1
Fear reduction towards conditioned fear cues via
instructed extinction
A. L. Dokht
Collegekaartnummer: 5694116
Adres: Hoofdweg 544
Telefoonnummer: 06-50999967
E-mail adres: Ayeh.Dokht@gmail.com
Begeleiders: Tom Beckers & Angelos-Miltiadis Krypotos Onderzoeksinstelling / locatie: Universiteit van Amsterdam
Datum: September 1st, 2014
2 Abstract
Fear extinction, the laboratory parallel of exposure therapy for anxiety disorders, is based on Pavlovian Extinction and encompasses the repeated presentation of neutral stimuli (e.g., a
geometrical object; Conditioned Stimulus or CS) previously associated with an aversive event (e.g., a shock; Unconditioned Stimulus or US) such as that the presentation of the CS will evoke fear
responses (i.e., subjective apprehension, physiological arousal and action tendencies), without the US. Such a procedure will typically lead to the depletion of fear responses Considering that verbal
instructions play an important role in everyday life and continuously guide our actions, we tested whether mere instructions could lead to the reduction of conditioned fear responses, without the experience of CS-noUS associations. Participants initially underwent fear conditioning, in which pictures of one neutral object were sometimes followed by the presentation of a shock (CS+) and pictures of another object were never followed by shock (CS-). Next, participants in the first group were informed that no more shocks would be administrated (Instructed Extinction group). The second group underwent a fear extinction procedure without receiving any instructions in advance. A third control group just underwent a fear acquisition procedure. In order to obtain a complete overview of fear reduction, we measured all three fear responses. A depletion of fear responses was found for the indices of subjective apprehension and partially for physiological arousal, the startle response, but not for electrodermal responding (physiological arousal) and avoidance tendencies. The present study helps to increase our understanding of how verbal instructions affect extinction learning. In turn increased understanding could lead to improved forms of exposure therapies and eventually lead to lower prevalence rates for anxiety disorders.
3 Introduction
Fear in essence helps individuals to analyze signals of impending threats and help organisms prepare for appropriate response (Beckers, Krypotos, Boddez, Effting & Kindt, 2013). However, when individuals start exhibiting excessive fear responses towards innocuous cues, like photographs of spiders, fear can derail into an anxiety disorder or phobia (Beckers, et al., 2013). Anxiety disorders have a lifetime prevalence of 19,6%, among which more than half have a permanent character (de Graaf, ten Have, van Gool & van Dorsselaer, 2012). This makes anxiety disorders the number one cluster of mental disorders afflicting the Dutch population (de Graaf, et al., 2012). As such, ways to prevent and treat anxiety disorders is a key research area in clinical psychology and neurosciences. A way to experimentally study anxiety disorders’ symptomatology is via the fear conditioning paradigm, which is based on Pavlovian learning (Mineka & Zinbarg, 2006). In the traditional fear-conditioning paradigm, initially neutral cues (a geometrical shape; Conditioned stimulus or CS) preceding an aversive outcome (electric stimulation; Unconditioned stimulus or US) comes to elicit (conditioned) fear (e.g., increased skin conductance levels; Conditioned Responses or, CR)(Pavlov, 1927). Those conditioned fear responses are reflected in the three indices of emotion, namely
subjective apprehension, physiological arousal and avoidance tendencies (Foa & Kozak, 1986; Mauss, & Robinson, 2009), with some theorists even arguing that emotions are primarily reflected in action tendencies (Frijda, 2010).
Once conditioned responses have been acquired, the presentation of the conditioned stimulus without the presentation of the US (i.e., fear extinction; CS-NoUS; Lovibond, et al., 2009; Pavlov, 1927), would typically lead to the depletion of the responses. Of importance, exposure therapy, the dominant intervention for anxiety disorders, is based on fear extinction (Culver, Stoyanova & Craske, 2011). As such, the experimental paradigm of fear extinction can help us to gain more insight on fear exposure. Although at this moment little is known about the translation of verbal information into actions, such as avoidance behavior, verbal, or explicit, instructions play a fundamental role in everyday life and often guide our actions (Hartstra, Kühn, Verguts & Brass, 2011). Recent studies have demonstrated that when combining extinction learning with explicit instructions (e.g., this picture will no longer be followed by an electric shock), a direct depletion in subjective apprehension was found and was found to be relatively effective in reducing physiological arousal, such as electrodermal responding. However, no effective reduction in startle response was found (Sevenster, Beckers & Kindt, 2012).
4 Research on Instructed Extinction has focused namely on the combination of fear extinction with explicit instructions (Sevenster et al., 2012) and not the mere influence of instructions. Previous research has also assessed physiological and subjective expressions of conditioned fear, but has not included avoidance tendencies as a dependent variable and therefore not assessing all indices of fear (Beckers et al., 2013). The goal of the current study was to investigate whether mere instructions could (i.e., Instructed Extinction) lead to similar or more reduction in fear symptomatology compared to the standard fear extinction procedure (i.e., Experiential Extinction).
If Instructed Extinction leads to corresponding or stronger depletion of conditioned responding as for Experiential Extinction, this can have useful implications for theory and practice. In theoretical settings, targeting conditioned fear responses through instructions, could translate into an increased understanding of therapeutic efficacy for anxiety disorders through different forms of exposure therapy. In turn, increased understanding could lead to improved therapies and lower relapse rates for anxiety disorders.
Participants initially underwent a fear acquisition procedure in which one initially neutral object was sometimes followed by an electric stimulus (Conditioned Stimulus; CS+) whereas another neutral object was not (CS-). Following fear acquisition, participants underwent two different forms of extinction learning. The first group (Experiential Extinction) underwent a standard extinction procedure, where the CSs were not followed by an electric stimulus. The second group (Instructed
Extinction) was instructed that the CSs would not be followed by an electric stimulus. The control
group (Fear Acquisition-only) waited for one minute. The study entailed 5 phases: fear acquisition, fear extinction, extinction testing, the Approach- Avoidance Task and measurement of subjective fear. Subjective apprehension and measures of physiological arousal (i.e., EMG and SCR) were obtained during the fear acquisition, fear extinction and extinction-testing phase. Action tendencies were assessed during the Avoidance Approach Tasks.
In order to obtain a complete view of fear symptomatology depletion, all three indices (i.e., subjective apprehension, physiological arousal and avoidance tendencies (Beckers et al., 2013; Foa & Kozak, 1986; Mauss & Robinson, 2009) were included. We expected that participants in the
Instructed Extinction group would experience less subjective apprehension than the participants in the fear acquisition-only group. It was also expected that subjective apprehension ratings from participants in the Instructed Extinction group would not differ from participants in the Experiential Extinction group, and for both groups would be less than for participants in the Fear Acquisition-only group. A reduction of physiological arousal was expected for participants in the Instructed
5 Extinction and Experiential Extinction group, which was similar to the expected reduction for subjective apprehension. For both groups the reduction is expected to be less than for participants in the Fear Acquisition-only group. Regarding action tendencies, it was expected that reaction time responses from participants in the Instructed Extinction group would differ from the responses of participants in the Fear Acquisition-only group. The participants in the Instructed Extinction group would show less difference in reaction times than the participants in the Fear Acquisition-only group. It was also expected that action tendency responses from participants in the Instructed Extinction group would not differ from participants in the Experiential Extinction group. For both groups resulting in fewer differences in reaction time than for participants in the Fear Acquisition-only group.
Method Participants
A total of 67 participants (14 males) participated in this study. All participants were screened for previous medical or psychiatric condition and excluded if criteria were met (i.e. pregnancy; seizure disorder; hearing disorder; cardiovascular disease; Attention Deficit Hyperactive Disorder; Post Traumatic Stress Disorder; Fear disorder; use of medication that could potentially influence attention span). After data collection, ten participants were removed (mean ± SD age, 22.4 ± 4.02 years), as they could not articulate the CS-US contingencies (i.e., contingency unaware participants (Lovibond, et al., 2009)). We defined contingency awareness as higher US-expectancy ratings for the CS+, compared to the CS-, in the last two fear acquisition trials, as well as the successful report of CS-US contingencies in the exit interview.
Participants were randomly assigned to thee different conditions; Fear Acquisition-only (n = 18), Experiential Extinction (n = 19) and Instructed Extinction (n = 20). The ethics committee of the University of Amsterdam approved the study and informed consent was obtained from all participants.
Apparatus
Stimuli
The pictures of a cube and a cylinder served as CS+ or CS-. The pictures were depicted from four viewpoints (50 mm × 50 mm) against a white frame. The allocation of the cube or cylinder pictures to the CS+ and the CS- was counterbalanced between participants. The same pictures were used for
6 the AAT, although the surrounding frame differed. For all groups, the surrounding frame had a landscape (100 mm × 55 mm) or portrait (55 mm × 100 mm) orientations with the CS centered in the middle of each frame (Krypotos et al., 2014). During the AAT phase, a white manikin figure (71 mm × 95 mm) was also presented (see below).
Electric stimulation was delivered to the wrist of the non-preferred hand through two Ag electrodes and the electric stimulation served as the US (Effting & Kindt, 2007). The electric stimulus was administered with the Digitmer DS7A constant current stimulator (Herfordshire, UK) with duration of 2 ms.
Fear potentiated startle
The potentiation of the eyeblink startle reflex in response to a loud noise (i.e., startle probe) was used as measurement for the conditioned fear startle response. The startle response was measured by electromyography (EMG) of the right orbicularis oculi muscle. Placement of EMG and conversion into data was done similar to previous studies in our lab (e.g. Sevenster et al., 2012; Soeter & Kindt, 2010). Peak amplitude of startle was measured between 0- 250 ms after the startle probe onset.
Noise alone (NA) trials were used for accessing baseline startle response activity and each CS presentation was paired with administration of the loud noise and during intertrial intervals (Noise Alone, NA). Participants also received habituation startle trials. All auditory stimuli were
administered binaurally through headphones and lasted for 40 ms at 104 dB. Apart from the startle probe, throughout the whole fear acquisition procedure, continuous background noise (70 dB) was delivered.
Skin conductance response
Electrodermal activity was measured using an input device with a sine-shaped excitation voltage (1 V peak- peak) of 50 Hz, which was derived from the mains frequency as was done in similar studies performed in this lab (e.g., see, Sevenster et al., 2012; Soeter & Kindt, 2010). The two Ag/AgCl electrodes of 20 × 16 mm were placed on the medial phalanges of the second and third fingers of the non-preferred hand. The electrodes were kept in place using adhesive tape. The SCR signal was sampled at 1000 Hz. The analogue output was digitized at 100 Hz by a 16-bit AD-converter (National Instruments, NI-6224) by leading the signal from the input device through a
signal-conditioning amplifier. Skin conductance responses towards each CS were determined by subtracting the average baseline (2 seconds before CS onset) to peak difference within 1-7 s window following
7 CS onset. This method has been extensively used in previous research (Sevenster et al., 2012; Soeter & Kindt, 2010).
Assessment of fear and US expectancy
During fear conditioning, fear expectancy ratings were measured trial by trial, within 7 seconds after CS onset, using an 11-point Likert scale, ranging from -5 (certainly no electric stimulus) to +5 (certainly an electric stimulus). Measurement of fear expectancy ratings was similar to previous studies (e.g., Krypotos, et al., 2014).
Subjective evaluation of fear was assessed on an 11-point Likert scale ranging from -5 (unpleasant) to +5 (pleasant). Also US intensity (weak through unbearable), US startlingness (“not through” to “too strong”) and valence (negative through positive) were measured after the AAT on the corresponding Likert scales. Measurement was similar to previous studies (e.g., Krypotos, et al., 2014).
Trait and state anxiety were measured using the state and trait anxiety inventories (STAI-S and STAI-T respectively) (Spielberger, Gorsuch & Lusthene, 1970). The tendency to react fearful to anxiety symptoms was measured with the Anxiety Sensitivity Index (ASI) (Peterson & Reiss, 1992). Procedure
Please view table 1 for a schematic overview of this study.
Preparation
Participants were seated in front of a computer screen. After reading the information brochure and signing the consent form, participants filled in the STAI-S. Afterwards, the EMG, SCR and shock electrodes were attached on the participant. Next, US intensity was determined individually and participant deemed the electric stimulus as ‘unpleasant but not painful’.
Fear conditioning
All participants received oral and on-screen instructions stating that pictures would be presented of two objects with one object sometimes followed by an electric stimulus whereas the other object would never be followed by an electric stimulus. Participants were asked to learn to predict which object would be followed by an electric stimulus, which object would not be followed by an electric
8 stimulus and to rate their expectancy of receiving an electric stimulus on an 11-point Likert scale ranging from -5 (certainly no electric stimulus) to +5 (certainly an electric stimulus).
Fear conditioning consisted of the presentation of pictures of two different objects (i.e., a cube and a cylinder). One of these objects, the CS+, was followed by an electric stimulus in semi-random order 6 out of 8 times (i.e., one unreinforced CS+ trial per blocks of four trials). The CS-, was never followed by an electric stimulus. Assignment of pictures of the cube and the cylinder to the CS+ or CS- was counterbalanced across participants. Each CS was presented for 8 s and CS order was semi-random (no more than 2 consecutive CS+, CS- or NA trials). After 7 s, the startle probe was administered, and in the case of CS+, after another 500 ms, the electric stimulus (US) was administered. Intertrial Intervals (ITI’s) varied from 15 to 25 seconds, with an average of 20 s.
Fear Extinction
The extinction phase differed between groups:
Experiential Extinction condition: For participants assigned to this condition, after a one-minute break,
the experiment continued similar to the fear acquisition phase. Main difference from the fear acquisition phase was that none of the CSs were reinforced. In total participants received 10 unreinforced CS+, 10 CS- stimuli and 10 NA trails during this phase. After a one-minute break, participants were instructed to continue as before and rate the US expectancy for the CS+ and CS-.
Instructed Extinction condition: Participants assigned to this condition were instructed that from that
point on, all presented pictures would not be followed by an electric stimulus.
Fear Acquisition-only condition: Participants assigned to this condition were asked to wait for one-minute
before the experiment continued. Similar to the Experiential Extinction condition, participants were asked to continue as before, to rate the US expectancy for the CS+ and CS- after a one-minute break.
Test
After the extinction phase, participants in all groups received 2 unreinforced CS+ and 2 CS- trials. In order to measure extinction, participants were asked to rate their US expectancy during each CS presentation.
9 After a waiting period of 3 minutes, electrodes were removed and the AAT started. In that task oral and on-screen instructions consisted of asking the participants to move a white manikin figure as fast and accurately towards or away from the depicted CS depending on the orientation of the white frame. On the first block, for example, participants had to move the manikin towards pictures with horizontal orientation and away from pictures with vertical orientation. Instructions were reversed for the second block.
The white manikin figure appeared centered on the bottom or top half of the computer screen. After 1500 ms, a CS picture was presented. To move the manikin participants had to press the “Y” button, labelled with the symbol ↑, for moving the manikin upward and “B”, labelled with the symbol ↓, for moving the manikin downwards. After 500 ms the manikin disappeared from the screen and in case of an incorrect answer, direct feedback followed. The AAT consisted of two blocks with 4 practice trials followed by 16 test trials per block. During the practice trials, 2 CS+ and 2 CS- were presented randomly. During test trials, each viewpoint was presented twice in semi-random order, with no more than two consecutive CS+ or CS- trials.
Table 1.
An overview of the experimental phases for all groups.
Groups Acquisition Extinction Test Break AAT
Experiential Extinction 6/8 CS+ (75% reinforcement 8CS-Short break (1 min) and 10 CS+ 10 CS-(no reinforcement) 2 CS+ 2 CS- 3 minutes 8CS+ 8CS- avoid/ approach Instructed Extinction 6/8 CS+ (75% reinforcement 8CS- Instructions 2 CS+ 2 CS- 3 minutes 8CS+ 8CS- avoid/ approach Fear Acquisition-only 6/8 CS+ (75% reinforcement 8CS- Waiting (1 minute) 2 CS+ 2 CS- 3 minutes 8CS+ 8CS- avoid/ approach
10
Note. CS = Conditioned Stimulus. AAT= Avoidance Approach Task. Exit Interview
Upon completion of previous stages, participants were asked to rate CS, US pleasantness, fill in the STAI-T, ASI and demographics. After completing these questionnaires participants were debriefed, rewarded and thanked.
Statistical analyses
All our statistical analyses were set a priori. Considering the small sample size, setting the analysis a priori, made sure that no confirmation bias would take place during the statistical analysis. For all statistical analysis, the alpha level was set at .05. All ANOVA assumptions were validated before main analysis was conducted (Field, 2009). In case of violation of the assumption of sphericity, Greenhouse-Geisser correction was used.
STAI-T and STAI-S, US intensity and US evaluation scores were subjected to separate one-way ANOVA’s with group (Fear Acquisition-only, Experiential Extinction and Instructed
Extinction) as independent factors.
US-expectancy ratings, startle responses and skin conductance responses were square transformed and subjected to a mixed ANOVA with stimulus (CS+ vs. CS-) and trial (acquisition phase: 8 levels; Extinction phase: 10 levels; Test phase: 2 levels) as within-subject factors and group (Fear Acquisition-only vs. Experiential Extinction vs. Instructed Extinction) as between-subject factors.
Median reaction times during the AAT were calculated for each participant depending on response type (approach or avoidance) and stimulus type (CS+ or CS-). Reaction times were analyzed comparing Experiential Extinction with Instructed Extinction and Instructed Extinction with Fear Acquisition-only. This was done with a 2 (stimulus: CS+, CS-) × 2 (response: approach, avoidance) × 3 (group: Experiential Extinction, Instructed Extinction and Fear Acquisition-only) ANOVA. Stimulus and response were the within-subject factors and group was the between-subject factor.
11 Results
Groups did not differ in terms of state anxiety, F(2, 54)= 1.72, p = .19, trait anxiety, F(2, 54)=1.04,
p= .36, anxiety sensitivity, F(2, 54)= 2.22, p= .12, US intensity, F(2, 54)= .09, p= .92, and US or CS
evaluation, Fs < 1. Furthermore, no between groups differences were found, for gender and age (all
t2s < 1).
3.1. US Expectancies
US-expectancies across groups and across the different conditioning phases are shown in Figure 1. During fear acquisition, participants learned to expect an electric stimulus after a CS+ and not after a CS-, Stimulus × Trial interaction, F(4.49, 242.59)= 58.11, p < .001, ŋp²= .52, which were similar
across groups, Stimulus × Trial × Group interaction (F<1).
Differential expectancy ratings decreased during the extinction phase for the Experiential Extinction group, Stimulus × Trial interaction, F(2.81, 53.33)= 14.30, p < .001, ŋp²= .43. Analysis
also confirmed a main effect for stimulus, F(1, 19)= 93.45, p < .001, ŋp²= .83. The combination of
the found main effect for stimulus with the decrease found for US expectancy ratings, concludes to partial extinction for the US expectancy ratings. When comparing first extinction trial with last extinction trial, using paired samples t-tests a successful decrease in CS differentiation was found. Although differences between the CS+ and CS- remained, differences between the US-expectancy ratings for participants decreased. The decrease in CS differentiation was due to a decrease in CS- ratings, t (20)=-1.95, p= .03, d= .44 but mainly due to a decrease in CS+ ratings, t(20)=-5.26, p< .001,
d=-.42. Finally, we observed a trend towards differences in differential US expectancy ratings during
the test phase, Stimulus × Trial × Group, F(2, 54)= 2.66, p= .08, ŋp² = .09.
Further analyses during the test phase showed that for the Experiential Extinction group and the Instructed Extinction group the main effect of Stimulus was significant, F(1, 37)= 19.17, p<.001, ŋp²= .34 and groups did not differ on US expectancy responding, Stimulus × Group, F< 1, and responding did not differ over time, Stimulus × Group × Trial, F <1. Participants in the Instructed Extinction group did report lower differential US expectancies, F(1, 18)=6.19, p= .023, ŋp²= .26,
than participants in the Experiential Extinction group, F(1, 19)=14.54, p= .001, ŋp²= .43, during the
test phase. When comparing with the Fear Acquisition-only group during the test phase, F(1, 17)=241.97, p<.001, ŋp²=.93, for both groups a decrease in main effect for US expectancy responding was found. In conclusion, although initially a trend was found during the test phase
12 between the groups, US expectancy ratings for participants in the Instructed Extinction group
decreased more than for participants in the Experiential Extinction group. (a)
(b)
(c)
Figure 1. Mean US-expectancy ratings to the CS+ and CS- during acquisition, extinction and test for
(a) Experiential Extinction, (b) Instructed Extinction, and (c) Fear Acquisition-only groups.
3.2 Fear Potentiated Startle Response
13 During acquisition, differential startle responses were established as reflected in a main effect of Stimulus, F(1, 54)= 9.23, p < .01, ŋp²= .15. Specifically, startle responses were higher in the CS+
compared to the CS- trials, a result which was similar across groups (F < 1).
A main effect for stimulus was found, F(1, 19)=11.81, p= .003, ŋp²= .38, during the extinction phase and no decrease in startle responses, Stimulus × Trial interaction (F< 1.2), was found for participants in the Experiential Extinction group. Although no extinction for startle responses was found at first, using paired samples t-tests, comparing the first extinction trial with last extinction trial, a successful decrease in startle differentiation was revealed. This was due to a decrease in CS- ratings, t (19)=-2.68, p<.05, d=.46 and a decrease in CS+ ratings, t(19)=-1.78, p<.05, d=-.75, resulting in partial extinction of the startle response for participants in the Experiential Extinction group.
During the test phase, the main effect for Stimulus remained, F(1, 54)=12.37, p <0.01,
ŋp²=.19. Although the groups differed in responding for stimulus, Stimulus × Group: F(2, 54)=6.37, p < .001, ŋp²= .19, no differences were observed between groups in differential startle responses
during the test phase, Stimulus × Trial × Group interaction, F< 1. In order to decompose the stimulus × Group interaction, further analysis were necessary.
Further analyses showed that for the Experiential Extinction and Instructed Extinction group differential startle response for stimulus was significant for both groups during the test phase, Stimulus, F(1, 37)=13.51, p < .001, ŋp²= .27. The Instructed Extinction group differed in responding
from the Experiential Extinction group for stimulus, Stimulus × Group, F(1, 37)=4.91, p = .03, ŋp²=
.12, but did not differ in startle response ratings during the test phase, Stimulus × Trial × Group (F< 1.0). For participants in the Experiential Extinction group lower startle responses were found (F(1, 19)=8.37, p=.009, ŋp²=.31) than for participants in the Instructed Extinction group (F(1,18)=9.17,
p=.007, ŋp²=.43) during the test phase. Participants in the Fear Acquisition-only group revealed that
no significant effect was found (F < 1) for the differential startle response. (a)
14 (b)
(c)
Figure 2. Mean startle response to the CS+, CS- and noise alone (NA) during acquisition, extinction
and test for (a) Experiential Extinction, (b) Instructed Extinction, and (c) Fear Acquisition-only groups.
3.3 Skin conductance responses
Skin Conductance responses across groups and across the different phases are shown in Figure 3.
During fear acquisition, analysis showed that differential Skin Conductance response ratings were established as reflected in a main effect for Stimulus; F(1, 54)=7.07, p = .01, ŋp²=.12, which was
similar in all groups, Stimulus × Trial × Group (F < 1). Looking into extinction for Experiential Extinction, no decrease in differential rating was shown, Stimulus × Trial (F < 1). In addition, no differential ratings of electrodermal responding for stimulus were found (F < 1). Further analyses, using paired samples t-tests, comparing first extinction trial with last extinction trial, did reveal a
15 successful decrease in differentiation. The decrease in differentiation was due to a decrease in CS+ electrodermal response ratings, t (20)=-2.78, p< .05, d= .34 but no significant decrease in CS- electrodermal response ratings (ts < 1).
During the test phase, no differences between groups in differential electrodermal response ratings was found, Stimulus × Trial × Group (F <1.2). Further analyses showed that when comparing the groups with one another (Experiential Extinction vs. Instructed Extinction; Experiential Extinction vs. Fear Acquisition-only), no differential electrodermal response ratings were found (Stimulus × Group × Trial, Fs <1.2). In contrast, when analyzing the groups separately, for participants in the Instructed Extinction group a differential electrodermal response rating remained for stimulus, (F(1,18)=5.41, p=.03, ŋp²=.23. For participants in the Experiential Extinction
and Fear Acquisition-only group differential responding was not found (Fs < 1). In conclusion, participants from all groups did not differ on electrodermal response ratings during the test phase. (a)
16 (c)
Figure 3. Mean skin conductance interval response times (EIR) to the CS+ and CS- during
acquisition, extinction and test for (a) Experiential Extinction, (b) Instructed Extinction, and (c) Fear Acquisition-only groups.
3.4 AAT
Median reaction times across groups and across the different phases are shown in Figure 4.
Median reaction times across groups during the AAT approached significance, Stimulus × Response × Group, F(2, 54)= 2.66, p = 0.08, ŋp²=.09. However, no differences in reaction times between
approaching or avoiding the stimuli were found for all participants, Stimulus × Response interaction (F(1, 540= 1.84, p = .18, ŋp²=.03).
To further decompose the previous findings, separate analysis was conducted for each group separately using a 2 (stimulus: CS+, CS-) × 2 (Response: Avoidance, Approach) ANOVA for each group. Participants in the Instructed Extinction, Experiential Extinction and Fear Acquisition-only group showed no differential ratings (F<1). These results suggest that there are no differences between approaching and avoiding the stimuli and that there are no differences between stimuli for participants.
17
(a) (b)
(c)
Figure 4. Mean median RT in ms for (a) Experiential Extinction, (b) Instructed Extinction, and (c)
Fear Acquisition-only groups.
Discussion
We tested whether providing instructions about CS-US contingencies would result in stronger fear extinction compared to a typical fear extinction procedure, in which no instructions are given. Our dependent variables referred to all three indices of fear (i.e., subjective experience, physiological arousal and action tendencies). Results showed that instructions led to a partial reduction of fear responses. The fear reduction was partial when considering the different indices of emotion; instructions only led to a reduction of the fear response for subjective apprehension and physiological arousal and not for action tendencies. The reduction in fear response found for
18 instructions was similar to the fear reduction found for Extinction learning alone. Taken together, in the current study instructions resulted in similar results as Extinction learning alone. However, those results should be interpreted cautiously as no differences in CS differentiation were found for the Fear Acquisition-only group as well. Although our design was similar to previous studies (Sevenster et al, 2012), the design could have caused uncertainty among participants in the Fear Acquisition-only group for the passage of time resulting in no differences between the groups and the manipulation of control. Nevertheless, participants did acquire fear learning and in turn showed depletion in differentiation between the CS+ and the CS- on the various indices.
The findings in this study cannot easily be explained by emotion theories (Mauss & Robinson, 2009; Frijda, 2010) and we acknowledge that we go beyond the data but will still try to interpret the findings of this study. In line with previous studies (Sevenster et al., 2012) a dissociation between startle responses and SCR ratings was also found. Previous research has shown that SCR ratings are associated with non-specific increases in arousal during conditioning experiments
(Sevenster, et al., 2012). The association with non-specific increases suggest that SCR ratings mirror current states of arousal and is triggered by anticipatory arousal and contingency learning (Sevenster, et al., 2012). Thus it can be seen as the cognitive expression of fear memory. Startle responses can be seen as the emotional aspect of fear learning as it is the reflexive and automatic response to feared stimuli. The found dissociation might indicate the different processes underlying the processing of emotion, which is more complex than merely explained by emotion theories. Also, to date research has not been able to demonstrate that verbal instructions alone lead to differentiated responding for action tendencies for the fear-conditioning paradigm. Nevertheless, this study demonstrates that instructions can play an important role in extinction learning and as to how, should be further explored.
In the current study, partial reinforcement during the acquisition phase could have led to ambivalence among participants. The unreinforced trials during the extinction phase could have seemed as a continuance of the first phase leading to higher expectations of receiving an electric stimulus. In combination with the amount of extinction trials, 10 trials for each CS, partial reinforcement could have resulted in the found partial extinction. Although fear acquisition was successful, in future research reinforcing 7 out of 8 trials with fixed order, first trial unreinforced, should control for possible ambivalence among participants in order not to influence the findings of future research.
19 In addition, as with the partial reinforcement, the possible occurrence of extinction learning during the test phase and reaction time task of the study could be a potential limitation of the current study. Because differences between the control and the manipulation groups were not found for all indices, further analyses were done after data collection in order to assure extinction learning had not occurred during other phases than the extinction phase. Since participants received no reinforcement during the test phase, the trials in the test phase potentially could have functioned as extinction trials and potentially have led to not finding any results during the AAT. One way of testing for the extinction occurrence was comparing the first two trials of the Experiential Extinction group with the two test trials of the Instructed Extinction and Fear Acquisition-only group. As no differences were found during the additional analysis, it can be said that the test trails did not function as extinction trials and extinction learning did not occur during the test phase. Still, in future research possible occurrence of extinction learning during the different phases should be taken into
consideration and alterations should be made in order to control for the possibility.
All in all, our study contributes by increasing the understanding of instructions in extinction learning. Targeting conditioned fear responses through instructions could translate into an increased understanding of verbal instructions on extinction learning. Increased understanding of extinction learning may eventually lead to increased therapeutic efficacy for anxiety disorders through different forms of exposure therapy and eventually leading to improved therapies and prevalence rates for anxiety disorders.
20 References
Barlow, D. H. (2002). Anxiety and its disorders: The nature and treatment of anxiety and panic (2nd ed.). New
York, NY: Guilford.
Beckers, T., Krypotos, A. M., Boddez, Y., Effting, M. & Kindt, M. (2013). What’s wrong with fear conditioning? Biological Psychology, 92, 90 – 96.
Bouton, M. E. (2002). Context, Ambiguity, and unlearning: Sources of relapse after behavioral extinction. Biological Psychiatry, 52, 976 – 986.
Effting, M. & Kindt, M. (2007). Contextual control of human fear associations in a renewal paradigm. Behavioral Research and Therapy 45, 2002 – 2018.
Field, A. (2009). Discovering statistics using SPSS. London: Sage Publications Ltd.
Foa, E. B. & Kozak, M. J. (1986). Emotional processing of fear: Exposure to corrective information.
Psychological Bulletin, 99, 20 – 35.
Frijda, N. H. (2010). Impulsive action and motivation. Biological Psychology 84, 570 – 579.
Graaf, R. de, Have, M. ten, Gool, C. van, & Dorsselaer, S. van (2012). Prevalentie van psychische aandoeningen en trends van 1996 tot 2009; resultaten van NEMESIS-2. Tijdschrift voor
Psychiatrie, 54, 27 – 38.
Hartstra, E., Kühn, S., Verguts, T., & Brass, M. (2011). The implementation of verbal instructions: An fMRI study. Human Brain Mapping 32¸ 1811 – 1824.
Houwer, J. de, Beckers, T., Vandorpe, S. & Custers, R. (2005). Further evidence for the role of mode-independent short-term associations in spatial Simon effects. Perception & Psychophysics,
67, 659 – 666.
Hugdahl, K. (1978). Electrodermal conditioning to potentially phobic stimuli: Effects of instructed extinction. Behavior, Research & Therapy, 16, 315 – 321.
Hugdahl, K., & Ohman, A. (1977). Effects of instruction on acquisition and extinction of electrodermal responses to fear-relevant stimuli. Journal of Experimental Psychology: Human
Learning and Memory, 3, 608 – 618.
Krypotos, A. M., Effting, M., Arnaudova, I., Kindt, M. & Beckers, T. (2014). Avoided by
Association: Acquisition, extinction, and renewal of avoidance tendencies toward conditioned fear stimuli. Clinical Psychological Science, 2, 336 - 343.
Liefooghe, B. & Wenke, D. (2012). Instruction-based task-rule congruency effects. Journal of
21 Lipp, O. V. & Edwards, M. S. (2002). Effect of instructed extinction on verbal and autonomic
indices of Pavlovian learning with fear-relevant and fear-irrelevant conditional stimuli. Journal f
Psychophysiology 16, 176 – 186.
Lovibond, P. F., Mitchell, C. J., Minard, E., Brady, A. & Menzies, R. G. (2009). Safety behaviors preserve threat beliefs: Protection from extinction of human fear conditioning by an avoidance response. Behavioral Research and Therapy, 47, 716 – 720.
Mauss, I. B., & Robinson, M. D. (2009). Measures of emotion: A review. Cognition and Emotion, 23, 209 – 237.
Mineka, S. & Zinbarg, R. (2006). A contemporary learning theory perspective on the etiology of anxiety disorders: It’s not what you thought it was. American Psychologist, 61, 10 – 26.
Olsson, A. & Phelps, E. A. (2004). Learned fear of “Unseen” faces after Pavlovian, observational, and instructed fear. American Psychological Society 15, 822 – 828.
Pavlov, I. P. (1927). Conditioned reflexes: An investigation of the physiological activity of the cerebral cortex. Oxford: Oxford University Press.
Peterson, R. A. & Reiss, S. (1992). Anxiety Sensitivity Index manual. Worthington: International Diagnostic System.
Schulz, D. H., & Helmstetter, F. J. (2010). Classical conditioning of autonomic fear responses is independent of contingency awareness. Journal of Experimental Psychology: Animal Behavioral
Processes, 36, 495 – 500.
Sevenster, D., Beckers, T. & Kindt, M. (2012). Instructed extinction differentially affects the emotional and cognitive expression of associative fear memory. Psychophysiology, 49, 1426 – 1435.
Soeter, M. & Kindt, M. (2010). Dissociating response systems: Erasing fear from memory.
Neurobiology of Learning and Memory, 94, 30 – 41.
Spielberger, C. D., Gorsuch, R. L. & Lusthene, R. E. (1970). Manual for the State-Trait Anxiety Inventory. Palo alto, CA: Consultin.
