اثرات استراتژی تنظیم احساسات بر روی ولع مصرف سیگار، تعصب توجه و تداوم کار

Abstract The goal of this study was to investigate the effects of different strategies for regulating emotions associated with smoking on subjective, cognitive, and behavioral correlates of smoking. Emotion regulation was manipulated by instructing participants to reappraise (n = 32), accept (n = 31), or suppress (n = 31) their emotions associated with smoking. The dependent measures included subjective reports of craving, negative affect, and attentional biases, as measured by a modified dot-probe task, and persistence during a task to measure distress tolerance. Individuals who were encouraged to reappraise the consequences of smoking showed diminished craving, lower negative affect, had reduced attentional biases for smoking-related cues, and exhibited greater task persistence than those who were instructed to accept and suppress their urge to smoke. These findings suggest that reappraisal techniques are more effective than acceptance or suppression strategies for targeting smoking-related problems.

Results Manipulation checks The ASQ scores confirmed that the participants successfully employed the strategy that they were instructed to use (Table 1). More specifically, participants in the Reappraisal group were more likely to adjust their emotions; individuals in the Acceptance group were likely to tolerate their emotions; and people in the Suppression group were more likely to conceal their emotions after receiving the instructions. Table 1. Validation of the experimental manipulation based on the ASQ. ASQ subscales Reappraisal group Suppression group Acceptance group Mean SD Mean SD Mean SD Adjusting 30.87a 2.19 8.22b 1.99 9.61b 2.06 Tolerating 10.78b 2.67 11.44b 3.19 17.87a 2.40 Concealing 9.50b 2.42 28.51a 4.31 9.38b 2.43 Note: The Table shows means and standard deviations (SD) for the Affective Style Questionnaire (ASQ) subscales (Adjusting, Tolerating, and Concealing) of participants in the reappraisal, acceptance and suppression groups. Different subscripts indicate significant differences at p < .05 of the post-hoc ANOVA group comparisons. Table options Randomization The three experimental groups were comparable in their level of positive affect, F (2, 91) = 0.16, p > .98, partial η² = 0.04 and negative affect, F (2, 91) = 1.46, p > .23, partial η² = 0.03 on the PANAS for baseline measure ( Table 2). Furthermore, no differences were observed on craving on the QSU-Brief, F (2, 91) = 0.12, p > .98, partial η² = 0.17 and nicotine dependence on the FTND scale, F (2, 91) = 0.44, p > .64, partial η² = 0.09. Table 2. Positive and negative affect, and craving at baseline. PANAS baseline scores and FTND Reappraisal group Suppression group Acceptance group Mean SD Mean SD Mean SD Positive emotions 6.59 2.69 6.61 1.66 6.51 2.18 Negative emotions 3.84 2.52 5.77 4.87 4.67 3.15 FTND 3.38 2.12 2.87 2.01 3.16 2.17 Note: The Table shows means and standard deviations (SD) for the positive and negative affect on the PANAS and nicotine dependence on the FTND of participants in the reappraisal, suppression and acceptance groups. None of the post-hoc pairwise group comparison was statistically significant at p < .05. Table options Effects of regulation strategies on craving In order to examine changes in craving across time, we included individual raw scores for craving from the QSU-Brief and used these values as the dependent variable in all subsequent analyses. Scores for craving were measured at four time points as follows: at baseline, after the manipulation (following the instructions for the three emotion regulation strategies), after the craving induction, and after the visual probe and PASAT tasks. In order to examine the impact of the three emotion regulation strategies (Reappraisal, Acceptance, and Suppression) on subjective craving as measured with the QSU-Brief, we conducted a 4 (Time) × 3 (Strategies) repeated measures ANOVA with the craving scores as the dependent variable, Time (baseline, post-manipulation, post-induction, and post-PASAT task) as a within-subjects variable, and the three emotion regulation strategies as a between-subjects variable (Strategies: Reappraisal, Acceptance, and Suppression). Complete data were available for 32 participants in the Reappraisal Group, 31 participants in the Acceptance Group, and 31 participants in the Suppression Group. The ANOVA test revealed a significant Time effect for craving, F (3, 89) = 14.71 (Wilks' Lambda), p < .001, partial η2 = 0.33, a significant Strategies effect, F (2, 91) = 7.02, p < .001, partial η2 = 0.13, and a significant Strategies by Time interaction effect, F (6, 178) = 2.69, p < .02, partial η2 = 0.83. We also found a significant linear contrast for Time F (2, 91) = 7.22, p < .001, partial η2 = 0.14. We explored the interaction term by examining the interaction between polynomial contrasts for Time and Helmert contrast for Strategies, such that Reappraisal was compared against the composite of the Acceptance and Suppression strategies. The results showed a significant linear contrast in the Reappraisal, F (1, 30) = 12.66, p < .001, partial η2 = 0.29, but not in the Acceptance and Suppression conditions, F (1, 30) = 0.42, p > .8, partial η2 = 0.001. Post-hoc tests showed only a significant linear growth for the Suppression, F (1, 120) = 12.41, p < .001, and Acceptance groups, F (1, 120) = 13.61, p < .001 (all other ps > .29). Fig. 1 depicts the mean craving scores of the three experimental groups at baseline, post-manipulation, post-induction, and post-PASAT task. Absolute craving scores at baseline, post-manipulation, post-induction and ... Fig. 1. Absolute craving scores at baseline, post-manipulation, post-induction and post-PASAT task for participants who were instructed to reappraise, accept or suppress their craving. The graph depicts means of scores. Figure options Effects of regulation strategies on negative affect Before and after the final round of the PASAT, participants were asked to fill out a six-item mood scale measuring subjective distress. We found strong positive associations between the six-item mood scale from the PASAT and the PANAS with correlations ranging between r (31) = 0.88, p < .001 (for the Reappraisal group) and r (32) = 0.89, p < .001 (for the Acceptance group). Because the PASAT and PANAS were highly correlated, we only analyzed the PANAS, which is the more reliable and standard measure. To examine the impact of the three emotion regulation strategies (Reappraisal, Acceptance, and Suppression) on negative affect, we conducted a 4 (Time) × 3 (Strategies) repeated measures ANOVA with the negative affect subscale of the PANAS as the dependent variable. The results revealed a significant Time effect, F (3, 89) = 41.21 (Wilks' Lambda), p < .001, partial η2 = 0.58, a significant Strategies effect, F (2, 91) = 15.32, p < .001, partial η2 = 0.25, and a significant Time by Strategies interaction effect, F (6, 178) = 3.96, p < .001, partial η2 = 0.19. We also found a significant linear contrast for Time, F (2, 91) = 11.51, p < .001, partial η2 = 0.21. We explored the interaction term by examining the interaction between polynomial contrasts for Time and Helmert contrast for Strategies, such that the Reappraisal group was compared against the composite of the Acceptance and Suppression conditions. The results revealed a significant polynomial linear contrast in the Reappraisal condition, F (1, 30) = 23.97, p < .001, partial η2 = 0.44, but not in the Acceptance and Suppression conditions, F (1, 30) = 1.38, p > .2, partial η2 = 0.04. Post-hoc tests showed a significant linear growth for the Suppression group, F (1, 120) = 33.08, p < .001, the Acceptance group, F (1, 120) = 29.47, p < .001, and Reappraisal group, F (1, 124) = 12.84, p < .001 (all other ps > .5). Fig. 2 depicts the mean negative affect scores of the three experimental groups at baseline, post-manipulation, post-induction and post-PASAT task. Absolute negative affect scores at baseline, post-manipulation, post-induction ... Fig. 2. Absolute negative affect scores at baseline, post-manipulation, post-induction and post-PASAT task for participants who were instructed to reappraise, accept or suppress their craving. The graph depicts means of scores. Figure options Effects of regulation strategies on attentional bias Attentional bias scores were computed following earlier recommendations by subtracting mean reaction times (measured in milliseconds) to probes replacing smoking pictures from the mean reaction times to probes replacing control pictures (Field, Duka, Tyler, & Schoenmakers, 2009; Mogg & Bradley, 1999). A paired t-test showed that there was a significant difference in the participants' mean reaction times to probes replacing smoking pictures compared to mean reaction times to probes replacing control pictures for all conditions, t (93) = 13.40, p < .001. In order to examine the impact of the regulation strategies on the attentional bias related to smoking cues, we conducted an ANOVA with participants' difference scores on the visual probe task as a dependent variable. Results showed a significant effect of the regulation strategies on attentional bias, F (2, 91) = 5.99, p < .004, partial η2 = 0.11. Post-hoc tests (Scheffe) indicated that participants in the Reappraisal group showed significantly lower levels of interference on the dot-probe task than those in the Suppression group (mean difference: −13.89, SE = 4.27, p < .007) and the Acceptance group (mean difference: −11.27, SE = 4.27, p < .035). We found no difference in task interference for the participants in the Acceptance and Suppression groups (mean difference: −2.62, SE = 4.30, p = .831). Fig. 3 depicts the differences in task interference between the three experimental groups. Means and standard errors of bias scores for the three experimental groups. Fig. 3. Means and standard errors of bias scores for the three experimental groups. Figure options Association between craving and attentional bias We further examined the association between subjective craving and attentional bias. In order to account for the non-normal distribution of the attentional bias index, we re-expressed the data using a logarithmic transformation. We observed a weak positive association between measures of craving and attentional bias index, r (94) = 0.41, p < .05, in the combined group. In the Reappraisal group, this correlation was r (32) = −0.37, p < .036. For the Acceptance group, the correlation was, r (31) = 0.39, p < .028 and for the Suppression group the correlation was r (31) = 0.35, p < .05. The difference in the correlation coefficients between the Reappraisal group and the Suppression group was statistically significant, z = −2.85, p < .05. We further found significant differences between the correlation coefficients of the Reappraisal and Acceptance group z = −3.02, p < .05, but no difference between the Acceptance and Suppression group z = 0.17, p > .8. Effects of regulation strategies on distress tolerance In order to examine the impact of the regulation strategies on task persistence, we conducted an ANOVA with participants' distress scores (measured as task persistence in seconds) in response to the PASAT task as the dependent variable. Results showed a significant effect of the three emotion regulation strategies on distress tolerance, F (2, 91) = 17.65, p < .001, partial η2 = 0.34. Post-hoc tests (Scheffe) showed that participants in the Reappraisal group persisted significantly longer with the distressing task than those in the Suppression group (mean difference: 120.82, SE = 21.93, p < .001) and the Acceptance group (mean difference: 102.40, SE = 22.11, p < .001). We found no difference in task persistence for the participants in the Acceptance and Suppression groups (mean difference: −18.41, SE = 22.28, p = .71). Fig. 4 shows the difference in task persistence between the three experimental groups. Means and standard errors of task persistence for the three experimental groups. Fig. 4. Means and standard errors of task persistence for the three experimental groups. Figure options Association between craving and persistence In order to examine the association between craving and task persistence, we computed correlations between cravings reported after finishing the PASAT task (post-PASAT) and the persistence on the PASAT. As predicted, we observed a significant negative association between level of craving and task duration, r (94) = −0.56, p < .05, in the combined group. In the Reappraisal group, the correlation between craving and persistence was r (32) = −0.44, p < .05. In contrast, in the Suppression and Acceptance groups, these correlations were r (31) = 0.35, p < .05 and r (31) = 0.31, p < .05, respectively. The differences in the correlation coefficients between the Reappraisal and the Suppression group were statistically significant, z = −3.16, p < .002. These differences were also statistically significant between the Reappraisal and the Acceptance group, z = −2.99, p < .003. The differences in the correlations coefficients were not statistically significant between the Acceptance and the Suppression group, z = 0.17, p > 0.8.