ارزیابی مجدد (اما نه حواس پرتی) مایه شرمساری: شواهد فیزیولوژیکی برای تلاش خود کنترلی

Abstract Previous studies of emotion regulation suggested that reappraisal (construing an emotional event in non-emotional terms) has no cognitive or physiological consequences, but in all these studies, reappraisal was instructed ahead of an emotional situation. The authors' recent work, using behavioral indices, showed that inhibitory self-control resources are challenged when reappraisal starts late during an emotional situation relative to late instruction of distraction (diverting attention through producing neutral thoughts). The present study provides converging physiological evidence in showing that instructing to use reappraisal but not distraction late in a sadness inducing film involved increased skin conductance and decreased finger temperature. Both of these results are indicative of increased sympathetic activation that has been previously found to accompany inhibitory self-control effort.

1. Introduction Emotion regulation is an important cognitive skill that has substantial implications to one's interpersonal conduct, well being, coping and appropriate functioning in general (see Gross, 1998a, for a review). For this reason, people may be struggling with the question which emotion regulatory strategy would prove most useful in each context. However, providing an answer to such question may not be simple since the same strategy that was proven effective in one context may not be effective in another context. In the present study we created one such context that poses a substantial challenge to the effectiveness of regulation strategies and as such is likely to involve physiological effort. Specifically, in the following sections we show that previous studies have established that reappraisal—construing an emotional situation in non-emotional terms—has no physiological or cognitive costs (see Gross, 2002 and Richards, 2004, for reviews). However, in a recent behavioral study, we showed (Sheppes and Meiran, in press) that initiating reappraisal late during a sadness mood induction increased the regulatory challenge, and resulted in the expenditure of inhibitory self-control resources relative to late distraction (diverting attention away from an emotional situation via producing neutral thoughts. e.g. Nolen-Hoeksema, 1991). In the present study, we sought to provide converging physiological evidence for the increased inhibitory self-control effort that we found for late initiated reappraisal but not for distraction. 1.1. Does reappraisal consume inhibitory self-control resources? Two lines of evidence have established that reappraisal does not involve an expenditure of inhibitory self-control resources as reflected in behavioral and physiological measures. First, several behavioral studies were inspired by the ego depletion theory, which views self-control as a limited resource which gets depleted when one tries to inhibit competing behaviors, urges or desires (see Muraven and Baumeister, 2000, for review). According to this theory, the exertion of self-control appears to depend on a limited resource. Just as a muscle gets tired after performing an effortful action, an initial act of an inhibitory self-control task causes impairments (ego depletion) in the performance of a subsequent self-control task. Applying this logic, it was shown that initiating reappraisal did not result in ego depletion (e.g., Baumeister et al., 2007 and Vohs and Schmeichel, 2003). Second, another study showed that as opposed to the increased inhibitory-related sympathetic activation found in suppression of facial expressions, reappraisal showed a physiological response profile that was not different from that of a control condition (Gross, 1998b). The basic argument for reappraisal's lack of inhibitory self-control cost is based on Gross's (1998a) process model of emotion regulation. In that model, reappraisal is considered an antecedent focused emotion regulation strategy, which is initiated early, before emotional response tendencies are fully activated. Such an early initiation diverts the emotional trajectory before the emotional response is fully blown, and hence does not tax self-control resources. Note that in all of these studies, reappraisal was indeed instructed very early during the emotion generative process (at the mood induction onset), making the inhibitory self-control challenge minimal. Indeed, we agree that in this case, reappraisal is what Gross, 1998a and Gross, 2001 defines as “an antecedent focused strategy”, changing the emotional trajectory early on and consequently at a minimum cost. However, in a previous work, we showed that not all forms of reappraisal come free of charge. Specifically, we introduced a new form of emotion regulation defined as online regulation—the attempt to change an emotion which starts and continuously operates during an emotional situation ( Sheppes and Meiran, 2007). In that study, we tested distraction and reappraisal in two manners. When we tested both strategies as antecedent (initiated at the mood induction onset) we replicated previous findings in showing that reappraisal and distraction were equally effective. However, initiating both strategies late (during the mood induction presumably after the emotion response has sufficiently evolved) revealed that reappraisal was less effective than distraction in down regulating the sad emotion. In a following study, designed to test the behavioral origin of this effect, we adopted a classical ego depletion procedure. We found that initiating reappraisal late (relative to late distraction), resulted in an expenditure of self-control resources, as reflected in a subsequent increased Stroop interference effect (a task involving inhibition, Sheppes and Meiran, in press). We suggested that initiating reappraisal late in an emotional situation may pose a high inhibitory self-control challenge, because it requires overcoming a strong tendency of identifying with the emotional content which had a chance to be well established prior to the late strategy initiation. Accordingly, applying reappraisal late involves using self-control resources as one has to stop and override the strong previous interpretation when transforming it to a neutral interpretation. By contrast, we showed that late distraction does not involve consuming self-control resources since it entails diverting attention away from the emotional situation and its contents by producing independent neutral contents. This feature of distraction was observed in memory decrements of the emotional situation once distraction was initiated; indicating reduced encoding of the emotional situation (Sheppes and Meiran, 2007 and Sheppes and Meiran, in press). Notice that this inhibitory self-control effort that we found for reappraisal is relatively indirect because we examined the delayed effects of self-control effort (the depletion of self-control resources as seen in subsequent Stroop performance), but did not yet show evidence for self-control challenge as it occurs online. One of the most straightforward ways to examine online processes as they evolve is via physiological measures. Several researchers argued that the cognitive effort that accompanies self-control demand in general and inhibition/suppression in particular is reflected in increased sympathetic activation, specifically in an increase in skin conductance level (SCL) and also to less extent in a decrease in finger temperature (FT). Below we provide a short review of this topic. 1.2. Inhibition, self-control effort and their relation to increased sympathetic activation One of the early demonstrations of the relation between SCL and inhibitory self-control effort was obtained by Elliott et al. (1970). These authors found the increase in SCL to be a reliable measure that positively correlated with increased inhibitory demand in the Stroop task. Two other studies have shown that SCL increased both when participants refrained from telling the truth in a guilty knowledge test (Pennebaker and Chew, 1985), and when participants avoided talking about personal and traumatic events (Pennebaker et al., 1987). SCL was also measured in studies that examined the inhibitory processes occurring during the suppression of emotional thoughts (e.g., Wegner and Gold, 1995 and Wegner et al., 1990). For example, Wegner and Gold (1995) investigated the effects of suppression of thoughts concerning a significant past romantic relationship. Participants had to either express or suppress thoughts regarding a desired past relationship (a high emotional difficult condition) or a no longer desired past relationship (a low emotional easy condition). Results for the expression and suppression groups in the desired relationship condition showed that those who were instructed to suppress their thought had indeed thought less about the desired past relationship but showed increased SCL relative to the group who were instructed to express their thoughts and feelings. Note that there were no SCL differences between express and suppress groups in the non-desired past relationship condition indicating that only taxing suppression results in SCL increase. In addition, behavioral studies provide converging support for the interpretation that SCL rise observed in thought suppression may denote the expenditure of inhibitory self-control resources. Specifically, it was repeatedly shown that initiating thought suppression consumes self-control resources and leads to ego depletion (see Muraven and Baumeister, 2000, for a review). Last, several studies measured the physiological profile of expressive suppression, defined as inhibiting ongoing emotion-expressive facial behavior (Gross, 1998b, Gross, 2002, Gross and Levenson, 1993 and Gross and Levenson, 1997). It was hypothesized that because suppression involves behavioral inhibition it should result in increased sympathetic activation as reflected in the physiological response. The results supported the hypothesis in showing higher SCL and lower FT in films inducing disgust (Gross, 1998b and Gross and Levenson, 1993) and sadness (Gross and Levenson, 1997). Converging behavioral support for the interpretation that increased sympathetic activation may relate to the expenditure of inhibitory self-control resources in suppression comes from Inzlicht and Gutsell (2007) who have directly shown that expressive suppression depletes self-control resources. Specifically, suppressing one's emotions resulted in subsequent increased Stroop effect relative to a control condition. Last, note again that just like Wegner and Gold (1995),Gross and Levenson (1997) did not find an increase in sympathetic activation when participants suppressed their ongoing facial behavior to a non-emotional neutral film. This last finding indicates that increased sympathetic arousal in suppression is only observed when the inhibitory demand is strong. 1.3. The present study In the present study, we set out to provide physiological evidence for the differential inhibitory self-control demands associated with online cognitive reappraisal and online distraction. We examined this issue under conditions that presumably challenge inhibitory self-control abilities. To that end, we measured physiological responses associated with the autonomic system while participants were watching a sadness inducing film. According to the online emotion regulation paradigm, participants were randomly assigned to conditions by receiving subtitles containing the core instructions at a late point during this film. This procedure enabled us to measure the strategies' response profiles before and after the strategy initiation. Reappraisal's profile was compared to a control condition in which participants were allowing their feelings to rise, and to late distraction which has not been shown to involve the expenditure of inhibitory self-control resources. Moreover, by showing that online distraction does not result in increased sympathetic activation we could demonstrate the specificity of the effect to reappraisal rather than to online emotion regulation attempt in general. Our main hypothesis was that once late reappraisal is initiated (after the emotion-related response tendencies are sufficiently evolved), it would involve inhibitory self-control costs, reflected in an increase in SCL relative to a pre-regulation period, as compared to distraction and control conditions. In addition, some studies (e.g., Gross, 2002 and Richards and Gross, 1999) have also looked into the decrease in FT serving as an index of greater vasoconstriction and hence greater sympathetic activation. We therefore incorporated a measurement of FT into our design and predicted that reappraisal would show a decrease in FT once it is initiated relative to distraction and control conditions. Last as mentioned above, in our previous studies we found that when reappraisal was given a sufficiently long time to operate, it was as effective as distraction in reducing subjective negative experience (see Sheppes and Meiran, 2007 Experiment 2, in press). Because we incorporated the same late initiation condition in the present study, we predicted that both strategies would be effective in reducing subjective negative experience relative to a control condition. Obtaining this result was important to rule out an alternative explanation according to which the increase in sympathetic activation in reappraisal denotes an improvement in negative mood rather than an inhibitory effort.

3. Results 3.1. Mood check As predicted, distraction (M = 5.93, SD = 0.38) and reappraisal (M = 5.53, SD = 0.38) conditions were associated with lower levels of negative experience relative to the control condition (M = 7.11, SD = 0.39) F(1, 41) = 8.47, p < 0.01 with no significant differences between distraction and reappraisal F(1, 41) < 1. 3.2. SCL 3.2.1. Preliminary analyses We did not find any meaningful SCL differences between groups during the pre-instruction and post-instruction baselines (both Fs < 1). These results are important since they show that our random assignment of participants to groups has been successful, and that the instruction procedure (which was somewhat different between reappraisal and distraction) did not create changes between groups prior to the mood induction. 3.2.2. Core analyses Our main prediction was that reappraisal would show an increase in SCL during the regulation period relative to the pre-regulation period as compared to distraction and control. Indeed the contrast analysis supported our prediction, F(1, 41) = 5.0, p < 0.04. In another focused contrast, we found that Distraction and Control did not differ from one another F(1, 41) < 1 and both actually showed a decrease in SCLs during the regulation period relative to the pre-regulation period (see Fig. 1). Furthermore, the aforementioned difference between Reappraisal and the other groups remained significant even when we directly compared reappraisal to distraction F(1, 41) = 4.3, p < 0.05. Note that there were no significant SCL differences between reappraisal, distraction, and control groups during the pre-regulation period 4 (all Fs < 1). Mean SCL change scores from baseline of the pre-regulation and regulation ... Fig. 1. Mean SCL change scores from baseline of the pre-regulation and regulation periods for the three instructional groups. Figure options 3.3. FT 3.3.1. Preliminary analyses Similar to the SCL preliminary analyses, we did not find any meaningful FT differences between groups during the pre-instruction and post-instruction baselines (both Fs < 1). 3.3.2. Core analyses Our main prediction was that reappraisal would show a decrease in FTs during the regulation period relative to the pre-regulation period as compared to distraction and control. The contrast analysis supported our prediction, F(1, 41) = 4.17, p < 0.05. A similar focused contrast indicated no significant differences between Distraction and Control, F(1, 41) < 1 and both actually showed an increase in FTs during the regulation period relative to the pre-regulation period (see Fig. 2). Furthermore, the difference between Reappraisal and distraction was marginally significant even when these conditions were directly compared in a focused contrast, F(1, 41) = 3.18, p = 0.08 5. Mean FT change scores from baseline of the pre-regulation and regulation periods ... Fig. 2. Mean FT change scores from baseline of the pre-regulation and regulation periods for the three instructional groups. Note that the ordinate's scale is such that increased sympathetic activation is upward. Figure options As can be seen in Fig. 2, there were some observable differences between groups during the pre-regulation period. However, these differences were found to be non-reliable, All Fs < 1). Nevertheless, one could still argue that the differences that we found (during the regulation period relative to the pre-regulation period) between reappraisal, relative to distraction and control derive from Wilder's (1958) law of initial values. Applying this law for the present focus, it could be that because reappraisal showed the highest initial value, it had the highest tendency to drop during the regulation period. However, this conclusion is unwarranted, because according to the same law, the control group who showed the lowest initial value should have showed the smallest drop during the manipulation. In fact Control showed an increase that was of the same magnitude as that of Distraction which had higher initial values. Importantly, the FT results provide partial converging evidence to the SCL results. Specifically, they show that late reappraisal relative to late distraction and control results in a decrease in FT levels, which indicates an increased sympathetic activation.