REPEATED PRESENTATION OF SUBOPTIMAL STIMULI
AND SUBSEQUENT "AFFECTIVE" RESPONSES

by
Michael Hoff
B.A., University ofVictoria, 1995

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
m

PSYCHOLOGY

©Michael Hoff, 1999
THE UNIVERSITY OF NORTHERN BRITISH COLUMBIA
July, 1999

All rights reserved. This work may not be
reproduced in whole or in part, by photocopy or
other means, without the permission of the author.

UNIVERSITY OF NORTHERN
BRITISH COLUMBIA

UBRARY

Prince George, BC

Repeated Presentation

n

ABSTRACT

A priming paradigm was employed to examine the veridicality of an independent affective and
cognitive processing system (Zajonc 1980). This assertion was tested by comparing participants
physiological reactions to neutral and affective priming stimuli. Cardiovascular reactivity was
recorded while participants (N=36) watched a computer monitor that presented stimuli of brief
duration of exposure (suboptimal). At suboptimal exposures, only affective primes produced
significant shifts in the participants physiological activity. One interpretation of these results is
that participants' manifested an affective reaction to emotional primes without the benefit of
conscious recognition (Kunst-Wilson & Zajonc 1980). The results are interpreted in light of
neurological (Le Doux 1986) and behavioral evidence (Roediger 1990) to suggest support of
interdependence of affect and cognition. This study showed that a physiological-affective
priming paradigm has utility for examining the interdependence of affect and cognition.

Repeated Presentation

m

TABLE OF CONTENTS

Abstract ......................................................................................................................................... n
Table of Contents .......................................................................................................................... iii
List of Tables................................................................................................................................ v
List of Figures ............................................................................................................................... vi
CHAPTER ONE
Introduction ................................................................................................................................... 1
Mere Exposure Effect ...... ................................................................................................. 1
Nonconscious Affective Priming ...................................................................................... 3
Physiological Evidences. ..................................................... ............ .............................. .... 4
The Independent Affective-Cognitive Processing Debate ... .. ............. ......... .. ..... .......... .. .. 6
Overview of Present Study ............................................................................................... 7
CHAPTER TWO
Method .......................................................................................................................................... 9
Participants ............................. ................................. ......................................................... 9
Apparatus and Materials .......... ..... .................................................................................... 9
Procedure .......................................................................................................................... 11
CHAPTER THREE
Results ........................................................................................................................................... 17
CHAPTER FOUR
Discussion ............................................................................................................... :; .................... 28

Repeated Presentation

iv

Visceral Response and Affective States ........................................................................... 28
Neuroanatomical Evidence ............................................................................................... 29
Behavioral Evidence ......................................................................................................... 36
Null Findings and Explanations ........................................................................................ 40
Conclusion ................. ...... ................................................................................................. 41
REFERENCES ..............................................................................................................................43
APPENDICES
A

Source monitoring test. ..................................................................................................... 50
B

Differential emotional scale ............................................................................................ 54

Repeated Presentation

v

LIST OF TABLES
Table
1

Mean change in participants
reported affective states (DES) ............................................................................. 27

Repeated Presentation
Figure

vi

LIST OF FIGURES

1

Stimulus presentation for phases 1-4 .................................................................... .12

2

Presentation of optimal and suboptimal
stimuli across a four stage procedure ................................................................... 16

3

Differential physiological registration
to stimuli ............................................................................................................... 21

4

Physiological responses across the target
and neutral conditions among the three groups .................................................... 22

5

Main effect for within subject condition of
neutral expression) vs. target (anger, disgust
and happy) ............................................................................................................. 23

6

Interaction for free recall of affective primes
(neutral and target) across the between
subject conditions (anger, disgust and happy) ..................................................... 24

7

Physiological responses across the within
subject conditions (target and neutral)
by group ................................................................................................................ 25

8

Main effect for within subject condition of
neutral (neutral expression) vs. target
(anger and disgust) ................................................................................................ 26

9

Amygdala projections to and from
cortical systems ..................................................................................................... 3 5

Repeated Presentation

vn

ACKNOWLEDGMENTS
I am deeply indebted to Dr Glenda Prkachin for allowing me to explore ideas without
prejudice and providing invaluable insight along every stage of this thesis. Thanks to Ken
Prkachin, a man who though always on a mission has time for the little people.
I would also like to thank those wild wacky people who volunteered there time and
effort during the experimental stages of this thesis. Primarily, I would like to thank Jill
Demontigny, Megan Piket, and Michelle Smith for slapping on electrodes, and for doing the
things that only they could do.
Most of all I would like to thank my wife for putting up with my sorry ass these past few
years, if possible, I would of left me long ago.

Repeated Presentation

1

Chapter One
Repeated presentation of suboptimal stimuli and subsequent affective responses
The affective priming hypothesis, asserts that affective reactions can be elicited with
minimal stimulus input and "virtually" no cognitive processing (Zajonc 1980). This hypothesis
challenges the notion that cognitive and affective processes are neurologically interdependent (Le
Doux 1992a). This thesis describes research that examines the legitimacy of a claim of
independent cognitive and affective processing. It also shows how various stimuli can elicit
physiological reactions that may or may not be interpreted as support for independent cognitive
and affective processing system.
To set the stage for the present study the following three avenues of investigation are
reviewed, 1) the mere exposure effect (Zajonc 1968), 2) the non-conscious affective priming
effect (Kunst-Wilson & Zajonc; Zajonc 1984; Zajonc 1980), and 3) physiological measures
(Tranel & Damasio 1988; Zajonc 1968).
Mere Exposure Effect
. With the publication ofZajonc's (1968) paper" the attitudinal effects of mere exposure"
the mere exposure effect became a topic of conversation in mainstream psychology.

Zajonc

defined the exposure effect as the observation that " mere repeated exposure of the individual to
a stimulus is a sufficient condition for the enhancement of his attitude toward it. By 'mere
exposure, is meant a condition which just makes the given stimulus accessible to the individual's
perception" (p.l ). Thus, Zajonc suggested that simple unreinforced repeated exposures lead to
liking for a stimulus. In a series of subsequent experiments, Zajonc then went on to demonstrate
the power of the mere exposure effect for affective experiences.

Repeated Presentation

2

In Zajonc's (1968) first experiment, participants were exposed to series of nonsense
words at frequencies ranging from 0 to 25. They then rated each stimulus word for "goodness"
of"meaning" (i.e. the extent to which the word connotes "good" vs."bad" affect) on a sevenpoint scale. Zajonc found a positive relationship between number of exposures and the average
goodness rating for a word. He then replicated this experiment using a similar procedure but
different stimuli: Chinese ideographs were substituted for the nonsense words used in
Experiment 1. The results of Experiment 2 were consistent with the findings of Zaj one's first
study; again, rated goodness of meaning was positively related to frequency of exposure.
Zajonc's (1968) third experiment investigated the extent to which typical exposure effects
could be obtained with socially relevant stimuli. In this experiment, subjects were shown a series
of faces (photographs of students taken from a college year book) at different exposure
frequencies, after which they were asked to make liking ratings of each stimulus person on a
seven point scale. A significant, positive relationship between frequency of exposure (25
presentations was the maximum number of presentations in this study) and mean liking rating of
the stimuli was found.
Stimuli presentation for the above experiments occurred at optimal levels of exposure,
meaning the participants could clearly see the stimulus and therefore were aware of it. However,
a related phenomenon, the affective priming effect, has consistently been found even when the
participant has claimed that he or she was unaware that a stimulus had been presented. Stimuli in
these experiments are presented at suboptimal levels of exposure. That is, the stimulus is either
presented at a level so degraded that the participant cannot recognize the stimulus or,
alternatively, presented so fast that he or she cannot see the stimulus. Manipulating the image,

Repeated Presentation

3

size and focus of a stimulus constitutes suboptimal presentation of a stimulus in a degraded state
(Kunst-Wilson & Zajonc 1980). The standard duration for suboptimal presentation of a stimulus
is anywhere from four to five milliseconds (Kunst-Wilson & Zajonc 1980).
Nonconscious Affective Priming
Change in preferences for stimuli as result of repeated suboptimal exposures to those
stimuli is known as the nonconscious affective primacy effect.
One example of the nonconscious affective primacy effect is found in research performed
by Kunst-Wilson and Zajonc (1980). In this experiment, similar to the mere exposure
experiments (1968), stimuli were presented repeatedly. However, instead of presenting the
stimuli at optimal levels of exposure, stimuli were presented at varying levels of degradation by
manipulation of the images' size and focus. The clear finding was a direct increase in
participants' preference for an object after repeated exposures to that object. Interestingly, the
effect was also obtained even when the exposures were so degraded that the person was not aware
that anything at all had been presented. They were unable to describe why they did or did not like
the stimuli they were previously exposed to. That familiarity does not appear to be involved in
these effects is shown by the fact that liking scores depended almost entirely on the person' s
objective experiences with the object rather than on their perceptions of familiarity with the object
(Moreland & Zajonc 1977; Matlin 1971). It is also of some interest that Bomstein (1987)
reported stronger exposure effects with suboptimal presentations than with optimal presentations.
The clearest instance of non-conscious affective priming is found in experiments in which
stimuli presentation occurs with short duration. In this type of experiment (Zajonc 1980), a
photograph of a smiling face is presented at very short intervals, say 4 milliseconds, just before

Repeated Presentation

4

another neutral and unrelated stimulus (a Chinese ideograph) is shown for 1 second. Results
consistently show that participants' stated preference for the neutral stimulus is influenced by
whether or not that stimulus is preceded by a smile or an angry expression. It is important to
note that this same procedure involving the optimal presentation of the same affective prime
produces no result (Kitayama 1991 ; Niedenthal1990).
One way of interpreting these results is to allow for the possibility that total affective
discriminations can be made virtually without awareness, whereas cognitive discriminations
require greater access to stimulus information (Kunst- Wilson & Zajonc 1980; Murphy 1990;
Zajonc 1980). Indeed, the affective primacy hypothesis (and its assumption of independent
affective and cognitive processing) hinges on the conjecture that the simple affective qualities of
stimuli, such as good versus bad or positive versus negative, can be processed more readily than
their non-affective attributes.
Physiological Evidence
Physiological evidence is often provided (Murphy 1990; Zajonc 1980) as converging
support for the assumption that cognitive and affective processing systems are independent.
Zajonc (1968), for example tested the hypothesis that repeated presentation of a word would lead
to a decrease in galvanic skin response (GSR) fluctuations that result from stimulus exposures.
Fifteen nonsense words were presented at optimal levels of exposure with a frequency between
one to twenty five times. Zajonc found a negative relationship between exposure frequency and
mean GSR change in response to the final stimulus presentation, suggesting that repeated
unreinforced exposure to a word result in a decreased autonomic arousal following later stimulus
presentations.

Repeated Presentation

5

Supplementary research is provided by Tranel and Damasio (1988). In a study of four
patients with face agnosia, Tranel and Damasio focused on the use of GSR and self-report
measures in response to familiar and unfamiliar faces. Results showed a strong dissociation
between the indices. Participants generated more frequent and significantly larger amplitude
skin conduction responses to familiar faces than to unfamiliar ones but were unable to give
discriminatory verbal ratings to familiar faces versus unfamiliar faces. In other words, skin
conductance revealed that face agnostic patients responded to familiar faces although according
to their self-reports they did not appear to perceive them. In a subsequent study Tranel,
Damasio, and Damasio (1995) tested whether nine patients with ventromedial frontal damage
could discriminate familiar faces (family) they had exposure to prior to brain damage from
familiar faces (psychologists) they had exposure to following the damage. Patients were also
asked to rate each face for familiarity. Participants with bilateral frontal damage recognized the
identity of familiar faces, yet failed to generate discriminatory skin conductance to those same
familiar faces. The findings showed electro-dermal activity to facial stimuli that patients could
not recognize and for which a sense of familiarity was non-existent.
Zajonc and colleagues (Kunst- Wilson & Zajonc 1980; Murphy 1990; Zajonc 1980,1984)
have suggested that physiological responses comparable to those found in the above studies can
be equated with emotion. This assumption, that physiological responses are synonymous with
affect, will be explored later in this paper.
In sum, collectively, mere exposure, affective priming, and physiological evidence are
cited (Kitayama 1991 ; Kunst-Wilson & Zajonc 1980; Murphy 1990; Niedenthal1990; Zajonc
1980; Zola-Morgan, Sqiure, Avarez-Royo, & Clower 1991) as support for separate cognitive and

Repeated Presentation

6

affective processing systems. However this evidence alone is insufficient to convince all
theorists that an independent cognitive-affective processing system exists.
The Independent Affective-Cognitive Processing Debate
For theorists advocating a non-conscious process of emotion, emotion is often separated
from cognition. Zajonc (1984) argued that affective influences should resist attribution
interventions because the affective system responsible for preferences is separate from the
cognitive system responsible for inferences. He suggested that early affective processes are
automatic and therefore inaccessible to higher-order interventions. As a result, preliminary
affective responses are not represented as conscious feelings but are diffuse and can "spill over"
from one stimulus to another. In contrast non-conscious cognition is always context specific and
must be about something.
Theorists such as, Ledoux (1989, 1986), and Shwarz and Clore (1987) who argue that
non-conscious states of emotion are really subjective states of awareness oppose the idea of a
separate affective system. In other words emotions are conscious states. In their
feelings-as-information model Schwarz and Clore (1987) declare that judgments are based on
perceptible feeling. Feelings are the central component of emotion and feelings are by definition,
conscious. Like Schwarz and Clore, Ledoux (1986, 1989) asserts that non-conscious emotions
do not exist. However, Ledoux does indicate that conscious emotional states are products of
unconscious processes. He further suggests that processes that are themselves not permeable to
consciousness are responsible for separating the substances of consciousness. Given the
controversy surrounding the nature of non-conscious affect it is apparent that the question of
non-conscious affect is as yet unresolved.

Repeated Presentation

7

Overview of Present Study
The present thesis describes an experiment that further explores the question of nonconscious affect. The major objective of this experiment is to determine if awareness of our
mental or implicit representation is necessary to produce physiological responses and recognition
of those expressions. By including a physiological measure for detection of non-conscious
affect, it was possible to circumvent the traditional method of asking for stimulus preference
(Zajonc 1980) and instead assess directly implicit responses to stimuli. Additionally, the present
study addresses whether facial affect type may produce differing levels of physiological arousal.
Exploring this was made feasible through the use of physiological measures, in this instance,
hemodynamic activity assessed by impedance cardiography (Sherwood, Allen, Fahrenberg,
Kelsey, Lovallo, & Van Doomen 1990).
Another objective was to determine if repeated suboptimal presentation of facial

.

expressions of emotion would result in increasing the threshold for detection of that emotion. In
addition, it is quite possible that some facial expressions (e.g. anger)will result in greater
physiological arousal than other facial expressions (e.g. disgust). Use of repeated presentation
and impedance cardiography would help facilitate detection of participant's differential response
to facial expressions.
In sum, the study examined the following predictions: (1) IfZajonc's (1984) affective
priming hypothesis has merit then there will be a dissociation between physiological and
conscious reports. Subjects exposed to the sub-optimal presentation of emotional stimuli will be
unable to consciously report those stimuli. However, they will generate more frequent and
significantly changed heart rate responses to facial expressions of emotion, than too neutral facial

Repeated Presentation

8

expressions. (2) Some emotions will be easier to label than others. Happiness and anger for
example, will be easier to label than disgust (Prkachin & Prkachin 1994). (3) Some emotions will
be easily confounded. In particular, it is predicted that fear is likely to be mistaken for surprise
and disgust for anger (Prkachin in preparation). (4) Finally, some emotions will be easier to
detect than others at sub-optimal levels of exposure. Happy expressions, for instance, may have
a higher probability ofbeing detected (Ladavas, Umilta, Ricci-Bitti 1980) at sub-optimal
exposures than any other emotion. However, detectability does not necessarily translate into
greater physiological response. What expression will result in the greatest physiological
response is also a question of this research, although no explicit predictions can be made from
the available literature.

Repeated Presentation

9

CHAPTER TWO
Method
Participants
Thirty-six introductory psychology students (18 men and 18 women) participated in the
following experiment for supplementary course credit. Participants were equally distributed
across three between group conditions (12 participants per condition). The between group
variable was the affective priming conditions of anger, disgust and happiness. A within group
condition included exposure to optimal and suboptimal primes.
Apparatus and Materials
Using conductance electrodes and a BoMed Cardiodynamic Data Processing System the
impedance ECG derived signals of heart rate, were saved to disk for later analysis of heart rate
variability. Electrode placement was preceded with the washing of upper and lower thoracic
regions with isopropyl rubbing alcohol. Redeux, a form of saline paste was then applied to
facilitate conduction. Neck electrodes were attached frrst, with one half of a dual electrode
attached at the intersection at the base of the neck and the other half of the electrode attached
directly above. This procedure was repeated for the other side of the neck. A similar procedure
followed for attachment of lower thoracic electrodes. Electrodes were attached bilaterally to
either side of the thorax with the upper half of a dual electrode attached parallel to the xiphoid
process and the other half attached just below.
For assessment purposes, participants completed a modified version of Lindsay &
Johnson's (1989) source monitoring test (described more completely in procedure section) and
Izard's (1972) Differential Emotion Scale (DES). The source-monitoring test assesses

Repeated Presentation

10

participants' ability to accurately memorize the representation of an event (see Appendix A). The
DES, consists of a series of affect words (e.g., elated, tense) presented along with a four point
Iikert scale, which participants used to self rate the emotions they were experiencing following
stimulus presentation. At the end of each DES, a question was included asking participants to
recall the stimuli they were exposed to in the preceding phase (see Appendix B).
Participants were seated in a comfortable chair approximately 50 em away from a 1 color
TRL computer monitor. Mel 2 computer programming language and a 133 MHz Pentium
processor facilitated presentation of stimulus photographs. Photographs advanced automatically
according to preprogrammed modules designed with Mel 2 software. In total, participants
viewed ninety presentations depicting facial expressions chosen from Ekman and Friesen's
(1976) Pictures of Facial Affect. Photographs included one male and one female face expressing
the emotions, sadness, happiness, anger, surprise, disgust, fear and the expressionless face of
neutral. The expressions happiness, disgust and anger were selected as affective primes during
one of two suboptimal phases. By "suboptimal" I mean that the stimuli were presented for an
abbreviated time thus preventing their complete processing. The expression 'neutral' functioned
as a control stimulus during one of two suboptimal phases. Stimulus faces were 175 X 225
pixels X 75 resolution.
In phase one, participants viewed one face (either male or female) and 4 of 7 possible
expressions at optimal level (10,000 milliseconds for each face). By "optimal" I mean that the
stimuli were presented for a sufficient duration to be processed fully. They viewed each of these
expressions 5 times for a total of 20 presentations. In phase two, a "suboptimal" eondition, they
viewed the same person as in phase 1 depicting one expression, not seen in phase 1, 25 times at 5

r - - - - - - - - - - - - - - - - - - - - - - - - ---------

Repeated Presentation

11

milliseconds per face. In phase three, another suboptimal condition, participants viewed the
same face as in phase 1 and 2, and one expression not seen in phases 1 and 2 the same number of
times and duration as in phase 2. In the fourth and final phase, an "optimal" condition,
participants viewed the same face and expressions as in previous phases plus the additional
expression of fear, which functioned as a foil. Each of the expressions was viewed three times
for a total of 21 presentations. Expressions were presented for 14,000 milliseconds each,
allowing participants time to provide verbal answers for the source monitoring test. See Figure 1
where each of the phases is presented graphically. The expressions are presented in cartoon
form.

Repeated Presentation

Sadness

Anger

Surprise

12

Disgust

xs

Phase 1

10,000 ms

10,000 ms

10,000 ms

10,000 ms

I time=200,000ms
Total

~~imal~~~~~~~~~~~~~~~~~~~~~~

Exposure

Happiness
Phase 2
Sub-optimall--~~~--.-~~~~.1...--~~~-.-~~----1

Exposure

X25

Sms

Total
time=125ms

Neutral
Phase 3
Sub-optimal J--~~~--.-~~~~......__~~~--r~~-----1
Exposure

X25

Sms

Total
time=125ms

Fear=foil
Phase 4
~ptimal

Exposure
14000ms
Figure 1
Stimulus presentation for phases 1-4

Total
time=294, OOOms

Repeated Presentation

13

Procedure
Informed consent was obtained from each participant prior to the experimental session.
Following agreement to participate, the impedance electrodes were attached. Participants were
then instructed to rest quietly for an acclimatization period of 5 minutes.
The experiment consisted of four phases, an explicit exposure phase, a neutral phase, an
implicit exposure phase and a final test phase. These ran consecutively and participants
performed the same task in each. Participants were not told the distinctions between the four
phases. A five-minute rest period was included between phases one, two and three. Following
the completion of each phase, participants completed the DES. During the (ourth and fmal phase
participants answered the modified version of Lindsay & Johnson' s (1989) source monitoring
test.
Following completion of informed consent, participants were told that their task was to
observe 20 photos of facial expressions. Participants were advised that each facial expression
would be in view for a full 10 seconds and that they should look at it for the whole period as they
would be queried at the end of the task regarding the expressions they had seen. A fixation point
was projected for 2000 ms at the center of the screen, immediately prior to each of the facial
expressions. Participants were exposed to a random order of facial expressions. In total,
participants viewed four of seven possible facial expressions at optimal exposure.
For phases two and three, counterbalancing was performed by randomly assigning
participants to either a target priming condition in which participants were exposed 'to an
emotional facial expression (affective stimuli) or a neutral priming condition in which they were
exposed to a neutral expression (neutral stimuli). For brevity, explanation of phases two and

Repeated Presentation

14

three assumes exposure too neutral stimuli first followed by target (see figure 2), but the reader
should keep in mind that these phases were counterbalanced across participants.
The stimulus for the second phase of the experiment consisted of a neutral facial
expression. Neutral expressions are included as a control measure. These expressions were not
observed in phase one. Neutral expressions were presented to participants repeatedly and at suboptimal levels of exposure. Each neutral expression was presented for 5 milliseconds, followed
immediately by the blank backward mask. Stimuli were presented to the left visual field as
research ( Borod 1992; Davidson 1992; Gardner, Brownell, Wapner & Michelow 1983) indicates
that the right hemisphere is dominant with respect to processing of emotional responses. To
ensure that participants attended to the screen during sub-optimal exposure, a fixation point was
projected for 1000 ms at the center of the screen immediately prior to target presentation.
Participants viewed in total 25 presentations.
Stimuli for the third phase of the experiment were the target facial expressions of
emotion. Target facial expressions, were not observed by participants in phase 1. For example
participants exposed to the facial expressions of sadness, surprise, happiness, and disgust in
phase 1 viewed the anger expressions in phase 3. Target expressions were presented to
participants repeatedly and at sub-optimal exposures. Each target expression was presented for 5
ms, followed immediately by the backward mask. As in the second phase, the stimuli were
presented to the left visual field to maximize potential physiological reaction. To ensure that
participants attended to the screen during sub-optimal exposure a fixation point was projected
for 1000 ms at the center of the screen immediately prior to the target presentation. PartiCipants
viewed in total 25 presentations.

Repeated Presentation

15

In phase 4, participants were exposed to both the optimal (phase 1) expressions and the
sub-optimal (phase 2 and 3) facial expressions but they were all presented at optimal fourteensecond duration. Following presentation of each face participants were asked to identify
expression type. All participants were informed that each facial expression would be in view for
a full fourteen seconds and that they should look at it for the whole period, then make their
response, even if they recognized the expression before the period was concluded. A record was
kept of each participant's ability to identify the individual expressions of emotion. Participants
were also asked to the source of each expression (e.g. where and if they had previously seen the
expression) and to indicate how confident they were (on a scale of0-100) that they had seen the
expression where they said they had seen it. See Figures 1 & 2 for further clarification of
procedure.

~

-+'

•

~

•

A

L..-

DES

Phase 2 & 3 suboptima

16

Source monitoring test

J

(4optimal + 1 neutral+
1 target + 1 foil) x 3
14 seconds per
expression

_J__________t-----~------ Phase 4 optimal

•

r

1 expression x 25
5ms per expression

Presentation of optimal and suboptimal stimuli across a four stage procedure

Figure 2

L

~

1 expression x 25
5ms per expression

I
•
Phase 1 optimalj__ _

4 expressions x 5
1o seconds per
expresson

1

,...-- Counterbalancei neutral x target -

.....-------------,....-- 5 minute rest periods - - - - , - - - - - - - - - - - - - - - .

Repeated Presentation

Repeated Presentation

17

CHAPTER THREE
Results
The analyses of the results of this experiment are limited to the exploration of the
responses to the suboptimal stimuli. Investigation of the responses to the optimal stimuli is
unnecessary to address the hypotheses raised in the introduction. Differences in physiological
responses to optimal presentation of comparable stimuli have been explored by others (Bomstein
1987; Cabeza, Burton, & Kelly 1997; Schweinberger, Pfuetze, & Sommer 1995) and are not
relevant to issues addressed in this report.
In order to determine whether participants were affected physiologically during the
different stimulus conditions, it was necessary to select an index that would logically reflect
physiological perturbations produced by stimuli. It is well known that heart-rate decelerates
during exposure to stimuli that evoke attention (Schwartz 1971). Ordinarily such responses can
be evaluated by comparing heart rate levels before and after exposure to a given stimulus, and
heart rate change (deceleration) is taken as an indication of the "registration" of the stimulus.
Heart rate change is usually measured over an epoch of several seconds. In the present case,
stimuli were exceptionally brief and operated in rapid succession. Moreover, it was not possible
to link stimulus presentation to the heart-rate time series in order to determine precisely when
heart rate changes occurred. It was reasoned, however, that if the briefly presented stimuli were
being "registered" at the physiological level, then it would be expected that the participants
would show a series of decelerations and recoveries of heart rate over the phase of the
experiment. If stimuli were not being registered, or not to the same extent, then such a series of
decelerations and recoveries would not occur. Consequently, it would be possible to detect the

Repeated Presentation

18

registration of stimuli at the physiological level by measuring the variability of the heart rate
during the relevant phases of the experiment. See Figure 3 for an example of heart rate
variability. Given the duration of photo stimuli presentation (5ms) we were unable to assess
visceral activity for independent stimuli. Standard deviations were therefore extracted from
across the spectrum of participants' heart rate response for all physiological data sets.
In preliminary analyses heart rate variability was examined across gender and
counterbalance order. No differences were found for gender and counterbalance conditions.
Subsequent analyses therefore ignored subject gender and counterbalancing as a factor.
A 3 X 2 analysis of variance (ANOVA) was performed comparing physiological response
across affective priming conditions (neutral vs target) and the between subject exposure
conditions (anger vs disgust vs happy). The results (see Figure 4 & 5) revealed a significant
main effect for affective priming conditions, F (1, 33) = 18.61, p = .000. A paired t-test revealed
that when participants were exposed to target affective primes their heart rate variability was
significantly higher than when exposed to neutral target primes. The mean heart rate variability
for target primes was 1.68 heart rate SD units, in contrast with a mean of .34 for neutral primes,
t(35) = 4.36, p = .000. The between group difference in the preceding analysis was
nonsignificant. Participants' differential physiological response to target and neutral primes
provides tentative support for the notion that feature detection may be occurring without
conscious recognition. Explicit support for this assumption is possible by examining the
relationship between the affective priming conditions and the free recall of those conditions,
which was the purpose of the following analysis.

Repeated Presentation

19

A 3 X 2 ANOVA was performed comparing participant free recall of affective primes
(neutral and target) across between subject exposure conditions (happy vs. anger vs. disgust).
The results revealed a significant affective prime X group interaction, F (2, 33) = 20.92, p = .000.
Although no participant exposed to the suboptimal neutral condition reported being aware of the
primes, differential reports of awareness occurred depending on the target prime, participants
were exposed to, as evident in Figure 6. In order to treat the data as conservatively as possible, a
Scheffe pairwise comparison was selected for post hoc analysis. The Scheffe pairwise
comparison revealed that happy faces were consciously recollected significantly more than that
of either angry or disgusted faces. The mean recall for happy faces following suboptimal
presentation was 83%. In contrast the mean recall for anger and disgust faces were 17% and
.00% respectively. No group differences were found for recall of neutral expressions. As for the
affective prime condition only happy facial stimuli (M=. 83) were recalled significantly more
than that of neutral facial stimuli (M = .00), t (11) = 7.42, p = 000. The same pattern emerged for
participants given the source-monitoring test. There was an affective prime X emotion
interaction F (2, 33) = 16.41, p = .000 with only participants exposed to happy ~ pr

io

meeting the threshold of accurate designation t (11) = 4.91 , p=OOO. In short, participants exposed
to suboptimal exposures had some sense of conscious recollection but only for happy facial
photo stimuli.
Given participants' capacity for recalling happy face stimuli, a 2 X 2 ANOVA was
performed comparing physiological response across affective priming conditions (neutral vs.
target) and the between-subject exposure conditions of anger and disgust. This was done in order
to determine whether the results of the initial analyses might have been affected primarily by the

Repeated Presentation

20

processing of happy facial expressions. The results revealed a significant main effect for
affective priming conditions, F (1 , 22) = 15.90, p = .001. A paired t-test revealed that when
participants were exposed to target affective primes, their physiological response was
significantly higher than when exposed to neutral target primes. The mean physiological
response for target primes was 1.60 heart rate in beats per minute, in contrast with a mean of .10
for neutral primes, t (23) = 3.96, p = .001 (see Figures 7 & 8). Results for recall of neutral and
target primes (anger vs. disgust) replicated those reported previously for free recall. Moreover,
even though participants were informed of the presence of the degraded primes, they
nevertheless still maintained that they were not aware of them.
Finally, there were no significant differences in the participants' report of affective
mental states corresponding with anger and happiness, as measured on Izard's (1972) DES.
However, a trend was seen in the data, which supports the notion that observing other
individuals' emotional expressions could influence a person's own mood. Although the trend
was slightly higher for happy than for a neutral stimulus, this was not the case for anger or
disgust as compared to the neutral stimulus. The opposite pattern of results emerged when the
affective state of irritation was examined as reported with exposure to anger. Participants
exposed to happy expressions reported irritation and anger substantially less than participants
exposed to anger and disgust expressions. For the pattern of results see Table 1. It important to
note that the affective state disgust was not adequately represented on the DES scale. Therefore,
the emotion disgust was excluded from analysis and hence is not included in Table 1 as a factor
for participants' change in emotional response.

Repeated Presentation

A Stimuli being "registered' physiologically

Q)

r'-o
t::
ro
Q)

::c

A Stimuli not being registered physiologically

....
Q)

~

t::
ro
Q)

::c

Figure 3
Differential physiological registration of stimuli.

21

Repeated Presentation

..-..

~
co

22

2.0

c

1.5

1.0
Q)

en
c
0

c..

en

.5

&
m

u

0>
0
0

·u;

£'
~

0.0
Suboptimal
-.5

o~~

c

m

~

-1.0 - - - - _ _ _ , . - - - - - - - - - - . , . . . - - - - - - - - - . . - - - - - - - - - ' •
Anger
Disgust
Happiness

Between Subject Condition

Figure 4
Physiological r~ po

across the target and neutral conditions among the three groups.

NT

Repeated Presentation

Anger, Disgust, and Happy Facial Stimuli
(5ms repeated Presentations)

23

Neutral Facial Stimuli
(5ms repeated Presentations)

1.8

........
:::2!

a_

1.6

co

c:
0

-....

1.4

(f)

Q)

ro

1.2

t

ro

Q)

I

'-"

1.0

Q)

en

c:
0

c..
en

.8

Q)

0::::

ro

.6

(.)
C)

0
0

~
.c:

.4

a_

.2

ro
Q)
:::2!

0.0

c:

Target

Neutral
Suboptimal

Figure 5
Main effect for within subject condition of neutral (neutral expression) vs. target (anger, disgust,
and happy).

Repeated Presentation

24

..........

;:R

0

c

~

80

:::::!

C/')

ro
E
......
Q)

0

C/')

60

c

:::>
.........

rn
(.)

Q)

0:::

40

C/')

:::::!
:::::!

E

+='
C/)

c

ro

Suboptimal

20

Q)

DTARGET

~

Anger

Disgust

Happiness

Suboptimal Affective Priming Condition

Figure 6
Interaction for free recall of affective primes (neutral and target) across the three between subject
conditions.

Repeated Presentation

~

c..

25

2.0

Ill
0

(/)

.£

1.5

(].)

1§
t:

m 1.0

-

:c

Suboptimal

DTARG

Anger

Disgust
Between Subject Condition

Figure 7
Physiological responses across the within subject conditions (target and neutral) by group.

Repeated Presentation

Anger and Disgust Facial Stimuli

Neutral Facial Stimuli

(5 ms repeated presentations)

(5 ms repeated presentations)

26

.--...

~

a..

1.5

CD

.£
0

(J)

-e
<0

Q)

:r:
Q)

1.0

f/)

c:

0
0..

f/)

Q)

0:::

Target

Neutral
Suboptimal Affective Priming Condition

Figure 8
Main effect for within subject condition of neutral (neutral expression) vs. target (anger and
disgust)

Repeated presentations

27

Table 1

Mean change in participants' probability of reporting affective states (DES).
Group

conditions
Anger

Disgust

Happiness

Reported affect

Exposure
Happiness

Anger

Neutral

8%

8%

Anger

8%

10%

Neutral

13%

8%

Disgust

13%

9%

Neutral

11%

5%

Happy

14%

3%

Note. Affective states of anger, irritation, and mad were collapsed to form the single emotional
state of anger.

Repeated Presentation

28

CHAPTER4
Discussion
To summarize our experiment, suboptimal affective target primes (in the form of the facial
expressions of anger; happiness; disgust) presented for 5 ms to the right hemisphere, generated
significant shifts in participants' physiological responses compared to neutral expressions
presented for the same duration and at the same location. In addition, participants exposed to
anger and disgust facial expressions failed to designate and report those expressions on subsequent
source monitoring and free recall test measures. In contrast, participants exposed to happy
expressions accurately designated and reported the expression on subsequent source monitoring
and free recall tests. Finally, there were no significant differences in the participants' reports of
affective mental states during any phase ofthe study, as measured on Izard' s (1972) DES. A trend
was seen in the data, which supports the notion that observing other individuals' emotional
expressions could influence a person' s own mood. Irritation was often reported with exposure to
an anger expression and less often to happy faces.
The salient finding in this investigation, was that exposure to affective primes elicited more
variability in physiological responses than exposure to neutral stimuli. Those responses occurred
largely without participant awareness that an emotional expression had been presented (with the
exception of happiness). The most parsimonious explanation for the results is provided by the
affective priming hypothesis (Zajonc 1980).
The affective priming hypothesis asserts that positive and negative affective reactions can
be evoked with minimal stimulus input and "virtually" no cognitive processing. In short, cognition ·

Repeated Presentation

29

and affect are independent processes that perform discrete operations on the same stimulus
information.
The findings are consistent with independent cognition and affect model if we can accept
the validity of the following assertions: 1) visceral responses are equated with affective states and
only affective states, 2) neurological architecture supports an independent cognition and affect
paradigm, and 3) behavioral research supports an independent cognition and affect model. The
remainder of this paper will explore evidence both consistent and inconsistent with cognitiveaffective independence.
Visceral Response and Affective States
The relationship between visceral response and affective state is difficult to explain without
reference to the context in which it occurs. As this relationship is strongly associated
with neuroanatomy, the majority of the debate regarding this assertion will be addressed
concurrently with neuroanatomical evidence. Any remaining questions about this relationship will
be discussed in the conclusion.
Neuroanatomical Evidence
Support for the assumption that cognitive and affective processes are independent (Zajonc
1980, 1984) is found in the form of recent neuroanatornical discoveries. For example, separation
of affective processes on the one hand and recognition and categorization of faces on the other is
suggested in cases of prosopagnosia (PA). Many prosopagnosics are completely incapable of
making even the most basic categorizations of faces, such as race, age, and gender, (Pallis 1955),
although they retain their ability to make appropriate affective responses to distinct facial
expressions (Ellis 1986). In fact, PA patients who suffer bilateral cerebral lesions are characterized

Repeated Presentation

30

by their inability to recognize the faces ofthe persons with whom they are familiar. Interestingly,
several studies have demonstrated that PA patients display elevated skin conductance (GSR) when
presented with faces of persons they had previously known but could not recognize (Bauer 1984;
Tranel & Damasio 1985). Some theorists (Kunst-Wilson & Zajonc 1980) suggest that, as with the
mere exposure phenomenon (Zajonc 1968), prosopagnosics manifest a positive reaction to
familiarity without the benefit of conscious recollection. These theorists therefore equate the
physiological skin conductance response with affect.
Interpretation of the PA results is provided by Bauer ( 1984) who has proposed a model
involving at least two anatomically and functionally distinct pathways. He has concluded that the
prosopagnosics' bilateral lesions selectively impair the ventral visuolimbic pathway (implicated in
object recognition) while sparing the dorsal visuolimbic connections. These spared visuolimbic
connections allow for a preliminary or pre-attentive analysis of the emotional significance of the
visual stimulus. In other words, prosopagnosics seem to retain their preferences while losing their
ability to discriminate (Zajonc 1980).
For additional support, researchers (Kitayama 1991; Kunst-Wilson & Zajonc 1980; Murphy
1990; Niedenthal 1990) advocating the separation of affective and cognitive processes often cite
research performed by Zola-Morgan, Squire, Alvarez-Royo, and Clower (1991). These researchers ·
conducted tests of emotional reaction and memory function on four groups of monkeys: intact
monkeys, monkeys whose amygdala had been removed, monkeys whose hippocampus had been
removed (LeDeoux, 1987), and monkeys whose amygdala and hippocampus had been removed.
Monkeys with amygadalectomies performed well on memory tasks but lost their emotional reactions
to emotion-inducing stimuli. In contrast, damage to the hippocampal formation resulted in memory

Repeated Presentation

31

deficits while leaving the emotional processes intact. Monkeys with lesions in both the hippocampus
and the amygdala lost both their emotional reactivity and their ability to retain newly learned
discriminations.
There is a final line of converging neuroanatomical research cited as support for affectivecognitive independence (Zajonc, 1984, 1989). LeDoux and colleagues (Iwata, Ledoux, Meely,
Americ, & Reis 1986; Iwata, Chida, & LeDoux 1987; LeDoux, Iwata, Cicchetti, & Reis 1988)
found a direct pathway between the thalamus and the amygdala that is just one synapse long. The
direct access from the thalamus to amygdala allows the amygdala to respond faster to some stimuli
than to other. As the hippocampus is separated from the thalamus by several synapses, according
toLe Doux et al, the response in the amygdala can occur as much as 40 ms faster. This
neuroanatomical architecture would apparently allow us to like an object even without knowing
what that object is.
I would be remiss in not pointing out that the evidence as presented is anything but
definitive support for neurological separation of cognitive and affective states. The supposition
that the physiological skin conductance response can be equated with an affective state is
speculative at best. Any number of factors may influence the outcome of a physiological response.
In the present, for instance, it is possible to equate cardiographic perturbations with affect.
However, these responses may be as much cognitive as they are affective. Cognitive research, for
example, has repeatedly shown that cognitive processing can occur at a nonconscious level of
awareness (Johnson, Hashtroudi & Lindsay 1993; Loftus 1979a, 1979b, 1981; McClosky &
Zaragoza 1985a, 1993; Roediger 1990; Tulving & Schacter 1990). Similar to procedures in this
experiment, implicit memory (Roediger 1990), repetition priming (Tulving & Schacter 1990), and

Repeated Presentation

32

misinformation effect (Loftus 1978a) procedures demonstrate that participants can perform
cognitive tasks and not be aware on subsequent testing that they had performed that task. It is
likely, given this cognitive research, that both the present procedure and those previously
mentioned (e.g. skin conductance studies) rely heavily on participants' cognitive processes.
Furthermore, it is equally likely that this processing plays a role in the subsequent physiological
response. We might argue then for a cognitive-affective interaction in the production of
procedurally driven physiological responses.
Evidence for a cognitive-affective interaction is also found in the form of recent
neuroanatomical discoveries. Research (LeDoux 1986, 1987; Amaral, Price, Pitkanen, and
Carmichael 1992) has shown that the amygdala receives inputs from areas in the thalamus, the
association cortex, the perirhinal cortex, and the hippocampus. Animal studies have implicated
each of these inputs to the amygdala in different aspects of fear processing (LeDoux 1992a).
Thalamic to amygdala projections are important in processing the emotional significance of simple
sensory features as opposed to complex objects. The association cortex to amygdala connection is
important for processing the emotional significance involved in the recognition of objects. The
perirhinal cortexes to the hippocampus connections and subsequent cortex and amygdala
projections are important in the emotional processing of complex representations, such as the
memory for context (Kim & Fanselow 1992). See Figure 9 for amygdala projections to and from
cortical systems.
It is apparent that the cognitive prerequisites for fear processing can operate on several
levels and can be minimal, extensive or both. Le Doux (1996) suggests that thalamic sensory
processing areas are the "gateway" to the neocortex, where object representations are constructed

Repeated Presentation

33

from incoming sensory signals. Furthermore, thalamic areas are a "gateway" to the amygdala,
where emotional significance can be attached to sensory signals. Functionally then, the amygdala
can be activated by the thalamus at about the same time the cortex can be activated by the
thalamus. Consequently, object recognition and emotional representation can proceed in parallel.
This neuroanatomical architecture would apparently allow us to like or fear something without
knowing what it is. However, this means that the representations that activate the emotional
system can be based on incomplete information rather than on clear and complete perceptions.
This may account for a person having an emotional response without a cognitive characterization
for that response. For example, sense or moods that can accompany a complex sensory event such
as the change in season yet have no apparent evidential representation. In other words, one's mood
such as sadness could be a r~ po

to a single component (leaves falling) of a complex stimulus

event (the season), but the individual is not capable of identifying this relationship. This kind of
architecture may be particularly useful for initiating a response to a threatening situation. Of course
this kind of architecture may also lead to emotional errors. However, emotional responses that are
inappropriately initiated by the thalamic sensory inputs to the amygdala can be ~o i i

by

cortical inputs, which provide multiple levels of representation to activate the amygdala (Amaral et
al1992).
In addition, Amaral et al (1992) have noted that the amygdala projects back to neocortical
systems (see Figure 9). Processing in the amygdala can thus influence many cognitive processes
organized at the many cortical levels, and at subcortical levels which include the hippocampus.
Projections to the cortex from the amygdala may serve as pathways through which emotional
processing can influence cognitive processing (Rolls 1990). The amygdala also projects to the

Repeated Presentation

34

brainstem (Takeuchi, Mclean & Hopkins 1982) and the forebrain (Price & Amaral1981). Both the
brainstem level and the forebrain cortex play a role in cortical arousal.
In summarizing the cognitive-affective neurological relationship, there appears to be a
demonstrable two-way interaction between emotion and cognition (Le Doux 1989; Amaral et al
1992). Bi-directional connections appear to exist between the neocortex and the amygdala that
permit upstream emotion related input from the amygdala to modulate cortical activity and
downstream cognitive input from the cortex to modulate the amygdalas' emotional information
processing (Ledoux 1989).

Repeated Presentation

Perirhinal
Cortex

Association
cortex

Primary
Cortex

35

2
Hippocampus
Lateral
Sensory
Thalamus

1

Amygdala
Central

1:stimulus features
2:perceptual cortex
3:perirhinal cortex
4:contexts
(From Le Deux 1996)

Emotional
Behavior

Autonomic
Nervous
System

Figure 9
Amygdala projections to and from cortical systems.

NeuroEndocrine
System

Repeated Presentation

36

Behavioral Evidence
Behavioral research supporting affective-cognitive independence has largely been
generated by Zajonc's (1980) affective priming procedure, a procedure in which participants by
virtue of repeated exposures, develop affective preferences for previously novel Chinese
ideographs. In this procedure a priming stimulus with some emotional connotation such as a
picture of a smiling or angry face is presented briefly (4 milliseconds) and is then followed by a
masking stimulus. The masking stimulus displaces the prime from consciousness essentially by
blanking it out. Shortly afterward, a target stimulus (a Chinese ideograph) is presented for several
seconds (optimal exposure) and is consciously perceived. Later when given a forced choice
recognition test, participants are unable to distinquish the priming stimulus from a new stimulus
they had never seen. Yet despite this lack of overt recognition, when asked which of the two
ideographs they liked better, subjects consistently preferred the one primed by the happy face.
A number of criticisms have been leveled at experiments reporting priming effects obtained
for stimuli presented at suboptimal levels. For the most part, these critics doubt that there actually
is a total absence of conscious detection or identification (Bolender 1986; Purcell, Stewart, &
Stanovitch 1983). In his experiment, Zajonc (1984) used a forced choice test of awareness to
determine whether primes remained outside participants' conscious awareness. For this test,
participants were exposed to two faces, one the suboptimal prime (either a happy or an angry
expression), and the alternative, a foil that has never been seen before. Participants were then
asked which of the two faces was the prime. Typical results showed that participants were unable
to select the prime (happy or angry) from the incorrect alternative at a level greater than chance.
The results in the present study are inconsistent with those reported above. In the present

Repeated Presentation

37

study, participants were asked to freely recall stimulus presentations. As noted, participants
recalled the expression "happy" at a level significantly above that of chance, whereas recall for
angry and disgust faces was not above chance.
One possible explanation for differences in the recollection of affective primes corresponds
with the test procedure used by Zajonc (1984) compared to the test procedure used in the present
experiment. As indicated, Zajonc used a forced choice test of awareness. The rationale underlying
a forced choice test is that if the subject truly cannot detect the prime, he or she should do no better
than chance at recognizing it. However, a potential problem with the forced choice test is the
introduction of an unnecessary artifact. Instead of having participants recollect one prime, the
veridical prime, they are place in a position whereby they must consider a fabricated item. The
introduction of this artifact may result in interference with the conscious recollection of the
veridical prime. That is, participants may have some vague sense of familiarity regarding the
prime, but that familiarity was suppressed or interfered with by the artifact. Therefore participants
exposed to the forced choice test may have at some level been aware of the prime, however that
awareness may have been masked.
Indeed, this is essentially what was found in this study, using the free recall procedure.
With a free recall procedure no additional artifact is introduced. Participants are merely asked to
recall the expression(s) they saw previously. In this experiment, nine out of twelve participants
recalled the expression "happy". The majority of these participants attributed the ability to recall
the expression to some 'vague sense' ('I just sensed that the expression was happy' ) that
accompanied the recollection. Only a few participants indicated physical features ('a flash of
white') as the primary reason for recollection. Since the present study did not include a forced

Repeated Presentation

38

choice comparison group to directly compare our methods with those of Zajonc ( 1984),
veridicality of familiarity interference notion is, at this point, a question that cannot be answered.
Since the emotional expression of happiness produces physiological changes with awareness while
angry and disgust produce physiological changes without awareness it can be concluded that there
is a differential effect of the emotions on awareness or the memory of what has been seen. The
differential effect of the three emotional expressions on memory clearly suggests that the free
recall method is more successful than the forced choice method. This finding also leads to some
potentially interesting comparisons between the two methods that should be conducted in the
future.
Another explanation for differences in recall of affective priming stimuli correlates with
duration of stimulus presentation. For the typical affective priming procedure, suboptimal stimulus
presentation (Zajonc 1980, 1984) occurs over 4 ms durations. This falls one ms short of the 5 ms
duration in which affective primes were presented in the present study. It is conceivable that the
additional 1 ms in the present experiment resulted in participants meeting a level of detection
above that of threshold. However, this does not explain their inability to recall the negative
affective primes of angry and disgust.
The reason for the apparently increased saliency of happy faces over those of anger and
disgust is best explained by the following observations. It has been repeatedly observed that happy
faces are recognized more accurately than any other facial expression (Ekman, Friesen, Ellsworth
1982; Kirouac & Dore 1983; Ladavas, Umilta, Ricci-Bitti 1980).

Several researchers have

suggested that the origin of this happy face advantage is a result of hemispheric differences in
recognizing facial expressions, although they tend to disagree as to the content of those differences.

Repeated Presentation

39

For example, while Suberi and McKeever (1977) claimed right hemispheric dominance in
recognizing all facial expressions, Reuter-Lorenz and Davidson (1981), suggested that the
dominance of the two hemispheres should depend on the polarity (positive vs. negative) of the
facial expressions. Reuter-Lorenz and Davidson maintained that whereas the positive expression
of happiness was recognized predominantly by the left hemisphere, the negative expressions of
anger and disgust were recognized predominantly by the right hemisphere. Furthermore, Stalans
and Wedding (1985) obtained faster reaction times (RTs) in the right visual field with all facial
expressions, which would support the idea of a left hemispheric advantage. Finally, Hirschman
and Safer (1982) found no asymmetry in the recognition of facial expressions.
In regard to the present experiment, both negative and positive affective primes were
presented to the left visual field (right hemisphere). In the typical affective priming procedure
stimuli are presented to the center of the visual field. One could effectively argue for a right
hemispheric advantage in the present experiment. However, a right hemispheric advantage is
reported to occur for negative affective primes only, primes for which participants in the present
experiment had no conscious recollection. Thus, the obtained patterns of results in this experiment
are opposite to those which would have been predicted assuming a right hemispheric advantage.
Therefore it can be concluded that right hemispheric advantage is not an adequate explanation of
the recall of the positive affective prime "happy" in the present study.
It is interesting to note that as stimuli were presented to the right hemisphere alone, the

present study cannot provide conclusive evidence regarding hemispheric differences in evaluation
of emotional polarity. However, the present findings are inconsistent with the hypothesis of a right
hemispheric preference for recognition of negative facial expressions. As indicated previously,

Repeated Presentation

40

participants in this experiment were, at some level, cognizant of the positive expression of
happiness, but were not cognizant of the negative expressions of anger and disgust. It should be
emphasized that this latter conclusion is only tentative as the stimuli were presented at suboptimal
levels of exposure.
Finally, in some respects the present findings parallel those found by Murphy (1990).
Murphy examined participants' ability to discriminate among six specific facial expressions
(Ekman 1972) at suboptimal levels of exposure. Participants were only able to differentiate among
emotions that differed in pleasurable polarity. Happiness could be distinguished at a better than
chance level from the negative emotions fear, disgust, sadness and anger. No reliable differences
were observed among the four negative emotions.
Null Findings and Explanations
The absence of consciously reported feelings (DES) as a result of the priming manipulation
is a null finding. One possible explanation is that affect produced by subliminal facial primes is
rudimentary and possibly unconscious (Zajonc 1994). Another explanation for participants'
absence of consciously reported feeling is that pictures of emotional facial expressions were
insufficient to trigger emotional feelings.
The absence of differences in the physiological data for participants exposed to either
happiness, anger, or disgust conditions is also a null finding. Accepting a null finding requires that
the study gave a reasonable chance for the variable to be manifest. As there were only twelve
participants per condition, this null finding may very well reflect a power deficiency. Increasing
the number of participants might resolve this deficiency and reveal a putative difference in
physiological responding to different emotional expressions.

Repeated Presentation

41

Conclusion
In conclusion, the present study provides suggestive, but inconclusive support for an
independent affect and cognitive model. Neurological and behavioral studies provide both
consistent and inconsistent evidence for differential affective and cognitive systems and states.
The debate primarily centers on whether neuroanatomical and behavioral evidence supports
distinct and separate systems ( Zajonce 1980), or, alternatively, that cognitive and affective
systems are interdependent (Ledoux 1992a)
With respect to the present results, support for an independent cognitive and affective
model is contingent on the following 1) physiological activity is a synonymous representation of
affect and 2) affect produced by suboptimal facial primes is unconscious. Under these conditions,
absence of consciously reported feelings emerging with an affective physiological response is
tantamount to a disassociative relationship between cognition and affect. However, a
physiological unconscious representation of affect necessitates that physiology can be equated with
affect and that affect can then be equated with that unconscious representation. As discussed
previously, this is a theoretical question with no current resolution.
The present research raises other important theoretical questions. For instance, what are the
implications of affective stimuli for memory? How might the present results be integrated with
current work on indirect memory (Merikle & Reingold 1991), implicit memory (Roediger 1990),
repetition priming (Tulving & Schacter 1990), and the misinformation effect (Loftus 1978a). The
common feature of these extensive lines of research on memory is that subtle tests, similar to those
in this study, reveal memories of which the participant may not be aware. If participants show a
physiological response to a stimulus of which they may not be aware, can that response be

Repeated Presentation

42

considered a more subtle indicator of memory (Merikle & Reingold 1991 ), a distinct process,
namely affect (Zajonc 1994), or could this response reflect an interactive process between
cognition and affect? Apparently the present research could be interpreted with any one of these
theoretical frameworks in mind. However, it is the author' s opinion that the interactive
interpretation is the most reasonable for the following two reasons. One, the procedure and those
described elsewhere (Zajonc 1980, 1997) requires both cognitive and affective input for the
acquisition of stimuli. Secondly, neuroanatomical research (Ledoux 1992) supports an interactive
interpretation.
It is also necessary to explore whether affective stimuli other than faces (which may have

very unique properties), result in a different pattern of results. For example, would presenting
photos of a snake suboptimally change physiological activity as well as conscious reports of
feeling.
Future research in this domain will, it is hoped, lead to a more systematic understanding of
the dynamics of the interactions between physiology, affect and cognition. Above all, the methods
presented here, namely, the comparisons between the effects of suboptimal primes on
physiology and subsequent recall might stimulate future exploration of the degree to which
cognition and affect operate interdependently to modify physiology.

Repeated Presentation

43

REFERENCES
Amaral, D. G. , Price, J. L. , Pitkanen, A., & Carmichael, S. T. (1992). Anatomical
organization of the primate amygdaloid complex. In J.P. Aggleton (Ed.), The amygdala:

Neurobiological aspects of emotion, memory, and mental dysfunction (pp. 1-66). New York:
Wiley-Liss.
Bauer, R. M. (1984). Autonomic recognition of names and faces in prosopagnosia: A
neuropsychological application of the guilty knowledge test. Neuropsychologia. 22, 457-469.
Bornstein, R. F. (1987). Subliminal mere exposure effects and conscious cognition: A

study of attitude changes in response to stimuli perceived without awareness. Doctoral
Dissertation. State of University ofNew York at Buffalo.
Borod, J.C. (1992). Interhemispheric and intrahemispheric control of emotion: A focus on
unilateral brain damage, Journal of Consulting and Clinical Psychology, 60, 339-348.
Cabeza, R., Burton, A.M., Kelly, Stephen, W. (1997). Investigating the relation between
imagery and perception: Evidence from face priming. Quarterly Journal of Experimental

Psychology: Human Experimental Psychology, 50A, 274-289.
Clore, G. L. , Ortony, A., & Foss, M. (1987). The psychological foundations ofthe
affective lexicon. Journal of Personality and Social Psychology, 53, 751-766.
Davidson, R. (1992). Prolegomenon to emotion: Gleanings from Neuropsychology,

Cognition and Emotion, 6, 245-268.
Ekman, P. & Friesen, W. V. (1976). Measuring facial movement. Environmental

Psychology and Nonverbal Behavior, 1, 54-56.
Ekman, P., Friesen, W. V., & Ellsworth, P. (1982). Does the face provide accurate

Repeated Presentation

44

information? In: P. Ekman (Ed.), Emotion in the human face (2nd ed.). Cambridge: Cambridge
University Press.
Ellis, H. D. (1986). Processes underlying face recognition. In R. Bruyer (Ed.), The

neuropsychology offace perception and facial expression. Hillsdale, NJ: Erlbaum.
Gardner, H., Brownell, H, Wapner, W. & Michelow, D. (1983). Missing the point: The
role of the right hemisphere in the processing of complex linguistic materials, in E. Pericman,
(Ed)., Cognitive processes and the right hemisphere. New York: Academic Press.
Harrison, A. A. (1977). Mere exposure. In L. Berkowitz (Ed.), Advances in experimental

social psychology, Vol. 10 (pp. 39-83). New York: Academic Press.
Hirschman, R.S. & Safer, M.A. (1982). Hemisphere differences in perceiving positive and
negative emotions. Cortex, 18, 569-580.
Holender, D. (1986). Semantic activation without conscious identification in dichotic
listening, parafoveal vision, and visual masking: A survey and appraisal. Behavioral and Brain

Sciences, 9, 1-66.
Iwata, J., Chida, K., & LeDoux, J.E. (1987). Cardiovascular responses elicited by
stimulation of neurons in the central amygdaloid nucleus in awake but not anesthesized rats
resemble conditioned emotional responses. Brain Research, 418, 183-188.
Iwata, J., LeDoux, J. E., Meeley, M.P., Arneric, S., & Reis, D. J. (1986). Intrinsic neurons
in the amygdaloid field projected to by the medial geniculate body mediate emotional responses
conditioned to acoustic stimuli. Brain Research, 383, 195-214.
Izard, C. (1972). Patterns of emotions.. New York. Academic Press. 85-97.
Jacoby, L. L., & Dallas, M. (1981). On the relationship between autobiographical memory

Repeated Presentation

45

and perceptual learning. Journal of Experimental Psychology: General, 110, 306-340.
Kim, J. 1., & Fanselow, M. S. (1992). Modality-specific retrograde amnesia of fear.
Science, 256, 675-677.
Kirouac, G. & Dore, F. Y. (1983). Accuracy and latency of judgment of facial expressions
of emotions. Perceptual and Motor Skills, 57, 683-686.
Kitayama, S. (1991). Impairment of perception by positive and negative affect. Cognition
and Emotion, 5 255-274.

.

Kunst-Wilson, W. R. & Zajonc, R. B., (1980). Affective discrimination of stimuli that

cannot be recognized. Science, 207, 557-558.
Ladavas, E., Umilta, C., & Ricci-Bitti, P.E. (1980). Evidence for sex differences in right
hemisphere dominance for emotions. Neuropsychologia, 18, 361-366.
LeDoux, J. E. (1986). Sensory systems and emotion: A model of affective processing.
Integrative Psychiatry, 4 237-248.
LeDoux, J. E. (1987). Emotion. In F. Plum, (Ed.), Handbook ofphysiology: 1: The
nervous system. Volume V: Higher functions of the brain (pp. 419-460). Bethesda, MD:
American Physiological Society.
LeDoux J. E. (1989). Cognitive-emotional interactions in the brain. Cognition and
Emotion, 3, 267-289.
LeDoux, J. E. (1990). Information flow for sensation to emotion: Plasticity in the neural
computation of stimulus values. In M. Gabriel & J. Moore (Eds.), Neurocomputation and
learning: Foundation and adaptive networks (pp. 3-52). Cambridge, MA: MIT Press.
LeDoux, J. E. (1991). Emotion and the limbic system concept. Concepts in Neuroscience,

Repeated Presentation

46

2, 169-199.
LeDoux, J. E. (1992a). Brain mechanisms of emotion and emotional learning. Current
Opinion in Neurobiology, 2, 191-198.
LeDoux, J. (1996). The Emotional Brain. New York, NY. Simon & Schuster.
LeDoux, J. E., Iwata, J. , Cicchetti, P., & Reis, D. J. (1988). Different projections ofthe
central amygdaloid nucleus mediate autonomic and behavioral correlates of conditioned fear.
Journal ofNeuroscience, 8, 2517-2529.
Lindsay, D.S., & Johnson, M. K. (1989). The eyewitness suggestibility effect and memory
for source. Memory and Cognition, 17, 349-358.
Matlin, M. W. (1971). Response competition, recognition, and affect. Journal of
Personality and Social Psychology, 19, 295-300.
Merikel, P.M., & Reingold, E. M. (1991). Comparing direct (explicit) and direct (implicit)
measures to study unconscious memory. Journal ofExperimental Psychology: Learning,
Memory, and Cognition, 17, 224-233 .
. Moreland, R.L., & Zajonc, R.B. (1979). Exposure effects may not depend on stimulus
recognition. Journal of Personality and Social Psychology, 37, 1085-1089.
Murphy, S. T. (1990). The primacy of affect: Evidence and extension. Unpublished
doctoral dissertation, University of Michigan.
Niedenthal, P.M. (1987). Unconscious affect in social cognition. Ph. D. Thesis,
University ofMichigan.
Niedenthal, P.M. (1990). Implicit perception and affective information. Journal of
Experimental Social Psychology, 26, 505-527.

Repeated Presentation

47

Pallis, C. A. (1955). Impaired identification of faces and places with agosia for colour.

Journal ofNeurology, Neurosurgery, and Psychiatry, 18, 218-224.
Price, J. L. , & Amaral, D. G. (1981). An autoradiographic study ofthe projections ofthe
central nucleus of the monkey amygdala. Journal of Neuroscience , I , 1242-1259.
Purcell, D. G. , Stewart, A. L. , & Stanovitch, K. K. (1983). Another look at semantic
priming without awareness. Perception and Psychophysics, 34,65-71.
Reuter-Lorenz, P. A. & Davidson, R. J. (1981). Differential contributions ofthe two
cerebral hemispheres to the perception ofhappy and sad faces. Neuropsychologia, 15, 609-614.
Roediger, H. L. (1990). Implicit memory: Retention without remembering. American

Psychologist, 45, 1043-1056.
Schwarz, N . & Clore, G. L. (1988). How do I feel about it? The informative function of
mood. InK. Fiedler & J. Forgas (Eds.), Affect, cognition, and social behavior (pp. 44-62).
Toronto: C.J. Hogrefe.
Schwartz, G. E. (1971). Cardiac responses to self-induced thoughts. Psychophysiology, 8,
462-467.
Schweinberger, S. R., Pfuetze, E. M. & Sommer, W. (1995). Repetition priming and
associative priming of face recognition: Evidence from event-related potentials. Journal of

Experimental Psychology: Learning, Memory and Cognition, 21 , 722-736.
Sherwood, A., Allen, M. T, Fahrenberg, J. , Kelsey, R. M., Lovallo, W. R., & Van Doomen,
L. J. P. (1990). Methodological guidelines for impedance cardiography. Psychophysiology, 27, 123.
Stalans, L. & Wedding, D. (1985). Superiority of the left hemisphere in the recognition of

Repeated Presentation

48

emotional faces. International Journal ofNeuroscience, 25,219-223.
Suberi, M. & Mckeever, W.F. (1977). Differential right hemispheric memory storage of
emotional and nonemotional faces. Neuropsychologia 5, 757-768.
Takeuchi, Y., Mclean, J. H., & Hopkins, D.A. (1982). Reciprocal connections between the
amygdala and parabrachial nuclei: Ultrastructural demonstration by degeneration and axonal
transport of horseradish peroxidase in the cat. Brain Research, 239, 538-588.
Tranel, D. & Damasio, A. R. (1988). Non-conscious face recognition in patients with face
agnosia. Behavioral Brain Research, 30, 235-249.
Tranel, D., Damasio, H., & Damasio, A.R. (1995). Double dissociation between overt and
covert face recognition. Journal of Cognitive Neuroscience, 7, 425-432.
Tulving, E., & Schacter, D. L. (1990). Priming and human memory. Science, 247, 301306.
Winkielman, P., Zajonc, R. B. & Schwarz, N. (1997). Subliminal affective priming resists
attribution interventions. Cognition and Emotion, 11, 433-465.
Zanjoc, R. B. (1968). Attitudinal effects of mere exposure. Journal ofPersonality and

Social Psychology [Monograph] 9, 1-27.
Zanjoc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American

Psychologist, 35, 151-175.
Zajonc, R. B. (1984). On primacy of affect. InK. R. Scherer & P. Ekman (Eds.),

Approaches to emotion (pp. 259-270). Hillsdale, NJ: Lawrence Erlbaum.

Repeated Presentation

49

Zola-Morgan, S., Squire, L. R., Avarez-Royo, P., & Clower, R. P. (1991). Independence of
memory functions and emotional behavior: Separate contributions of the hippocampal formation
and the amygdala. Hippocampus, 1, 207-220

Repeated Presentation

APPENDIX A
SOURCE MONITORING TEST

50

Repeated Presentation

51

Memory Test

On the page that follows, please write down the name of the facial expression (e.g what type of
emotion is expressed) presented on the computer monitor before you and indicate (by
checkmark) where you seen that expression in the designated columns. In addition we would
like you to indicate (in column provided) on a scale of 0-100 ( 0 being not confident at all, 100
being absolutely certain) how confident you are that your choice of designation was the correct
one. You have 14 seconds to make your decisions so please work quickly as possible.

Repeated Presentation 52

Name

FacialExpression
Type

.

Saw

i .r~ io

Saw
Expression

Saw
Expression

1

~

Never
Saw

Expression
During
During
urmg
Fast
Fast
Slow
Presentation Presentation Presentation

Confidence
Rating
0-100

1.
2.

3.
4.
5.
6.
7.

8.
9.
10.
11.
12.
13.
14.
15.
16.
17.

..

Repeated Presentation 53

Name

FacialExpression
Type

.
18.
19.
20.
21.
22.
23.
24.
25.

Saw

~r~ io

Saw

~r~io
urmg

~

Saw

io
urmg

Never
Saw

Expression
unng
Fast
Slow
Fast
Presentation Presentation Presentation

!

1

Confidence
Rating
0-100

Repeated Presentation

APPENDIXB
DIFFERENTIAL EMOTIONAL SCALE

54

Repeated Presentation

55

EMOTIONS SCALE
page.

A number of words that describe different emotions or feelings are listed on the following

Read each statement and then circle the appropriate number to the right of the emotion
word to indicate the extent to which it describes the way you felt during the performance of the
preceding task.
There are no right or wrong answers. Do not spend too much time on any one emotion
word, but give the answer that seems to best describe your feelings as you recall them.
Note: Mod. = Moderately
Consid. = Considerably

Repeated Presentation

Phase

ld:

56

Not at
All

Slightly

Mod.

Consid.

Very
Strongly

1.

Attentive

0

1

2

3

4

2.

Joyful

0

1

2

3

4

3.

Surprised

0

1

2

3

4

4.

Sad

0

1

2

3

4

5.

Irritated

0

1

2

3

4

6.

Guilty

0

1

2

3

4

7.

Bashful

0

1

2

3

4

8.

Afraid

0

1

2

3

4

9.

Sluggish

0

1

2

3

4

10.

Concentrating

0

1

2

3

4

11.

Enthusiastic

0

1

2

3

4

12.

Amazed

0

1

2

3

4

13.

Scornful

0

1

2

3

4

14.

Shy

0

1

2

3

4

Repeated Presentation

Not at
all Slightly

Mod.

57

Very
Strongly

Consid.

15.

Scared

0

1

2

3

4

16.

Fatigued

0

1

2

3

4

17.

Delighted

0

1

2

3

4

18.

Startled

0

1

2

3

4

19.

Angry

0

1

2

3

4

20.

Fearful

0

1

2

3

4

21 .

Happy

0

1

2

3

4

22.

Mad

0

1

2

3

4

23.

. Frightened

0

1

2

3

4

24.

Excited

0

1

2

3

4

25.

Jittery

0

1

2

3

4

What expression(s) did you see?