Pavlovian, Observational and Instructed Fear Learning:

1
Pavlovian, Observational and Instructed Fear
Learning Emotional Responses to Unmasked and
Masked Stimuli Andreas Olsson, Kristen Stedenfeld
Elizabeth A. Phelps, New York University
INTRODUCTION ? The flexibility of human biology
and culture allows for a variety of ways to form
associations between objects or events
(conditioned stimuli, CS) and naturally aversive
outcomes (unconditional stimuli, UCS). Among the
ways to learn to predict the aversive outcome of
a stimulus are (1) Pavlovian conditioning
Associations are acquired through direct
experience of the UCS paired with the CS. (2)
Observational learning Knowledge of the
properties of the UCS and the CS-UCS contingency
is acquired through observing another individual
having a first-hand experience of the CS in
combination with the UCS. (3) Instructed
learning Information about the UCS and its
association with the CS is transferred solely
through symbolic means, such as language. ?
Earlier studies have shown that expectancies
resulting from Pavlovian conditioning1
observational2 and instructed learning3,4
produces similar emotional responses, as
indicated by a differential skin conductance
response (SCR) to the predictive stimulus (CS)
vs. a control stimulus (CS-) when the CSs are
presented to subjects full awareness. ? When
fear-learning is acquired through Pavlovian
conditioning, subjects also display a
differential SCR to subsequent exposures of CSs
that are presented briefly and followed by a mask
to prevent explicit awareness5. ? However, it is
still unknown to what degree fear acquired
through observation or verbal instruction,
produces learning that modulates emotional
responses without explicit awareness of the CSs.
GOALS This study investigates the relation
between emotional and cognitive processes in
fear-learning, specifically to what degree a
learned fear response is modulated by (1) type of
learning (Pavlovian vs. observational vs.
instructed) and (2) explicit awareness of the
conditioned stimuli (unmasked vs. masked).
DISCUSSION
? Our results confirm earlier findings that shock
expectancies acquired with and without direct
experience of the shock elicit similar emotional
responses when the target stimuli allows for
explicit awareness. ? In accordance with
earlier data, we also found that CSs which were
presented without explicit awareness resulted in
a differential SCR in the Pavlovian learning
group. No such effect was found in the instructed
group. ? Interestingly, although not
significant, the observational group displayed a
trend towards a differential SCR to masked CSs.
This indicates that observing emotional reactions
in another person might provide a more efficient
route to emotional fear-learning than that by way
of symbolic representations only. ? Taken
together, these results indicate that emotional
responses to stimuli are dependent on how the
stimuli acquired their emotional value and
whether they are presented to allow for explicit
awareness or not.
RESULTS Differential Skin Conductance Responses
to Unmasked vs. Masked Stimuli (Acquisition Phase)
p lt 0.01
p lt 0.01
p lt 0.01
METHOD Subjects 126 paid participants were
randomly assigned to one of the three learning
groups. Twelve subjects were excluded from
further analysis after having reported to have
seen the masked CSs. Subjects that did not
display any SCR (non-responders, N14) or showed
no signs of learning in the unmasked condition
(N33) were also excluded. The final sample
consisted of 67 participants. Materials Two
angry faces served as CS (associated with shock)
and CS- (control). A neutral face was used as a
mask. All images were taken from the Ekman
Friesen6 set of emotional faces. Electric shocks
were delivered to the right wrist and SCR were
measured on two fingers on the left hand.
Procedure Each experimental session consisted
of three phases (1) Habituation (baseline SCR
was estimated), (2) Acquisition (learning was
measured when shocks were expected) and (3)
Extinction (rate of extinction was sampled as
shocks were no longer expected). Differential SCR
to CS vs. CS- served as the dependent variable.
Here, only the acquisition phase is discussed. ?
The acquisition phase consisted of 24 trials,
equally divided into four types unmasked (CS
and CS-) and masked (CS and CS-). In all groups,
trial duration was 6 seconds and the inter trial
interval varied between 7 to 15 seconds. In
masked trials, the target was exposed for 30 ms
and immediately followed by the mask. ? In the
Pavlovian conditioning group, the CS was
associated with the shock through direct
experience of an uncomfortable, but not painful,
electric shock that co-terminated with the
unmasked CS (fig.1). ? In the observational
learning group, participants watched a movie
(before the experiment) of a confederate
participating in the Pavlovian conditioning group
(fig.2). No shocks were administered. ? The
instructed learning group only received verbal
information about the UCS and its association
with the CS (fig.3). No shocks were given.
p lt 0.05
() p .085
n.s.
Fig.1. In the Pavlovian group, participants
received a shock paired with unmasked CS.
SELECTED REFERENCES 1. Lang, P. J., Ohman, A.
Simons R. F. (1978). The psychophysiology of
anticipation. In J. Requin (Ed.), Attention and
performance VII (pp. 469-485). Hillsdale,
NJErlbaum. 2. Hygge S, Ohman A. (1978). Modeling
processes in the acquisition of fears vicarious
electrodermal conditioning to fear-relevant
stimuli. Journal of Personality and Social
Psychology, 36, 271-9. 3. Hugdahl, K. Ohman, A.
(1977). Effects of instruction acquisition and
extinction of electrodermal responses to
fear-relevant stimuli. Journal of Experimental
Psychology Human Learning and Memory, 3,
608-618. 4. Phelps, E. A., OConnor, K. J.,
Gateby, J. J., Grillon, C., Gore, J. C., Davis,
M. (2001). Activation of the amygdala by
cognitive representations of fear. Nature
Neuroscience, 4, 437-441. 5. Esteves, F.,
Dimberg, U. Ohman, A. (1994). Automatically
elicited fear Conditioned skin conductance
responses to masked facial stimuli. Cognition and
Emotion, 8, 5, 393-413. 6. Ekman, P., Friesen, W.
(1976). Pictures of facial affect. Palo Alto
Consulting Psychologists Press .
? A 3 (learning group) x 2 (masking condition)
ANOVA showed a significant effect for masking
condition F(1, 64) 68.15. No other effects were
significant. ? T-tests of the differential SCR
of both unmasked and masked trials in each
individual group revealed the following
Fig.2. In the observational group, participants
watched a movie of a confederate submitted to the
same experimental conditions as in the Pavlovian
group. Subjects received no shocks.
Pavlovian conditioning (N24) ? Unmasked CS
elicited a greater SCR than unmasked CS-, t(23)
6.43, plt.001 (M.41, SD.31). ? Also on masked
trials, there was a significant differential
response, t(23)2.28, plt.05 (M.12, SD.26).
Observational learning (N16) ? SCR to unmasked
CS was greater than to unmasked CS-, t(15)
6.277, plt.0001 (M.45, SD.29). ? On masked
trials, there was a trend towards a greater SCR
to CS vs. CS-, t(15)1.86, p.83 (M.07,
SD.12).
Instructed learning (N27) ? Unmasked CS
elicited a significantly greater SCR than
unmasked CS-, t(26)9.9, plt.0001 (M.38,
SD20). ? There was no significant difference
between CS and CS- in the masked condition.
Acknowledgements We want to thank Laura Thomas
for her assistance with programming. This
research was supported by the National Institute
of Health, MH62104 to EAP and a Fulbrigh
Fellowship to the first author.
Fig.3. In the instructed group, learning was
acquired through verbal instructions given by the
experimenter. No shocks were administered.