واکنش پذیری عاطفی و مقررات شناختی در کودکان مضطرب

Abstract Recent models of anxiety disorders emphasize abnormalities in emotional reactivity and regulation. However, the empirical basis for this view is limited, particularly in children and adolescents. The present study examined whether anxious children suffer both negative emotional hyper-reactivity and deficits in cognitive emotion regulation. Participants were 49 children aged 10–17 with generalized anxiety disorder, social anxiety, or separation anxiety disorder as their primary diagnosis, as well as 42 age- and gender-matched non-anxious controls. After completing a diagnostic interview and self-report questionnaires, participants were presented with pictures of threatening scenes with the instructions either to simply view them or to use reappraisal, a cognitive emotion regulation strategy, to decrease their negative emotional response. Emotion ratings, content analysis of reappraisal responses, and reports of everyday use of reappraisal were used to assess negative emotional reactivity, reappraisal ability, efficacy and frequency. Relative to controls, children with anxiety disorders (1) experienced greater negative emotional responses to the images, (2) were less successful at applying reappraisals, but (3) showed intact ability to reduce their negative emotions following reappraisal. They also (4) reported less frequent use of reappraisal in everyday life. Implications for the assessment and treatment of childhood anxiety disorders are discussed.

Results Participant characteristics Participants in the AD and NAC did not differ significantly in gender, χ2(1, N = 91) = .09, p > .75, or in age, (M = 13.74, SD = 1.93, M = 13.42, SD = 2.38 in the NAC and AD accordingly), t(89) = .69, p > .49. All participants' scores on reading and vocabulary tests were within normal range. The three AD subgroups did not differ in clinical symptoms (i.e., severity of anxiety, severity of depression and overall severity of illness), however the SAD subgroup was younger than the other two subgroups, F(2,46) = 8.34, p < .01. Correlation between parent report on child's anxiety severity (SCARED-P) and the severity of anxiety according to child (SCARED-C) in our sample was high, r = .75, p < .001 for the whole sample. In the AD group, parent report on child's anxiety severity positively correlated not only with child's report on anxiety severity, r(46) = .40, p < .01 but also interviewer's rating of global illness severity, r(47) = .30, p < .05. In addition interviewer rating's of global illness was positively correlated with child's report of anxiety severity, r(48) = .69, p < .01. Table 1 shows the clinical characteristics of the two groups. Table 1. Clinical characteristics of participants. Variable Anxious group (n = 49) Non-anxious control group (n = 42) Difference M SD M SD df t 1. Severity of anxiety-child report 34.44 14.42 9.30 6.76 86 −10.43*** 2. Severity of anxiety-parent report 35.29 15.23 6.33 5.33 88 −11.69*** 3. Severity of depression 59.00 9.93 43.75 4.95 89 −9.02*** 4. Global illness severity 4.85 .95 1.19 .39 89 −23.15*** Note. Severity of anxiety-child and parent reports are the scores in the Screen for Anxiety and Related Emotional Disorders questionnaire (SCARED-C and SCARED-P, respectively). Severity of depression is the T-score calculated for the Children's Depression Inventory (CDI) and the Beck Depression Inventory (BDI-II). Global illness severity is the score in the Clinical Global Impression scale (CGI). Two AD participants' scores in the SCARED-C, were removed from analyses as they were more than 2.5 SD lower than parents' equivalent reports and were lower than would be expected based on the ADIS-C. * = p < .05, ** = p < .01, *** = p < .001. Table options Manipulation checks The images provoked mild-to-moderate negative emotional responses in our participants, as evidenced by a mean negative emotional rating for the whole sample of 3.5, SD = 1.7 for the images in the ‘view’ condition. When prompted to reappraise, participants generated reappraisals to 90% of the images on average, SD = 10.7%. As expected, the mean intensity of negative emotion in the ‘reappraise’ condition, following the instruction to reappraise, (M = 2.39, SD = 1.33) was lower than in the ‘view’ condition (mentioned above). This decrease in negative emotion was significant, t(88) = 8.74, p < .001, suggesting a reappraisal effect on participants' intensity of negative emotional response. One possible consequence of presenting ‘view’ and ‘reappraise’ conditions in a fixed order is that habituation might be responsible for the observed decrease in negative emotion ratings in the second (reappraise) block. To address this possibility, we performed several post hoc analyses examining the pattern of decreases in negative emotion ratings across the task. We began by splitting our 14-trial blocks for each condition into two blocks of seven, based on trial order. This yielded four blocks: view-early (the first seven trials), view-late (the second half of the view condition), reappraise-early (the first seven trials of the reappraise block), and reappraise-late (the second half of the reappraise block). We then calculated average negative emotion ratings for each of these four blocks. Negative emotion ratings decreased from view-early to view-late, t(73) = 2.8, p < .007, in the absence of any change in instruction, suggesting habituation. However, the decrease from view-late to reappraise-early is highly significant, t(73) = 5.8, p < .001 and is significantly greater in magnitude than the first decrease, t(73) = 2.1, p < .05. This result suggests that habituation alone cannot explain the reduction in negative emotion associated with the use of reappraisal in the second (reappraise) block. We used Pearson correlations to assess inter-correlations among emotional reactivity and cognitive regulation indices. Across all participants, reappraisal frequency was negatively correlated with emotional reactivity, r(89) = −.25, p < .05, meaning that individuals who reported greater everyday use of reappraisal showed lower levels of negative emotional reactivity in the task. Negative emotional reactivity was not correlated with the ability to generate reappraisals. Emotional reactivity was also not correlated with reappraisal efficacy; however, this lack of correlation is a necessary consequence of our strategy for operationalizing reappraisal frequency and should not be interpreted as a substantive finding. Reappraisal ability was correlated with reappraisal efficacy, r(86) = .21, p < .05, meaning that individuals who managed to reappraise more images showed greater decreases in negative emotion upon using reappraisal. Reappraisal ability in the task however, did not correlate significantly with reappraisal frequency in everyday life. Importantly, reappraisal efficacy was positively correlated with reappraisal frequency, r(89) = .27, p < .01, meaning that individuals who reported greater everyday use of reappraisal showed greater benefits from reappraisal in the laboratory. Baseline negative affect and negative emotional reactivity As expected, compared to the NAC group, AD participants reported higher levels of negative affect in the STAI-C before the REAR-I task started, (M = 24.95, SD = 3.95, M = 31.1, SD = 6.86), t(89) = 5.2, p < .001, d = 2.7. Also as hypothesized, the AD presented higher negative emotional response to the images in the ‘view’ condition, t(87) = 4.0, p < .001, d = .89, as presented in Fig. 2. The group difference in negative emotional response in ‘view’ remained significant after controlling for baseline negative affect, F(1,89) = 6.1, p < .05, ηp2 = .06, suggesting that the observed group difference is not explained by the AD's higher baseline negative affect. Negative emotional reactivity (A), reappraisal ability (B), reappraisal efficacy ... Fig. 2. Negative emotional reactivity (A), reappraisal ability (B), reappraisal efficacy (C), and reappraisal frequency (D) in anxious (AD) and non-anxious (NAC) participants. *p < .05. Error bars = standard errors of the means. Figure options Reappraisal ability, efficacy, and frequency As shown in Fig. 2, compared to the NAC group, AD participants were able to reappraise fewer of the images, t(85) = 3.0, p < .01, d = 0.64. 3 Even so, both groups were able to successfully generate reappraisals for the great majority of the pictures shown (87% for AD and 93% for NAC). With respect to reappraisal efficacy, there was no difference between the groups, t(87) = .4, p > .5. This result indicates that when reappraisal scores are adjusted for differences in emotional reactivity, the use of reappraisal was followed with similar decrease in negative emotion for the anxious and non-anxious children. 4 Finally, compared to the NAC group AD participants reported significantly less frequent use of reappraisal in everyday life according to the ERQ, t(88) = 3.6, p < .01, d = 77. Group differences in reappraisal ability and in reappraisal frequency persisted after controlling for baseline negative affect and emotional reactivity, (F(1,89) = 6.0, p < .05, ηp2 = .06 5 and F(1,86) = 5.2, p < .03, ηp2 = .05, respectively). Secondary analyses Age effects on emotional reactivity and reappraisal Although not the primary focus of the present investigation, one interesting question is whether there are age-related differences in emotional reactivity and cognitive regulation. Using simple regressions, age did not predict reappraisal ability, efficacy or frequency, but it did predict emotional reactivity. Specifically, age was negatively correlated with negative emotional reactivity, r(89) = −.36, p < .001, meaning that younger participants reacted with higher negative emotion to the threatening images. Importantly, when age was used as a covariate in group comparisons, AD and NAC still differed in negative emotional reactivity, reappraisal ability, and reappraisal frequency (p = .001, .004, .002 accordingly). Emotional reactivity, reappraisal, and symptom severity To examine whether symptom severity was related to emotional reactivity or to reappraisal indices, we correlated each of these measures with symptom severity in the AD group. Table 2 shows that negative emotional reactivity was positively correlated with of anxiety, depression and global illness. Reappraisal ability was not correlated with any of the clinical measures, reappraisal efficacy was not correlated with anxiety or depressive symptoms, but it was negatively correlated with global severity of illness. Everyday use of reappraisal was negatively correlated with severity of anxiety and depressive symptoms. Table 2. Correlation matrix of emotional reactivity, cognitive regulation indices and clinical symptoms within the anxious group. Subscale 1 2 3 4 5 6 7 1. Severity of anxiety–child report – 2. Severity of depression .59** – 3. Global illness severity .69** .33** – 4. Negative emotional reactivity .40** .37** .25* – 5. Reappraisal ability .10 .18 −.10 .02 – 6. Reappraisal efficacy −.11 −.00 −.35** −.03 .29* – 7. Reappraisal frequency −.28* −.38** −.20 −.14 .07 .28* – Note. N = 49. * = p < .05, ** = p < .01, *** = p < .001.