Culture and affect: the factor structure of the affective style questionnaire and its relation with depression and anxiety among Japanese

Recent affective neuroscience, psychopathology, and psychotherapy research has focused on the dysregulation of emotion as an underlying transdiagnostic process of depression and anxiety [1–7]. Emotion regulation refers to “the processes by which individuals influence which emotions they have, when they have them, and how they experience and express these emotions” [8]. While cognitive behavioral therapies (CBT) were originally developed from learning theory and cognitive theory [9], the application of modern CBT incorporates the findings of recent advances in emotion regulation research to improve the therapeutic strategy [5]. Research on the topic of emotion regulation could enhance the translation and utilization of the neuroscience findings to improve the intervention strategy of CBT.

Hofmann et al. [5] reviewed the existing findings of emotion regulation research and proposed the emotion dysregulation model of depressive and anxiety disorders. This theory assumes that the person’s affective style, which is influenced by the interaction between external or internal triggers and the individual’s diathesis, determines the experience of positive and negative affect; if certain affective styles contribute to the maladaptive regulation to maintain negative affect, emotional disorders, such as depression and anxiety, can occur. According to this theory, the most effective treatment strategies for mood and anxiety disorders are by (1) targeting emotion dysregulation by promoting adaptive emotion regulation strategies, (2) decreasing negative affect and increasing positive effect, and (3) promoting adaptive affective styles [5]. This theory assumes that a person’s affective style is of central importance in mood and anxiety disorders. Affective style refers to the inter-individual differences in emotional sensitivity and regulation [10]. Thus, it is a trait-like individual difference variable that determines a person’s habitual use of emotion regulation strategies to specific emotional experiences.

Based on the existing literature on the topic of emotion regulation and affective style, Hofmann and Kashdan [11] developed the Affective Style Questionnaire (ASQ) to measure this individual difference variable. Factor analyses, using two separate United States (US) samples of university students, provided a three-factor structure: concealing, adjusting, and tolerating. Concealing refers to the tendency to suppress or avoid intrapersonal and interpersonal emotions after the generation of the emotion. Adjusting refers to the ability to be aware of and utilize the emotion as information, and modulate the emotional experience and expression in response to the demands of a given context. Finally, tolerating reflects the tendency to respond to an emotion in a nondefensive manner, even if the negative (or positive) emotion leads to some degree of distress.

These affective styles are one of the processes targeted by modern CBT in the treatment for depression and anxiety. Because both concealing and emotional avoidance involve attempts to not experience an emotion, these two affective styles share the conceptual similarity. The goal of the exposure strategies in CBT is to prevent the manifestation of the avoidant pattern and to allow the patient to experience the entire range of emotional experiences [12–14]. One of the core aspects of adjusting is the utilization of emotion as important information in a given situation. Recently, emotion-focused transdiagnostic CBT [13] specifically educates patients about the functional importance of emotions and to increase their emotional awareness of their emotional experiences. The goal of Acceptance-and-Commitment Therapy, Dialectical Behavior Therapy, and various types of Exposure Therapies are to develop the clients’ ability to tolerate distressing emotions [12, 15, 16].

It is assumed that these therapeutic strategies reduce depression and anxiety by preventing maladaptive concealing styles and by fostering more adaptive adjusting and tolerating styles. Hence, these affective styles have been hypothesized to have a significant effect on depression and anxiety. In fact, suppression, which is a form of concealment, has positive effect on anxiety and depression in Western samples [17]. Although the findings for adjusting and tolerating are limited, both could be conceptualized as adaptive forms of emotion regulation [5]. Hence, we hypothesized that the subscales of ASQ would be related to the anxiety and depression symptoms.

However, the clinical importance of affective styles and findings regarding ASQ may be culturally-specific. Although separate factor analyses from two samples of US university students revealed the cross-validity of the three factor structure, these results may be limited to a US population and may not be apply to other cultures. To the best of the author’s knowledge, there is no study that examined this factor structure in the non-western population. Before examining the relationships of affective styles with depression and anxiety, it is needed to confirm the factor structure, reliability and convergent and discriminant validity in the non-western culture.

Some findings suggest that culture has a profound impact on the construal of the world view and the self, which could result in differences in the processing of emotion and brain function within and between individuals [18, 19]. For example, the degree of reappraisal and suppression and the relationship between these two emotion regulation strategies were different among samples obtained from 23 different countries [20]. In this study, participants from cultures that emphasized the maintenance of social order, such as Japan, tended to have higher scores on suppression than did those from cultures that emphasized individual autonomy. Moreover, reappraisal and suppression tended to be positively correlated in the former culture, and tended to negatively correlate in the latter culture [20]. Culture may posit the display rules of emotion [21], and this could influence the degree or function of emotion regulation. For example, it has been shown that the social consequences of suppression are dependent on the person’s cultural values [22].

Given these considerations, we conducted this study to examine the three purposes. First purpose is to examine the cross-validation of factor structure with a sample comprised of Japanese university students. Second purpose is to examine the convergent and discriminant validity of the ASQ with related and unrelated variables. Third, we examine the influence of affective styles on depression and anxiety.

For the convergent and discriminant validity, we hypothesized that the correlation pattern of ASQ subscales and other emotion regulation variables observed in the original validation study [11] would be replicated in the Japanese sample. Specifically, we expect that adjusting would positively correlate to reappraisal [23] and psychological flexibility [24], and negatively correlated to difficulty in identifying and describing feelings [25]. As mentioned, one aspect of adjusting is express emotion in accordance with the circumstances. This ability is considered to be required to reappraisal and flexibility. Moreover, the ability to use emotion as information, which is other part of adjusting, requires the identifying and describing emotion without engaging in rumination. For concealing, we expected that this variable would be positively correlated to suppression [26] because of the conceptual commonality [11]. Tolerating would be positively correlated to psychological flexibility because it is suggested that one aspect of psychological flexibility is accepting emotion. In addition to these hypotheses, we also examined the relationship with the Self-Construal Scale (SCS; [27] to exploratory examine the cultural aspects of the ASQ. We did not set the hypotheses a priori for the relation between ASQ and SCS.

The purposes of this study were to (1) conduct a cross-validation of factor structure of the ASQ in a Japanese university sample, (b) to examine the reliability, convergent and discriminant validity of the ASQ subscales, and (c) to examine the influence of affective styles on depression and anxiety in a Japanese sample. Some of our findings describing the factor structure differed from those of the original study, which used a sample of American participants. Consistent with the factors obtained in the US study, we also identified the concealing, adjusting, and tolerating factors. However, the tolerating factor had poor reliability. Moreover, a new factor, labeled holding, emerged. Correlation analyses indicated that these subscales had acceptable convergent and discriminant validity. Regression analyses showed that, in addition to the conventional measure of emotion regulation (reappraisal and suppression), the factor adjusting was negatively associated with depression and anxiety, and the factor holding was negatively associated with depression. These results underscore the importance of focusing on other types of affective styles that have not been systematically examined by previous studies and that may be culture specific.

Our factor analyses, using the same method and rotation as that described in the original paper, produced a four-factor structure when applied to a sample of Japanese participants. The concealing, adjusting, and tolerating factors consisted of the items that loaded onto the same factor as described in the original paper. In addition to these replicated factors, a new factor, labeled holding, was obtained. This factor consisted of three items that originally loaded onto the factors concealing, adjusting, and tolerating. This factor appears to reflect the ability to control emotions through self-restrain. It further appears that the holding strategy is different from simply tolerating and emotion. Although the test-retest correlation coefficient of tolerating was acceptable, the internal consistency was poor. Because this subscale consisted of only three items, one way of improving its reliability might be developing and adding other items reflecting the meaning of tolerating. In terms of test-retest reliability, none of the four factors of the subscale could not be concluded as stable (r = .60 − .76). These results mean only 36–58% of variance could be explained by the score at the time of one month ago. This finding suggests that the affective style could vary over time and, hence, be targeted for enhancement or prevention by CBT.

The affective styles (concealing, adjusting, tolerating, and holding) were weakly or moderately correlated with each other. Thus, all three styles might reflect some degree of regulating emotion. Interestingly, adjusting was moderately correlated (r = ~.35) with the other three styles. Thus, adjusting could be an important regulation style that is involved in many emotion regulation types. For example, it might be needed to adjust emotion in order to tolerate, suppress, or hold.

Convergent and discriminant validity is generally supported for ASQ subscales. The correlational data for the ASQ subscales of concealing and adjusting were consistent with the previous findings of studies conducted in the US. Adjusting was correlated with the AAQ-II (r = .51), reappraisal (r = .39), difficulty identifying feelings (r = -.33), and difficulty describing feelings (r = -.25). These results support the conceptualization of adjusting as the tendency for emotional awareness and to utilize emotion as information important for guiding behavior. The finding of a moderate correlation between concealing and suppression (r = .54) was also consistent with the results of those that used a sample of American participants. Whereas concealing was found to be positively correlated with difficulty identifying feelings (r = .18) and with difficulty describing feeling (r = .38) in the earlier US study, the correlations were not significant in the present study, which used a sample of Japanese participants. Thus, concealing might not be as maladaptive in an Asian culture as it is in a Western culture [22]. Holding, a new subscale identified in the present study, seems to have a similar meaning as concealing. In fact, the moderate relationship found between these affective styles (r = .49, p < .01) was similar to the pattern of the relationship with other instruments. However, only holding was negatively correlated with venting coping and positively correlated with positive reframing coping. These results suggest that holding might be a more adaptive style of regulating emotions in the Japanese culture.

Concealing and adjusting were positively correlated with independent self-construal, and holding was positively correlated with interdependent self-construal. Along with the results of the factor analyses, these results imply that holding is one of the unique styles of emotion regulation used in an Asian culture.

Our regression analyses showed the importance of focusing on affective style in considering its adaptive or maladaptive function in emotion regulation. Specifically, reappraisal, holding, and adjusting were negatively associated with depression, whereas suppression was positively associated with depression. These reappraisal and suppression findings were consistent with the previous findings [17]. Moreover, holding and adjusting were negatively associated with depression. In the correlational analyses, both affective styles were negatively correlated with rumination. Hence, these affective styles might have negative association with depression through the mediating role of rumination. Holding is also similar to distress tolerance. As distress tolerance is associated with low depression [42], this factor may contribute to the inhibitory effect of holding on depression. Adjusting and reappraisal were associated with low anxiety. As with the depression findings, the reappraisal finding is in line with the existing literature [17]. Interestingly, adjusting had a stronger negative association with anxiety. These results support the view that emotional awareness and emotion utilization play important roles in the treatment of anxiety disorders [13].

These results, however, must be evaluated while considering the limitations of this study. First and foremost, the participants were university students living in an urban area of Japan. Consequently, we do not know if the results of this study generalize to clinical populations, such as those suffering from depression or an anxiety disorder. Moreover, because affective styles seem to play a key role in the maintenance of such disorders [5], it is necessary to replicate the present findings by using clinical samples. Second, though we employed the prospective design to examine the influence of ASQ on depression and anxiety, the time interval was relatively short (i.e., one month). Third, we did not have a priori hypothesis for the cultural difference of factor structure or other findings regarding ASQ. Holding, the new factor obtained in this study, might be culturally specific affective style for Japanese, or Asian culture. Further study should be conducted to examine this factor conceptually and psychometrically.

Despite of these limitations, this study provides the first findings on factor structure of the ASQ and its relation with depression and anxiety among Asian population. This study adds important information to the expanding field of emotion regulation research by highlight the importance of considering cultural factors, which may lead to novel affective styles (i.e., holding) associated with depression and anxiety.