ارزیابی مجدد شناختی و پذیرش: مقایسه تجربی از دو استراتژی تنظیم احساسات

Abstract The purpose of the present study was to compare the effect of cognitive reappraisal and acceptance on subjective distress, physiological reactions and behavioral avoidance in relation to aversive emotional states elicited by film-clips. Ninety-four participants were randomized to one of three groups. The Reappraisal group was instructed to think about what they saw in a way that minimized negative emotional reactions, the Acceptance group was told to let their feelings come and go without trying to control or avoid them, while the Watch (control) group was told just to watch the film-clips. Compared to the control condition, both reappraisal and acceptance led to significant reductions of subjective distress, physiological reactions associated with aversive emotions and behavioral avoidance. On the three types of measures there were few significant differences between the Reappraisal and Acceptance groups, but when such differences existed they were to the benefit of the Reappraisal condition. In the reappraisal condition there was however a positive correlation between elicited aversive emotion and avoidance, while no such correlation existed in the acceptance condition. The results are interpreted and discussed in relation to the theories underlying reappraisal and acceptance as well as the conceptual framework for emotion regulation established by Gross (2007).

Introduction The empirical study of emotion regulation has been a growing research area the last decade (Rottenberg & Gross, 2007) and the field has seen an increasing number of both experimental (e.g. Eifert and Heffner, 2003, Feldner et al., 2003, Gross, 1998, Hofmann et al., 2009 and Liverant et al., 2008) and individual differences studies (e.g. Gross and John, 2003, Kashdan et al., 2006 and Stewart et al., 2002) regarding the process and outcome of different strategies for emotion regulation. In parallel with this development clinical psychology, and especially the field of cognitive and behavioral therapies, has seen a rising discussion regarding the clinical usefulness of strategies focusing on changing versus accepting experiential content as a process of change in psychotherapy (e.g. Arch and Craske, 2008, Clark, 1999, Hayes, 2008, Hayes et al., 1999 and Mathews, 2006). In traditional cognitive behavioral therapies cognitive restructuring, i.e. modifying dysfunctional cognitions that contribute to psychological distress, is a corner stone of the therapeutic process (Beck et al., 1979 and Clark, 1999) and is supposed to be a central mediator of the therapeutic outcome (Clark, 1999 and Wishman, 1993). In contrast, facilitating acceptance of experiential content is a central part of the therapeutic process in the so called “third wave of behavioral therapies” (Hayes, Folette, & Linehan, 2004) and most notably so in Acceptance and Commitment Therapy (Hayes et al., 1999). Acceptance in this context is defined as an approach that encourages individuals to experience their emotions, thoughts, and bodily sensations fully without trying to change, control, or avoid them, entailing openness to internal experiences and willingness to remain in contact with those experiences even if they are uncomfortable (Hayes et al., 1999). When comparing psychotherapeutic paradigms, a complimentary approach to randomized controlled trials of broad treatment protocols is to conduct what Hayes, Luoma, Bond, Masuda, and Lillis (2006) called “micro studies”, i.e. experimental studies on key processes of different therapies to see if they are psychologically active and work in a fashion that accords with theory. The present study seeks to provide such an experimental comparison of the effects of cognitive reappraisal and acceptance on subjective distress, avoidance tendencies, and physiological reactions in relation to aversive emotional states. This is done in the context of Gross, 1998 and Gross, 2007 model of emotion regulation, where emotion regulation is defined as attempts individuals make to influence which emotions they have, when they have them and how these emotions are experienced and expressed (Gross, 1998 and Sloan and Kring, 2007). In this temporal model of emotion regulation which broadly distinguishes between antecedent focused and response focused emotion regulation. Antecedent focused emotion regulation involves attempts to alter emotional experiences before the emotion is fully generated; while response focused strategies involve attempts to alter emotional responding after the emotion has been generated. Cognitive reappraisal is a form of antecedent focused emotion regulation that involves cognitively construing a potentially emotion-eliciting situation in a way that changes its emotional impact (Gross & John, 2003). In an experimental study cognitive reappraisal was found to decrease both experiential and physiological expression of negative emotion compared to the response focused strategy of emotional response suppression (Gross, 1998). Compared to a neutral control condition, cognitive reappraisal was found to be associated with less experienced negative emotion, less expressed emotion and a non-significant tendency toward reduction of physiological responding. When conceptualized as an emotion regulation strategy within the framework established by Gross, 1998 and Gross, 2007 acceptance is most logically construed as a response focused strategy aimed at allowing the experience of emotion without attempts to alter or suppress it (Hofmann & Asmundson, 2008). Acceptance may also however be said to contain an antecedent focused cognitive change component regarding the acceptability of an emerging emotional experience (Liverant et al., 2008). In experimental research on the consequences of emotional acceptance (e.g. Campbell-Sills et al., 2006a, Campbell-Sills et al., 2006b, Eifert and Heffner, 2003, Feldner et al., 2003 and Levitt et al., 2004) it has been shown that acceptance is associated with experiencing less fear, catastrophic thoughts, avoidance behavior and better recovery from negative affect as compared to suppression. Interestingly, many of the experimental studies made on acceptance as emotion regulation strategy indicate that, in comparison to people with low levels of emotional acceptance, people with high levels do not experience less physiological arousal or bodily sensations related to the elicited reaction, but they report lower levels of subjective distress (Eifert and Heffner, 2003, Feldner et al., 2003 and Karekla et al., 2004). These findings indicate that acceptance is more related to how bodily arousal or other forms of physiological emotional responses are experienced and evaluated rather than how they actually occur, and thus support the supposition that acceptance is primarily a response focused strategy of emotion regulation. To our knowledge, only two previous studies have experimentally compared the effects of cognitive reappraisal and acceptance. Hofmann et al. (2009) studied the effects on anxious arousal of using acceptance, reappraisal or suppression in a situation where the participants were asked to give an impromptu speech in front of a video camera. The results suggested that that both reappraisal and acceptance strategies were more effective than suppression for moderating the physiological arousal while the reappraisal strategy was more effective for moderating subjective feelings of anxiety than attempts to suppress or accept the emotional experience. The authors of the study remarked however that the differences overall were small and that a more potent stimulus for emotion elicitation might have been used (Hofmann et al., 2009). None the less the study suggests that cognitive reappraisal generally was the most adaptive strategy in terms of reducing anxiety responses. Furthermore when it comes to acceptance, the study suggest an outcome pattern that somewhat contrasts with what has been found in other experimental studies of acceptance, in that the most obvious adaptive effect were found for the physiological outcome measure and not the subjective, thus to some extent strengthening the supposition that acceptance also has antecedent focused components. The second study (Szasz, Szentagotai, & Hofmann, 2011) compared the effects of reappraisal, acceptance and suppression on anger and frustration tolerance. In this study as well, reappraisal was found to have the most adaptive effects in that it led to lower self-reported anger and higher frustration tolerance than the other strategies. As far as we know, none of the studies on emotion regulation where participants have been randomly assigned to use a certain regulatory strategy have studied whether individual differences in habitual tendencies to regulate emotions in one way or another moderates the effects of the experimental manipulation, though Hofmann et al. (2009) suggested the need for future studies to examine the role of individual differences in this regard. No such effect would be expected if the different strategies are readily accessible to all and used repeatedly in everyday life, even if the relative extent to which we use them varies across individuals. If on the other hand the different strategies are viewed as skills or abilities that we need training to learn effectively or reflect more stable and rigid approaches to emotional responses, one might suspect that individual differences would moderate the effects of the experimental instructions in that participants would find it more easy to use a strategy they habitually use. The studies referred to above comparing groups defined by scoring high or low on experiential avoidance (Eifert and Heffner, 2003, Feldner et al., 2003 and Karekla et al., 2004) suggest that individual differences do affect the responses to experimental elicitations of emotional reactions. The construct of ‘avoidance’ refers to refraining from, or escaping from, an action, person or thing ( Ottenbreit & Dobson, 2004). Behavioral or situational avoidance is an important aspect in many emotional disorders, both with regard to descriptive diagnostic criteria ( American Psychiatric Association, 1994), and as an important maintaining factor in the pathological condition (e.g. Barlow, 2008 and Martell et al., 2001). Accordingly, both cognitive reappraisal and acceptance have distinct and important relationships to behavioral avoidance, especially within the frameworks of traditional CBT and ACT. In traditional CBT, an important aspect of working with cognitive reappraisal is to identify how a particular cognitive process (for example an automatic thought) in a given situation affects our subsequent emotions and behaviors and to develop and test alternative cognitions that lead to other behavioral response patterns (Beck et al., 1979). Behavioral avoidance is thus seen as an important target for therapeutic interventions, but the effect is assumed to occur indirectly, as the result of altered cognitive processes. From this perspective one would expect a causal chain from altered appraisals to reduced negative emotional reactions which in turn lead to reduced behavioral avoidance. Thus the default association between emotional distress and behavioral avoidance, that we tend to avoid situations or stimuli that we associate with aversive emotions, is likely to be maintained when using reappraisal and a reduction in avoidance would be achieved mainly by the fact that we change our cognitive and emotional reaction to the emotionally relevant situations or stimuli. By acceptance as construed within ACT on the other hand, behavioral avoidance is more directly addressed. In fact, one of the main purposes of establishing experiential acceptance is to alter the social or verbal context in such a way that our behavior is less regulated by aversive private events (Hayes, 2008). If such a strategy is employed and works according to the underlying theory, one would expect a deviation from the default association between behavioral avoidance and aversive emotions, in that the association of aversive emotional reactions to a given stimulus or situation and the tendency to avoid this stimulus or situation would be significantly weakened. So where cognitive reappraisal might be conceptualized as seeking to affect avoidant behavior patterns by altering the antecedent cognitions and emotions that give rise to those behaviors, acceptance seeks to weaken the very connection between aversive emotions and behavioral avoidance. If such a difference in the association of emotional reactions and avoidance tendencies between the two strategies can be empirically detected in an experimental context has, to our knowledge, never been tested and was a part of the purpose of the present study. The present study: purpose and hypotheses The purpose of this study was to experimentally compare the experiential and physiological consequences of cognitive reappraisal and acceptance as emotion regulation strategies with regard to aversive emotions elicited by film-clips and how the different emotion regulation strategies influenced tendencies of behavioral avoidance. The present study also sought to investigate whether individual differences in the habitual use of cognitive reappraisal and acceptance influenced the effects of the experimental instructions to use these strategies in regulating aversive emotions. Based on theory and previous research (Hofmann et al., 2009 and Szasz et al., 2011), we hypothesized (1) that both cognitive reappraisal and acceptance would lead to reductions in the subjective distress produced by aversive film-clips as compared to the control condition, but with cognitive reappraisal having a significantly larger effect. We also expected (2) that both strategies would lead to lower scores on a rudimentary measure of behavioral avoidance when compared to the control group, but that they would differ in terms of “the default association between emotional distress and avoidance”, cognitive reappraisal maintaining it, and acceptance deviating from it. That is, we expected (3) that participants in the acceptance condition would show a relative absence of correlation between change in subjective distress and avoidance, whereas there would be a significant correlation between these two measures in the cognitive reappraisal and watch condition. With regard to the physiological measures we contrasted two rival hypotheses: The first hypothesis (4a), in line with seeing acceptance as a purely response focused emotion regulation strategy (Hofmann & Asmundson, 2008), was that cognitive reappraisal, but not acceptance, would lead to a reduced physiological reactivity, in comparison to the control condition. The other hypothesis (4b), in line with seeing acceptance as additionally containing an antecedent focused cognitive change component (Liverant et al., 2008), was that both cognitive reappraisal and acceptance would lead to reductions in physiological reactivity compared to the control condition, but with cognitive reappraisal having a significantly larger effect. Finally, with regard to the effects of individual differences in habitual use of cognitive reappraisal and acceptance, previous research (e.g., Eifert and Heffner, 2003 and Feldner et al., 2003) led us to expect (5a) that individuals would have greater effects when using the emotion regulation strategy they habitually use, both in the cognitive appraisal and the acceptance condition, and (5b) that habitual use of these strategies in the control condition would be associated with less elicited aversive emotion and less avoidance. Three different film-clips intended to elicit different emotional reactions were included. This was mainly done to get a broader test of the above listed hypotheses, and we had no prior hypotheses regarding different effects of the emotion regulation strategies across the different film-clips.

Results Pre-experiment group differences In order to get a measure of whether the random assignment procedure had worked as intended, the groups were compared on demographic characteristics, pre-film affective status and pre-film status on habitual use of emotion regulation strategies. No between group differences were apparent in gender, χ2 (2, N = 94) = .66, p = .72 or age, F (2, 91) = .810, p = .45. Furthermore, the groups did not differ on negative affect (PANAS-N) scores, F (2, 91) = 1.01, p = .37 or habitual use of reappraisal, F (2, 91) = 1.16, p = .32 and acceptance, F (2, 91) = .019, p = .98. Manipulation checks The manipulation checks suggested that participants in the experimental conditions understood the instructions and were able to successfully distinguish between approaches of acceptance and reappraisal: When asked questions about how to relate to the content of the film-clips and elicited emotions, the participants in the acceptance condition to a significantly larger extent endorsed statements describing acceptance strategies (t (61) = 12.95, p < .01) while participants in the reappraisal condition to a larger extent endorsed statements describing reappraisal strategies (t (61) = −17.96, p < .01). Furthermore, participants in the experimental conditions did not differ significantly in how successful they reported themselves to be in applying the instructions after each of the emotion elicitation episodes (Disgust: t (61) = 1.36, p = .21; Sadness: t (61) = .12, p = .90; Fear: t (61) = −1.98, p = .054). Hypothesis 1: effects of reappraisal and acceptance on self-reported aversive emotion Means and standard deviations for self-reported aversive emotion and avoidance tendencies for the experimental conditions are presented in Table 1. Table 1. Means and standard deviations for self-reported aversive emotion and avoidance tendencies. Measure Condition Reappraise (N = 31) Accept (N = 32) Watch (N = 31) M (SD) M (SD) M (SD) PANAS-N Pre-film 14.71 (2.61) 15.09 (3.74) 14.00 (2.77) Disgust 19.26 (6.97) 24.13 (6.79) 26.38 (7.74) Sadness 20.90 (6.22) 21.41 (4.52) 26.39 (6.01) Fear 22.13 (7.99) 22.81 (6.37) 30.68 (8.79) Avoidance Disgust 1.97 (1.14) 1.94 (1.11) 3.84 (1.44) Sadness 1.55 (.81) 1.84 (1.27) 2.52 (.99) Fear 1.55 (1.06) 1.47 (.62) 3.55 (1.46) Table options Group differences in self-reported aversive emotion across the film-clips were examined using a 3 (Film-clip) × 3 (Group) mixed model ANCOVA with scores at baseline as covariate and film-clip scores as repeated measure. The α-level was Bonferroni corrected to adjust for the four omnibus tests performed in the study, rendering a critical α-level of .013. The omnibus test revealed a significant between subjects effect of Group (F (2, 90) = 16.35, p < .01, η2 = .27) while the within-subjects effect of Film-clip was non-significant (F (2,180) = 1.39, p = .25, η2 = .02). The interaction effect of Group × Film-clip was significant but small in comparison to the Group effect (F (4,180) = 4.59, p < .01, η2 = .08). Post-hoc Sheffé corrected pairwise comparisons for each film-clip with scores at baseline as covariate (univariate ANCOVAs), revealed that the participants in the watch condition reported significantly greater experiences of negative emotion compared to participants in the reappraisal condition to all three film-clips (Disgust film-clip: Mean difference = 7.13, p < .01; Fear film-clip: Mean difference = 8.55, p < .01; Sadness Film-clip: Mean difference = 5.48, p < .01). The same results were found when comparing the watch and acceptance groups on the Fear (Mean difference = 7.86, p < .01) and Sadness (Mean difference = 4.98, p < .01) film-clips. For the Disgust film-clip however, the difference between the watch and acceptance group was non-significant (Mean difference = 2.26, p = .25). Finally, participants in the acceptance condition reported significantly greater experiences of negative emotion compared to participants in the reappraisal condition to the Disgust film-clip (Mean difference = 4.87, p < .01), while there were no significant differences for the other film-clips (Fear film-clip: Mean difference = .68, p = .92; Sadness film-clip: Mean difference = .50, p = .89). Thus, in sum, the first part of Hypothesis 1, that participants using cognitive reappraisal and acceptance would report less experiences of negative emotion when compared to the control condition, was supported (except when comparing acceptance and watch on the Disgust film-clip). The second part of the hypothesis, that the effect would be larger for the participants using reappraisal however was only supported for the Disgust film-clip. Hypothesis 2: effects of reappraisal and acceptance on avoidance Group differences in avoidance tendencies across the film-clips (see Table 1) were examined using a 3 (Film-clip) × 3 (Group) mixed model with avoidance score for each film-clip as repeated measure. Once again, the α-level was Bonferroni corrected to adjust for the four omnibus tests performed in the study, rendering a critical α-level of .013. The omnibus test revealed a significant between subjects effect of Group (F (2, 91) = 38.00, p < .01, η2 = .46) and a significant within-subjects effect of Film-clip (F (2,182) = 14.77, p < .01, η2 = .14). The interaction effect of Group × Film-clip was significant but small when compared to the Group effect (F (4,182) = 3.27, p = .01, η2 = .07). Post-hoc Sheffé corrected pairwise comparisons for each film-clip revealed that the participants in the Watch condition reported significantly greater avoidance tendencies than participants in the reappraisal condition for all three film-clips (Disgust film-clip: Mean difference = 1.87, p < .01; Fear film-clip: Mean difference = 2.00, p < .01; Sadness Film-clip: Mean difference = .97, p < .01). The same results were found when comparing the Watch and Acceptance conditions (Disgust film-clip: Mean difference = 1.90, p < .01; Fear film-clip: Mean difference = 1.71, p < .01; Sadness Film-clip: Mean difference = 1.05, p < .01). No significant differences in avoidance tendencies were found between participants in the reappraisal and acceptance conditions. In sum, the hypothesis (2) that cognitive reappraisal and acceptance would lead to reductions in behavioral avoidance when compared to the control condition was supported for each film-clip. Hypothesis 3: relationship between self-reported negative emotion and avoidance To examine Hypothesis 3, that participants in the acceptance condition would show a weaker connection between elicited subjective distress and avoidance than those in the cognitive reappraisal and watch conditions, bivariate correlations were computed between avoidance tendencies and change in negative emotion from baseline to emotion elicitation. As can be seen in Table 2, the performed analysis revealed significant positive correlations between induced negative emotion and avoidance tendencies for each film-clip in the reappraisal and watch conditions, but for none of the film-clips in the acceptance condition. Using Fisher r-to-z transformation to test if the differences between the correlation coefficients in the acceptance and reappraisal conditions were significant, showed significant differences between the coefficients for all three film-clips (Disgust: z = 4.35, p < .01; Fear: z = 2.14, p = .03; Sadness: z = 2.30, p = .02). In sum, the hypothesis (3) that the association between elicited aversive emotion and avoidance tendencies would be weaker in the acceptance than in the reappraisal condition was supported. Table 2. Correlations of avoidance tendencies and changes in self-reported aversive emotions. Measure Condition Reappraise (N = 31) Accept (N = 32) Watch (N = 31) r p r p r p Avoidance × ΔPANAS-N disgusta .67 <.01 −.33 .07 .70 <.01 Avoidance × ΔPANAS-N feara .61 <.01 .14 .18 .80 <.01 Avoidance × ΔPANAS-N sadnessa .55 .03 .01 .98 .58 .01 a Change in PANAS-N score from baseline to after emotion elicitation. Table options Hypothesis 4: effects of reappraisal and acceptance on physiological responses Means and standard deviations for each physiological response are presented in Table 3. Table 3. Means and standard deviations for physiological responses. Measure Condition Reappraise (N = 31) Accept (N = 32) Watch (N = 31) M (SD) M (SD) M (SD) Disgust film-clip Corrugator superciliia Baseline 162.16 (77.09) 184.44 (105.82) 204.79 (227.57) Elicitation 180.91 (90.22) 218.63 (128.91) 386.58 (286.33) Skin conductance level Baseline 314.25 (310.82) 307.41 (176.56) 340.04 (233.00) Elicitation 294.22 (192.65) 313.01 (180.43) 450.70 (270.26) Fear film-clip Corrugator superciliia Baseline 192.25 (272.92) 202.54 (132.45) 190.75 (272.92) Elicitation 202.85 (99.38) 208.04 (130.61) 305.69 (227.58) Skin conductance level Baseline 299.31 (179.15) 305.54 (212.12) 333.34 (287.89) Elicitation 356.10 (175.64) 356.89 (208.23) 592.04 (291.05) Sadness film-clip Corrugator superciliia Baseline 214.88 (162.51) 192.80 (101.57) 211.75 (245.74) Elicitation 203.72 (91.44) 216.38 (115.76) 329.16 (193.54) Skin conductance level Baseline 228.38 (130.44) 278.16 (192.64) 302.11 (200.66) Elicitation 237.34 (133.74) 309.80 (228.35) 408.91 (211.14) a The mean values refer to the root mean square (rms). Table options Two rival hypotheses were contrasted: The first hypothesis (4a) was that cognitive reappraisal, but not acceptance, would lead to a reduced physiological reactivity, in comparison to the control condition. The other hypothesis (4b) was that both cognitive reappraisal and acceptance would lead to reduction in physiological reactivity compared to the control condition, but with cognitive reappraisal having a significantly larger effect. To test these hypothesis two 3 (Film-clip) × 3 (Group) mixed models ANCOVAs were performed, one with Corrugator EMG and the other with skin conductance level as repeatedly measured dependant variable. In both analyses scores at baseline were used as covariates. The α-level was Bonferroni corrected to adjust for the four omnibus tests performed in the study, rendering a critical α-level of .013. For Corrugator EMG a significant between subjects effect of Group was found (F (2, 88) = 51.81, p < .01, η2 = .54) while the within-subjects effect of Film-clip (F (2, 176) = 3.54, p = .03, η2 = .01) and the interaction effect of Group × Film-clip (F (4,176) = 3.00, p = .02, η2 = .12) was non-significant at the adjusted α-level. Post-hoc Sheffé corrected pairwise comparisons for each film-clip with scores at baseline as covariate (univariate ANCOVAs), revealed significantly greater Corrugator EMG in the watch than in the reappraisal condition for all film-clips (Disgust film-clip: Mean difference = 263.72, p < .01; Fear film-clip: Mean difference = 102.84, p < .01; Sadness Film-clip: Mean difference = 177.05, p < .01). The same results were found when comparing the watch and acceptance conditions (Disgust film-clip: Mean difference = 226.01, p < .01; Fear film-clip: Mean difference = 97.65, p < .01; Sadness Film-clip: Mean difference = 164.39, p < .01). Finally, corrugator EMG was significantly greater in the acceptance condition compared to the reappraisal condition for the Disgust film-clip (Mean difference = 37.71, p = .01), while there were no significant differences for the other film-clips (Fear film-clip: Mean difference = 5.20, p = .89; Sadness film-clip: Mean difference = 12.66, p = .64). For skin conductance level there was a significant between subjects effect of Group (F (2, 88) = 44.98, p < .01, η2 = .51) and a significant within-subjects effect of Film-clip (F (2, 176) = 8.26, p < .01, η2 = .09) while the interaction effect of Film-clip × Group (F (4, 176) = 1.03, p = .40, η2 = .02) was non-significant. Post-hoc Sheffé corrected pairwise comparisons for each film-clip with scores at baseline as covariate (univariate ANCOVAs), revealed significantly greater skin conductance level in the watch than in the reappraisal condition for all film-clips (Disgust film-clip: Mean difference = 201.64, p < .01; Fear film-clip: Mean difference = 235.94, p < .01; Sadness Film-clip: Mean difference = 232.86, p < .01). The same results were found when comparing the watch and acceptance conditions (Disgust film-clip: Mean difference = 182.85, p < .01; Fear film-clip: Mean difference = 235.14, p < .01; Sadness Film-clip: Mean difference = 160.40, p < .01). Finally, skin conductance level was significantly greater in the acceptance condition compared to the reappraisal condition for the Disgust (Mean difference = 18.79, p = .02) and Sadness (Mean difference = 72.46, p < .01) film-clips, while there was no significant difference for the Fear film-clip (Mean difference = .79, p = 1.00). Hypothesis 5: effects of the habitual use of reappraisal and acceptance To examine whether the participants’ habitual use of acceptance and reappraisal based emotion regulation strategies influenced the impact of the experimental manipulation or in other ways had an effect on their performance in the experiment, three kinds of analyses were performed: initially, the repeated measures ANOVAs were rerun using habitual use of cognitive reappraisal and habitual tendencies of experiential avoidance as covariates. These analyses did not change the results from the ANOVAs without covariates, indicating the effects of the experimental manipulation remained when controlling for individual differences. Secondly, the analyses of the correlation between elicited aversive emotion and avoidance tendencies were rerun, but this time controlling for habitual use of reappraisal and experiential avoidance through calculating partial correlations. The results were the same as before, showing no effect of controlling for habitual use of reappraisal and experiential avoidance. In the last analysis the correlation between habitual use of reappraisal and experiential avoidance, elicited emotion and avoidance tendencies were investigated for each of the three conditions separately. These analyses showed no significant correlation between the variables in the reappraisal and acceptance conditions. In the control (watch) condition however, there were significant negative correlations between habitual use of reappraisal and elicited aversive emotion for the Fear (r = −.40, p = .026), Disgust (r = −.41, p = .023) and Sadness (r = −.36, p = .048) film-clips. In this condition there was also a significant negative correlation between habitual use of reappraisal and avoidance tendencies for the Disgust film-clip (r = −.46, p = .01) but not for the other film-clips. In the same condition there were significant positive correlations between experiential avoidance and induced negative emotion from the Fear (r = .38, p = .04), Disgust (r = .52, p < .01) and Sadness (r = .52, p < .01) film-clips. There were also significant positive correlations between experiential avoidance and avoidance tendencies for all the film-clips (Fear: r = .44, p = .01; Disgust: r = .54, p < .01; Sadness: r = .40, p = .03). To sum up the findings from this part of the study, the first part of our hypothesis (5a) was not supported, since differences in habitual tendencies to use reappraisal and acceptance did not interact with the effects of the experimental manipulation. The second part of our hypothesis however, that there would be effects of individual differences among participants in the control condition (5b), was supported.