that the prolonged exposure condition
also contributed to an increase in priming performance more so than the 40ms
condition and the 350ms condition.
Fig 2. Behavioural data. The mean magnitude of priming calculated by deducing the
mean RT for old objects from the response time of new objects. Maximal priming was
observed at 350ms.
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
45.00
50.00
40 ms 150 ms 250 ms 350 ms 500 ms 1900 ms
Priming:MeanRTnew-meanRTold
Exposure Duration
Priming performance over 6 exposure durations

21
Effect of exposure duration on recall ability
To be able to successfully evaluate the effect differing exposure durations had upon
recognition performance, scores for recognition performance were calculated with the
use of the “hit” and “false alarm” method. A “hit” is where a participant correctly
identifies the stimuli presented on the screen in front of them as either new or old. A
“false alarm” is where a participant incorrectly labels a new image old or old image
new. The difference in scores between “hits” and “false alarms” was used as a
measure of performance on recall tasks. Therefore the higher the difference between
the two scores, the better the recall performance.
As witnessed in the statistical analysis for the recognition task, a significant positive
correlation was found between exposure duration and recall performance. Because
there was an increase in exposure duration, the performance on the recall task
invariably, also increased. This rejects the null hypothesis supporting the notion that
as a stimulus is presented for a longer duration, consciously, it becomes easier to
recall information about that episode at a later time. The scores progressively
increased with exposure duration. The briefest exposure duration was 40ms and as
expected scored the lowest in terms of difference between the “hit” and “false alarm”
rate (0.31). The second lowest duration (150ms) scored a difference of 0.34, this was
as expected, an improvement upon the previous exposure duration. The 250, 350
and 500ms all scored very similarly, scoring 0.43, 0.41 and 0.42 respectively. Finally
the highest scoring exposure duration condition for recognition testing explicit
memory was the 1900ms condition.

22
Fig 3. Behavioural data for the mean magnitude of recall performance; this was
calculated by deducing the mean false alarm rate from the mean hit rate. Maximal
recognition ability was induced from the 1900ms stimulus presentation.
0.00
0.10
0.20
0.30
0.40
0.50
0.60
40 ms 150 ms 250 ms 350 ms 500 ms 1900 ms
Meanhitrate-meanfalsealarmrate
Exposure Duration
Recall ability over 6 exposure durations

23
Discussion
The present study intended to investigate whether exposure duration had an effect
on implicit and explicit memory performance. The procedure displayed a clear
difference in the way implicit and explicit memory respond to varying exposure
durations. The results demonstrated that varied exposure duration in the implicit
memory group caused varying magnitudes of priming whereas the explicit memory
recognition task produced a positive correlation between exposure duration and
recall ability. The results support the observation by Bar and Biederman (1998; 1999)
that prominent levels of priming are present even when a stimulus is presented for a
brief exposure duration. Notably, Perfetti and Bell (1991) also found significant levels
of priming to be present after a 45ms stimulus presentation .In our results, high levels
of priming were present after 40ms, supporting the notion that priming is extremely
prevalent even after brief exposure.
The present results demonstrate that in the implicit memory group, the 350ms
condition facilitated the maximal magnitude of priming. The threshold level is said to
be the point at which the cortical object representation is at its maximum .The
threshold value in our study for priming magnitude is not witnessed across much
previous research, whereby thresholds for maximal priming have been reported to be
set around the 200ms mark for stimulus duration (Crabb & Dark, 1999; 2003,
experiment 2). Larger thresholds have also been reported but only up to 250ms
(Zago et al., 2005). Despite the differences in threshold values, the present study
does present mutual trends with previous research investigating exposure duration.
For example, the magnitude of priming decreasing once the threshold value has
been breached (Crabb & Dark, 1999; 2003) is reflected in this study. Priming
magnitude was at its strongest in the 350ms condition (46.09) and then weakest in

24
the subsequent 500ms condition (22.34). This deviation in magnitude of priming is
referred to as the “rise” and “fall” phenomenon and is prevalent in the present study.
Despite features of “rise” and “fall” being present, aspects of the data fail to support
this phenomenon. For example, stimuli presented for 350-1900ms produces a
weaker priming magnitude when compared to the priming magnitude produced by
stimulus exposure duration between 40-250ms (Zago et al., 2005). This is contrary to
the results in the present study which exhibit large priming effects for exposure
durations post the threshold (1900ms) and little priming effects for exposure
durations prior to threshold, excluding the brief 40ms condition. Consequently, it
would be more appropriate to assume the way magnitude priming behaves over
varying exposure durations is not represented reliably in this study. Despite this,
similar phenomenon’s such as the “rise” and “fall” effect and priming being present
post brief exposure to a stimulus are still prevalent in the present study.
Another phenomenon that is associated with implicit memory is “fine-tuning”. Fine
tuning is a function that is part of the “sharpening” process used in implicit memory to
preserve and consolidate key characteristics of a stimulus for future identification
(Zago et al., 2005). Fine-tuning occurs early into the presentation of the stimulus.
With the use of fMRI, activity in inferior temporal coincides with the fine-tuning
function and begins to instigate at 130ms into the stimulus presentation and is
functioning at its optimal level around 240ms into the stimulus presentation (Tamura
& Tanaka, 2001). This is the around the same duration into stimulus presentation that
the threshold for priming performance is realised, suggesting that the two are
coexistent. In our research, the 350ms condition is the threshold peak, which would
mean that the fine-tuning process is functioning at optimal capacity when stimulus
presentation lasts 350ms.

25
However, it would be inappropriate to directly state our findings support the fine-
tuning phenomenon because magnitude of priming should increase until the
threshold peak is reached, however the results indicate that priming magnitude
declined between 150-250ms. Priming strength was higher at 40ms (37.82) than the
subsequent 150ms (29.21) and 250ms conditions (31.18). However, our results
should not be interpreted entirely as having an absence in fine-tuning but rather the
continuum of priming scores reflecting fine-tuning, were not performing in a
“traditional” manner as reported by Tamura and Tanaka (2001).
Results from the explicit memory group show that as exposure duration increases as
does the performance on the subsequent recognition task. This is consistent with
previous findings that state the duration of rehearsal has a significant increase on
recognition task performance (Greene, 1986) and other explicit memory tasks
(Debner and Jacoby, 1994). Another factor that influences conscious recollection is
the number of studied items, the larger the number study items, the weaker the
consequential recall (Mandler, 1985). Although this can reduce the performance of
conscious recollection, this confounding variable did not contaminate our results
because the numbers of studied items were standardized. Therefore, the effect
witnessed was a direct result of exposure duration manipulation.
The explicit memory group reacted very different to the varying exposure durations in
contrast to implicit memory. Unlike the implicit memory condition, a threshold peak is
not present in the explicit memory group. For the explicit memory condition there is
no optimal exposure duration, the only relationship is as duration increases,
recognition also surges. However, there are some mutual aspects of implicit and
explicit memory performance to varying exposure duration in the present study.
Unexpectedly the 1900ms condition produced the second largest priming score out

26
of the six conditions. This was an unforeseen effect as priming performance is meant
to decrease post the threshold being breached. Similarly, the condition with the
highest recall rate was the 1900ms condition. As a similar trend has been witnessed
for both memory types when exposed to a 1900ms stimulus, it would seem logical to
suggest both types may coincide, this interpretation suggests either; (i) after an
exposure duration has been breached, conscious memory aids implicit memory
function, or; (ii) explicit and implicit memory come from the same source and act
more similar than initially anticipated. This supports research concluding that regions
in the brain that specifically contribute to explicit memory, such as the hippocampus,
are activated during implicit memory tasks (Henke, 2010). Furthermore, implicit
memory, despite being completely independent from other forms of memory, actually
interacts with many types of memory (Tulving & Schacter, 1990).This high score of
priming in the 1900ms condition could reflect a behavioural effect of conscious
memory influencing implicit memory, or represents an overlap between conscious
and non-conscious memory sources.
This is a plausible suggestion as it’s not the first time implicit memory has been
potentially contaminated by the influence of explicit memory involuntarily (Klavehn et
al., 1994). An example would be when a participant becomes conscious that the
stimuli have been previously presented to them, causing explicit memory to aid their
judgement. This is a possible explanation as to why the 1900ms condition produced
such a high level of priming magnitude. As reported by Berntsten (2010), involuntary
explicit memory recall can develop with the observation of a stimulus that acts as a
cue. Although ecologically valid to real life circumstances, such as witnessing an
advertising board that caused a conscious memory of an experienced life event, the
present study can be considered to represent a behavioural response to the

27
contamination of implicit memory by conscious memory under controlled settings.
The high priming magnitude for the 1900ms condition supports the claim by Cabeza
& Dew (2011) whom suggest that implicit and explicit memory is initiated by the
repetition of stimuli and that a threshold is breached and when breached, priming
converts itself into a conscious “signal of oldness” (Page 180). If hypothetically
correct our research shows a potential point of transition between sub-conscious
priming into conscious recall occurs between 500ms-1900ms from stimulus onset.
The results suggest a clear difference in the performance of implicit and explicit
memory over varying exposure durations. Differences in behaviour over time,
suggests both types of memory derive from a multiple memory system that supports
each type individually. The results can be interpreted to support a multiple model
system; the implicit memory performance reflects procedural memory because
priming was present even with duration of 40ms. Even though the reversed U-shape
arc which is what is expected to be present when displaying priming over differing
exposure durations was not achieved, the two (explicit and implicit memory) behaved
contrastingly to each other. This supports previous models that divide memories by
level of consciousness (Squire & Knowlton, 1995; Squire, 2009). In contrast to
implicit memory, the explicit memory condition showed examples of how it revolves
around declarative memory. This was shown by the increase in performance being a
direct result of an increase in exposure duration. This is because an opportunity to
create a more distinctive traceable memory is presented, allowing the memory
system to attain more key components and detail from the stimulus, creating a more
memorable object representation. Future research should investigate the possible
activation of conscious memory aiding implicit memory and which factors such as
exposure duration create this effect in the human memory system.

28
References
Bar, M., & Biederman, I. (1998). Subliminal visual priming. Psychological Science,
9(6), 464-468.
Bar, M., & Biederman, I. (1999). Localizing the cortical region mediating visual
awareness of object identity. Proceedings of the National Academy of Sciences,
96(4), 1790-1793.
Barbot, A., & Kouider, S. (2012). Longer is not better: nonconscious overstimulation
reverses priming influences under interocular suppression. Attention, Perception, &
Psychophysics, 74(1), 174-184.
Berntsen, D. (2010). The unbidden past: involuntary auto-biographical memories as a
basic mode of remembering.Curr.Dir.Psychol.Sci.19: 138–142.
Berry, C.J., Kessels, R.P.C., Wester, A.J., & Shanks, D.R. (2014). A single-system
model predicts recognition memory and repetition priming in amnesia. The Journal of
Neuroscience, 34(33), 10963-10974.
Crabb, B. T., & Dark, V. J. (1999). Perceptual implicit memory requires attentional
encoding. Memory & Cognition, 27(2), 267-275.
Crabb, B. T., & Dark, V. J. (2003). Perceptual implicit memory relies on intentional,
load-sensitive processing at encoding. Memory & cognition, 31(7), 997-1008.
Debner, J. & Jacoby, L. (1994). Unconscious perception: Attention, awareness, and
control. Journal of Experimental Psychology: lzarning, Memory and Cognition,
20,304-317.
Dew, I. T., & Cabeza, R. (2011). The porous boundaries between explicit and implicit
memory: behavioural and neural evidence. Annals of the New York Academy of
Sciences, 1224(1), 174-190.
Dunn, J. C., & Kirsner, K. (1989). Implicit memory: Task or process. Implicit memory:
Theoretical issues, 17-31.
Faivre, N., & Kouider, S. (2011). Increased sensory evidence reverses nonconscious
priming during crowding. Journal of vision, 11(13), 16.
Greene, R. L. (1986). Word stems as cues in recall and completion tasks. The
Quarterly Journal of Experimental Psychology, 38(4), 663-673.
Henke, K. (2010). A model for memory systems based on processing modes rather
than consciousness. Nature Reviews Neuroscience, 11(7), 523-532.
Richardson-Klavehn, A. et al. (1994). Intention and aware-ness in perceptual
identification priming. Memory. Cognitive Psychology. 22:293–312.

29
Knight, D. C., Waters, N. S., & Bandettini, P. A. (2009). Neural substrates of explicit
and implicit fear memory. Neuroimage, 45(1), 208-214
Knowlton, B. J., & Squire, L. R. (1995). Remembering and knowing: two different
expressions of declarative memory. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 21(3), 699.
Mandler, G. (1985). Cognitive psychology: An essay in cognitive science. Hillsdale,
NJ: Lawrence Erlbaum Associates.
Nelson, C. A. (1995). The ontogeny of human memory: A cognitive neuroscience
perspective. Developmental psychology, 31(5), 723.
Nelson, C. A. (1997). The neurobiological basis of early memory development. The
development of memory in childhood, 41-82.
Nelson, C. A., Wewerka, S., Thomas, K. M., deRegnier, R. A., Tribbey-Walbridge, S.,
& Georgieff, M. (2000). Neurocognitive sequelae of infants of diabetic mothers.
Behavioural neuroscience, 114(5), 950.
Perfetti, C. A., & Bell, L. (1991). Phonemic activation during the first 40 ms of word
identification: Evidence from backward masking and priming. Journal of Memory and
Language, 30(4), 473-485.
Serafine, M. L., Crowder, R. G., & Repp, B. H. (1984). Integration of melody and text
in memory for songs. Cognition, 16(3), 285-303.
Squire, L. R., & Knowlton, B. J. (1995). Memory, hippocampus, and brain systems.
Squire, L. R. (2009). Memory and brain systems: 1969–2009. The Journal of
Neuroscience, 29(41), 12711-12716.
Tamura H, Tanaka K (2001) Visual response properties of cells in the ventral and
dorsal parts of the macaque inferotemporal cortex. Cereb Cortex 11:384--399.
Tulving E, Schacter DL (1990) Priming and human memory systems.
Science 247:301--306.
Voss, J. L., & Gonsalves, B. D. (2010). Time to go our separate ways: opposite
effects of study duration on priming and recognition reveal distinct neural substrates.
Frontiers in human neuroscience, 4.
Wiggs, C. L., & Martin, A. (1998). Properties and mechanisms of perceptual priming.
Current opinion in neurobiology, 8(2), 227-233.
Zago, L., Fenske, M. J., Aminoff, E., & Bar, M. (2005). The rise and fall of priming:
how visual exposure shapes cortical representations of objects. Cerebral
Cortex, 15(11), 1655-1665.

30
Appendix
Fig.1
Displaying the opening screen to the study that participants witnessed. Here, they
were instructed by the computer, what to expect from the study and what each trial in
the research was like.
Fig.2
This is the option screen participants in the priming group are faced with. Here they
are required to make a judgement on whether the image on the monitor is natural or
manufactured. They can make the judgement by pressing ‘Q’ and ‘P’ for natural and
manufactured objects respectively.

31
Fig.3
This is the screen that the participants in the recognition group were presented with
on the test phase of the study. Here they were required to provide us with a choice of
confidence ratings. The choices as seen below were; high confidence or low
confidence that the image had been seen before and again, high confidence and low
confidence that the image had not been seen before. Accuracy is the primary target
here.
Fig. 4 this table shows the Means and Std. Deviation scores produced by the
statistical analysis (repeated measures ANOVA) on the implicit memory priming task.
Descriptive Statistics
Mean Std. Deviation N
dur40ms 37.8000 39.17896 10
dur150ms 29.2200 28.76069 10
dur250ms 31.1800 32.62061 10
dur350ms 46.1000 38.59370 10
dur500ms 22.3400 25.66339 10
dur1900ms 43.0300 39.04516 10

32
Fig.5 this table shows the Means and Std. Deviation scores produced by the
statistical analysis (repeated measures ANOVA) on the explicit memory recognition
task.
Descriptive Statistics
Mean Std. Deviation N
dur40ms .3080 .20004 10
dur150ms .3360 .25444 10
dur250ms .4270 .24864 10
dur350ms .4080 .22905 10
dur500ms .4240 .25605 10
dur1900ms .5220 .28809 10
Fig.6 this table is a product of the repeated measures ANOVA; it displays
significance levels for the between-subject effects for implicit memory.
Tests of Between-Subjects Effects
Measure: MEASURE_1
Transformed Variable: Average
Source Type III Sum of
Squares
df Mean Square F Sig.
Intercept 73269.181 1 73269.181 12.945 .006
Error 50939.294 9 5659.922
Fig.7 This table is a product of the repeated measures ANOVA; it displays the
significance levels for the between-subject effects for explicit memory.
Tests of Between-Subjects Effects
Measure: MEASURE_1
Transformed Variable: Average
Squares
Intercept 9.801 1 9.801 33.551 .000
Error 2.629 9 .292

33
Fig.8 This table again was produced by the repeated measures ANOVA test and
displays to the significance levels of the implicit memory priming task, using the
Sphericity Assumed statistical analysis option.
Fig.9 This table again was produced by the repeated measures ANOVA test and
displays to the significance levels of the implicit memory priming task, using the
Sphericity Assumed statistical analysis option.
Tests of Within-Subjects Effects
Measure: MEASURE_1
Squares
exposureduration
Sphericity Assumed .287 5 .057 3.804 .006
Greenhouse-Geisser .287 3.824 .075 3.804 .012
Huynh-Feldt .287 5.000 .057 3.804 .006
Lower-bound .287 1.000 .287 3.804 .083
Error(exposureduration)
Sphericity Assumed .679 45 .015
Greenhouse-Geisser .679 34.418 .020
Huynh-Feldt .679 45.000 .015
Lower-bound .679 9.000 .075
Tests of Within-Subjects Effects
Measure: MEASURE_1
Squares
exposureduration
Sphericity Assumed 4037.891 5 807.578 2.806 .027
Greenhouse-Geisser 4037.891 2.154 1874.727 2.806 .082
Huynh-Feldt 4037.891 2.854 1414.843 2.806 .062
Lower-bound 4037.891 1.000 4037.891 2.806 .128
Error(exposureduration)
Sphericity Assumed 12950.624 45 287.792
Greenhouse-Geisser 12950.624 19.385 668.085
Huynh-Feldt 12950.624 25.686 504.199
Lower-bound 12950.624 9.000 1438.958

34
Fig.10 This is a graph displaying the difference in mean RT for new and old items for
varying exposure durations. This was completed by subtracting the mean RT for new
items away from the mean RT for old items.

35
Fig.11 This is a graph displaying the difference in correctly identifying the presented
stimulus as new or old against incorrectly labelling the stimulus new or old, for
varying exposure durations. This was completed by subtracting the “hit rate” away
from the “false alarm rate”.

36
Fig.12 The brief participants were presented with prior to the study commencing.
The University of Plymouth Faculty of Health and
Human Sciences
Brief
 In this study you will be presented with a series of images one after the other on a
computer screen in front of you. All experimental instructions will be clearly provided
on the screen. You will be given 5 practise trials for every test phase encountered. If
you have any questions about the research do not hesitate to ask the experimenter.
 The experiment should last around 30 minutes.
 You will be provided with an informed consent sheet if you choose to agree to
participate in the research.
 Contact details are provided should you require any answers about the research in
the future.
Contact Details:
Researcher: Matthew D Jones
Phone Number: 07801922320
Email: matthew.d.jones@plymouth.ac.uk
Supervisor: Christopher Berry
Supervisor Email: Christopher.berry@plymouth.ac.uk

37
Fig.13 Debrief participants were given post the studies completing, this was used to
inform the participants of the actual aim of the research and why it was a necessary
piece of research.
The University of Plymouth Faculty of Health and Human Sciences
Debrief
This study was an investigation into the behavioural differences between explicit and
implicit memory and the effect that exposure duration to a stimulus has upon the
performance of both of types of memory. The research undertaken was an
adaptation of Zago et al. (2005) research. The aim of the experiment was to
understand the difference nature of performance for both implicit and explicit
memory. The route of memory in the brain and its nature is a highly disputed area of
psychology, this research attempts to provide substantial findings that can support
either a single multi-functioning system, or two separate functioning systems entirely.
In the experiment, you the participant, were exposed to a series of images each
presented differently depending on which exposure duration condition you were
assigned too (40, 150, 250, 350, 500 &1900 Ms.) For the first part of the study you
were required to recall as many images as you possibly could from the series
presented previously. This is specifically testing your explicit memory system by
recognition. For the implicit memory test, priming was administered. Another series of
images were presented during the study phase and then on the test phase, the
participant is required to select “yes” or “no” to the question, “has this image been
seen before?” This is a test of your implicit (unconscious) memory. The purpose of
differing stimulus durations for each task on both explicit and implicit memory was to
examine the difference in performance for each over the same stagnated duration
blocks. Any significant change in performance from either from the other, would
suggest that they are two differing systems entirely and are not originated from the
same multi-functioning source in the brain. It is hypothesized in the research that
likes Zago et al. (2005), priming performance increases with exposure duration up
until the threshold of around 250 Ms, soon after and beyond this, performance will
begin to decline. As for the explicit memory task of recognition, it is expected that the
longer the stimulus duration, the increase in recognition performance. If both are
apparent in the results of this research, it does suggest that there are two distinct
functioning mechanisms in the brain in charge of implicit and explicit memory
respectively.
Please contact the researcher Matthew Jones at
matthew.d.jones@students.plymouth.ac.uk if you have any queries about the
research in any capacity.
Contact Details
Researcher: Matthew Jones
Telephone number 07801922320.
Supervisor: Dr Christopher Berry chris.berry@plymouth.ac.uk

38
Fig. 14 the consent form participants were given upon arrival to the study. Signature
gives the experimenter the participant’s permission to take part in the research and
collect the data.
RESEARCH INFORMED CONSENT FORM
Title of Project: Simple picture judgement task, manufactured or natural?
Investigator(s): Matthew David Jones
Researcher Email: matthew.d.jones@students.plymouth.ac.uk
Please read the following statements and, if you agree, sign on the
corresponding lines to confirm agreement:
I confirm that I have read and understand the information sheet for the above study.
I have had the opportunity to consider the information, ask questions and have had
these answered satisfactorily.
I understand that my participation is voluntary and that I am free to withdraw at any
time without giving any reason.
I understand that my data will be treated confidentially and any publication resulting
from this work will report only data that does not identify me.
I freely agree to participate in this study.
Signature: …………………………………………….
Name of participant (block capitals)
Signature: …………………………………………….
Date
…………………………………………….
Researcher (block capitals)
Signature: …………………………………………….
Date
…………………………………………….

Dissertation Implicit Explicit Memory

More Related Content

Similar to Dissertation Implicit Explicit Memory (20)

Dissertation Implicit Explicit Memory