زمان بندی اثرات سابق- و واکنش- راهبردهای تنظیم احساسات

Abstract Distraction and cognitive reappraisal influence the emotion-generative process at early stages and have been shown to effectively attenuate emotional responding. Inhibiting emotion-expressive behavior is thought to be less beneficial due to later implementation, but empirical results are mixed. Thus, the current study examined the temporal dynamics of these emotion regulation strategies at attenuating the late positive potential (LPP) while participants were shown unpleasant pictures. Results revealed that all strategies successfully reduced the LPP and self-reported negative affect. We confirmed that distraction attenuated the LPP earlier than cognitive reappraisal. Surprisingly, expressive suppression affected emotional responding as early as distraction. This suggests that suppression was used preventively and disrupted the emotion-generative process from the very beginning instead of targeting the emotional response itself. Thus, the obtained results point to the importance of considering the point in time when response-focused emotion regulation strategies are being implemented.

Introduction Emotions are useful in terms of signaling potentially significant events in the environment and mobilizing the organism to take appropriate action. However, sometimes emotions can also turn maladaptive (e.g., when being too intense or contextually inappropriate), and the ability to regulate emotional responses is crucial for goal-directed and adaptive behavior. Thus, emotion regulation involves the ability “to influence which emotions we have, when we have them, and how these emotions are experienced and expressed” (Gross, 1998a, p. 224). Moreover, emotion regulation capacities appear to be essential for mental health (Gross & Muñoz, 1995). According to the process model of emotion regulation, emotion generation is a dynamic process that is based upon specific cognitive processes and unfolds over time (Gross, 1998b). After encountering an emotive situation, attentional deployment towards the emotional stimulus is followed by the identification of the situation's meaning. These cognitive processes give rise to the emotional response that involves changes in physiological, cognitive-emotional, and behavioral response systems. According to this model, emotion regulation strategies can be divided into antecedent-focused and response-focused regulation (Gross, 1998b). While the first influences the emotion-generative process prior to initiation of the emotional response, the latter modulates the emotional response itself. Distraction and cognitive reappraisal are two well-studied forms of antecedent-focused regulation strategies. Distraction alters emotional responses by directing attention away from the emotional situation very early on before elaborative processing has taken place. Cognitive reappraisal requires individuals to reformulate the meaning of a situation in less emotional terms and can be implemented only after an emotional situation has been attended to and appraised. Thus, reappraisal is thought to affect the emotion-generative process later than distraction. It was shown that both distraction and cognitive reappraisal are effective in reducing negative emotion processing as evidenced by self-reported affect (e.g., Lieberman, Inagaki, Tabibnia, & Crockett, 2011), neural indices of negative emotion processing like amygdala activation (e.g., Kanske, Heissler, Schönfelder, Bongers, & Wessa, 2011), and peripheral physiology (e.g., Neumann et al., 2004 and Ray et al., 2010; but see Urry, 2009). A prominent example of response modulation involves the suppression of emotion-expressive behavior. Regarding the consequences of expressive suppression, two contradicting hypotheses have been suggested. According to a catharsis model, emotions are thought to “pile up” if not expressed (Freud & Breuer, 1960). Consequently, expressive suppression should enhance emotion experience. On the other hand, facial feedback theory states that inhibiting emotional facial expressions dampens emotion experience (Adelmann & Zajonc, 1989). Prior studies investigating the effectiveness of expressive suppression have yielded inconsistent results reporting both a decrease (Dunn et al., 2009, Gross and Levenson, 1997, Hayes et al., 2010 and Vrtička et al., 2011) and an increase (Dan-Glauser and Gross, 2011, Goldin et al., 2008, Gross, 1998a, Kunzmann et al., 2005 and Roberts et al., 2008) of emotional responding. Thus, the impact of expressive suppression on emotional processing remains to be further elucidated. Appraisal theories of emotion (Arnold, 1960 and Lazarus, 1991) supported by contemporary research (Phillips, Drevets, Rauch, & Lane, 2003) emphasize that identification of the emotional significance of a situation is a prerequisite for emotion generation. Hence, investigating how the suppression of emotion-expressive behavior affects emotional guidance of attention would complement existing research. The facilitated processing of emotional stimuli is reflected in a central-parietal slow positive deflection in the event-related potential (late positive potential, LPP). The magnitude of the LPP has been shown to be enhanced for emotionally arousing compared to neutral stimuli, beginning approximately 300 ms following stimulus onset and being sustained up to several seconds (for a review see Hajcak, MacNamara, & Olvet, 2010). The LPP is related to activity in occipital, inferotemporal, and parietal visual areas (Keil et al., 2002 and Sabatinelli et al., 2007) and reflects increased attention and perceptual sensitivity to motivationally relevant stimuli (Briggs and Martin, 2009, Hajcak et al., 2010, Lang et al., 1997, Schupp et al., 2000, Schupp et al., 2004 and Weinberg and Hajcak, 2010). The magnitude of the LPP is sensitive to voluntary emotion regulation. Thus, a reduction of LPP amplitudes was shown when participants were instructed to distract attention from unpleasant pictures either by generating unrelated neutral thoughts (Thiruchselvam, Blechert, Sheppes, Rydstrom, & Gross, 2011) or by focusing on non-arousing picture aspects (Dunning and Hajcak, 2009, Hajcak et al., 2009 and Hajcak et al., 2006). Likewise, changing the meaning of an emotional scene in less emotional terms has proven a powerful technique to attenuate the LPP to unpleasant pictures either via self-generated (Hajcak and Nieuwenhuis, 2006, Moser et al., 2009 and Thiruchselvam et al., 2011) or directed reappraisal (Foti and Hajcak, 2008, MacNamara et al., 2011 and Mocaiber et al., 2010). Further, LPP reductions have also been observed for positive emotion regulation using strategies targeting attentional or appraisal processes (Hajcak et al., 2006 and Krompinger et al., 2008). Suppressing emotional facial expressions has been shown to attenuate LPP amplitudes during positive emotion (Korb, Grandjean, Samson, Delplanque, & Scherer, 2012), but to our knowledge, has not yet been investigated during negative emotion. To date, only two studies addressed the question whether emotion regulation strategies differentially affect the temporal dynamics of emotional responding. It was shown that distraction reduced emotional responding prior to cognitive reappraisal (Thiruchselvam et al., 2011), and reappraisal in turn exerted its influence earlier than expressive suppression (Goldin et al., 2008), confirming predictions derived from the process model of emotion regulation (Gross, 1998b). However, Thiruchselvam et al. (2011) investigated emotion-related ERPs, whereas Goldin et al. (2008) focused on neural responses associated with the cognitive control of emotion, making comparisons difficult. Therefore, examining all three emotion regulation strategies within one experimental design using one index of emotion regulation would complement these findings. In this regard, the LPP constitutes a potentially powerful electrophysiological index to examine the differential temporal dynamics of the regulation strategies due to the high temporal resolution of the electroencephalogramm (EEG). This study aims at investigating effectiveness and timing of the response-focused emotion regulation strategy expressive suppression in relation to the two antecedent-focused strategies distraction and cognitive reappraisal. ERPs during processing of neutral and unpleasant pictures were recorded along with valence and arousal ratings while participants either maintained or reduced their emotion via cognitive reappraisal, distraction, or expressive suppression. We expected to replicate the previous finding that distraction modulates LPP amplitudes prior to cognitive reappraisal ( Thiruchselvam et al., 2011). Further, we hypothesized that expressive suppression should affect the time course of the LPP later than both distraction and cognitive reappraisal ( Goldin et al., 2008 and Gross, 1998b). Additionally, we examined whether emotion manipulation and regulation instructions modulated the LPP and self-reported affect in the expected direction in order to validate the experimental design. Specifically, we expected an emotion effect as reflected by increased unpleasantness and arousal ratings and enhanced LPP amplitudes for unpleasant compared to neutral pictures. Regarding emotion regulation, we predicted that distraction and cognitive reappraisal successfully decrease subjective emotion experience and LPP amplitudes. Because few studies have examined the effects of expressive suppression on emotion-related ERPs, we will explore how the LPP is modulated by this response-focused emotion regulation strategy.

3. Results 3.1. Validation of the experimental design To test whether the emotion manipulation and regulation instructions modulated the LPP and self-reported affect in the expected direction, mean LPP amplitudes (300–5000 ms) and mean valence and arousal ratings were submitted to repeated-measures analyses of variance (ANOVA) with the within-subjects factor condition (five levels: neutral-maintain, unpleasant-maintain, unpleasant-distract, unpleasant-reappraise, unpleasant-suppress). For significant main effects and interactions, post-hoc tests were performed using Bonferroni correction. The Greenhouse–Geisser correction was used to correct for violations of sphericity. 3.1.1. Behavioral results Mean valence and arousal ratings are displayed in Table 1. Analyses revealed a main effect of condition for both mean arousal and valence ratings, F(4, 76) = 119.29, p < .001, η2 = .86, ε = .56 and F(4, 76) = 81.51, p < .001, η2 = .81, ε = .54, respectively. Pairwise comparisons showed that unpleasant pictures paired with the instruction to maintain emotions were rated as more arousing and unpleasant than unpleasant pictures in all three regulation conditions (all ps ≤ .001). All regulated unpleasant pictures were rated higher in arousal as compared to neutral pictures (all ps < .001). Table 1. Valence and arousal ratings across experimental conditions (means and standard deviations). Condition Valence Arousal M SD M SD Neutral-maintain 3.79 0.99 3.57 1.05 Unpleasant-maintain 6.99 0.57 6.99 0.74 Unpleasant-distract 5.89 0.76 5.94 0.82 Unpleasant-reappraise 5.71 0.82 5.72 0.77 Unpleasant-suppress 6.44 0.63 6.37 0.82 Note. Ratings were made on a 9-point rating scale (Valence: 1 = pleasant, 9 = unpleasant; Arousal: 1 = low arousal, 9 = high arousal). Table options 3.1.2. ERP results Fig. 1 depicts grand average ERP waveforms for each condition at centro-parietal recording sites and the scalp distribution for unpleasant minus neutral pictures in the maintain condition. The emotion effect was clearly evident as a pronounced positivity for unpleasant compared to neutral trials. The analyses identified a significant main effect of condition, F(4, 76) = 9.93, p < .001, η2 = .34, ε = .65. Pairwise comparisons confirmed that the LPP was enhanced for unpleasant-maintain (M = 4.52, SD = 4.46) compared to neutral-maintain trials (M = 0.82, SD = 3.23), p = .003. Regarding emotion regulation effects, it was shown that relative to maintain trials LPP amplitudes for unpleasant pictures were significantly reduced in all three regulation conditions (distraction: M = 1.93, SD = 2.92, p = .002; reappraisal: M = 2.29, SD = 3.62, p = .004; suppression: M = 1.84, SD = 3.44, p < .001) and did not differ significantly from amplitudes for neutral pictures (all ps > .1). 3.1.3. Relationship between self-reported arousal and the LPP To examine how LPP responses elicited by unpleasant pictures relate to subjective arousal ratings, Pearson correlations between mean LPP amplitudes and arousal ratings were calculated for each condition. Significant correlations were obtained in the maintain (r = .45), distract (r = .57), and suppress (r = .48) condition, all ps ≤ .05. No significant effect was found for the reappraise condition, r = .25, p = .308. 3.2. Timing analysis of early LPP amplitudes To investigate the temporal dynamics of the LPP as a function of different emotion regulation strategies, the early time window of the LPP (300–1700 ms) was divided into seven time epochs of 200 ms. These specific timing characteristics were chosen to be in line with previous studies investigating temporal characteristics of the LPP (Hajcak and Olvet, 2008 and Thiruchselvam et al., 2011). While distraction was shown to reduce the LPP as early as 300 ms (Hajcak et al., 2009), reappraisal seems to attenuate LPP amplitudes from 1500 ms (Thiruchselvam et al., 2011). Early LPP amplitudes are shown in Fig. 2. Average amplitudes of the early LPP were subjected to a 5 (condition: neutral-maintain, unpleasant-maintain, unpleasant-distract, unpleasant-reappraise, unpleasant-suppress) × 7 (time epoch: 300–500 ms, 500–700 ms, 700–900 ms, 900–1100 ms, 1100–1300 ms, 1300–1500 ms, 1500–1700 ms) repeated-measures ANOVA. The Greenhouse–Geisser correction was used to correct for violations of sphericity. Early LPP amplitudes (300–1700ms) at centro-parietal recording sites. (For ... Fig. 2. Early LPP amplitudes (300–1700 ms) at centro-parietal recording sites. (For interpretation of the references to color in the artwork, the reader is referred to the web version of the article.) Figure options Differential timing effects were reflected in a significant Condition × Time Epoch interaction, F(24, 456) = 6.17, p < .001, η2 = .25, ε = .27. For post hoc comparisons, amplitudes for unpleasant-maintain trials were compared to each regulation strategy (unpleasant-distract, unpleasant-reappraise, unpleasant-suppress) for each of the seven time epochs. Thus, seven paired t tests were conducted for each of the three comparisons. To control the family-wise error rate, p values were adjusted with the Bonferroni–Holm correction. This method is less conservative and provides more statistical power than the simple Bonferroni method. 4 The results of the planned comparisons are displayed in Table 2. Table 2. Mean differences (MD) and corresponding p values of paired t tests between unpleasant-maintain trials and each regulation condition for every time epoch of early LPP amplitudes (300–1700 ms). Time epoch (ms) Maintain-distract Maintain-reappraise Maintain-suppress MD p MD p MD p 300–500 1.35 .020a 0.87 .056 1.89 .001a 500–700 1.45 .012a 0.51 .032 1.53 .001a 700–900 1.87 .002a 1.11 .001a 2.21 .001a 900–1100 2.46 .003a 2.56 .001a 2.84 .001a 1100–1300 2.30 .006a 1.93 .002a 2.39 .001a 1300–1500 2.39 .001a 2.33 .001a 2.88 .001a 1500–1700 2.33 .001a 2.46 .001a 2.77 .001a a p < 0.05 after Bonferroni–Holm correction. Table options Compared to maintain trials, LPP amplitudes for unpleasant pictures were attenuated for both distraction and suppression trials starting early from 300 ms following picture onset, t(19) = 2.53, p = .020 and t(19) = 4.33, p < .001, respectively. Reappraisal on the other hand was accompanied by an amplitude reduction not until 700 ms, t(19) = 4.17, p = .001. For all three regulation strategies, the LPP remained significantly attenuated for the following time epochs. 5