مفهوم سازی اجتناب تجربی از نشخوار فکری افسرده: سه آزمون از مدل

This study examined an experiential avoidance conceptualization of depressive rumination in 3 ways: 1) associations among questionnaire measures of rumination, experiential avoidance, and fear of emotions; 2) performance on a dichotic listening task that highlights preferences for non-depressive material; and 3) psychophysiological reactivity in an avoidance paradigm modeled after the one used by Borkovec, Lyonfields, Wiser, and Deihl (1993) in their examination of worry. One hundred and thirty-eight undergraduates completed questionnaire measures and participated in a clinical interview to diagnose current and past episodes of depression. Of those, 100 were randomly assigned to a rumination or relaxation induction condition and participated in a dichotic listening task, rumination/relaxation induction, and depression induction. Questionnaire measures confirmed a relationship between rumination status and avoidance; however, no significant effects were found in the dichotic listening task. Psychophysiological measures indicated no difference in physiological response to a depression induction among high ruminators (HR). However, low ruminators (LR) in the relaxation condition exhibited a larger IBI response than LR in the rumination condition. Overall, these results provide partial support for an avoidance conceptualization of depressive rumination. Implications of these findings are discussed.

Depressive rumination refers to the tendency of some individuals to recursively mull over the symptoms, causes and consequences of their negative affect (Nolen-Hoeksema, 1991). Rumination has been linked to longer and more severe depression, delayed recovery from depression, increases in suicidal ideation, and impairments in problem solving, motivation, and concentration (Eshun, 2000, Lyubomirsky et al., 2003, Lyubomirsky and Tkach, 2004, Siegle et al., 1999 and Smith et al., 2006). Rumination has also prospectively predicted major depression over a 2.5-year follow-up (Nolen-Hoeksema, 2000 and Spasojevic and Alloy, 2001). Given that depression affects 1 in every 5 adults (Angst, 1995), a greater understanding of the nature of rumination may aid in the design and implementation of intervention strategies that target ruminative response styles, and, thus, lessen and/or prevent depression. Several theories have been proposed to explain the negative effects of rumination. The most widely known is Nolen-Hoeksema’s (2004) Response Style Theory which suggests that rumination negatively impacts individuals in four crucial ways: 1) through the activation of negative thoughts and memories, rumination exacerbates the impact of depressed mood on thinking and increases the likelihood that individuals will make depressogenic inferences in regard to their current circumstances, 2) rumination interferes with problem solving and, thus, 3) impedes the implementation of successful strategies to overcome negative mood, and 4) rumination alienates potential social supports resulting in a loss of support. Other models have proposed that rumination is harmful because of its relationship to right hemisphere activation, its activation of negative beliefs regarding rumination and oneself, and/or its interference with positive restructuring of negative memories (Martin et al., 2004, Matthews and Wells, 2004 and Papageorgiou and Wells, 2004). Although support for some of these models has been obtained (e.g. rumination has been linked to impediments in problem solving and activation of negative self-relevant information), an alternative hypothesis is that rumination is a manifestation of experiential avoidance (Lyubomirsky and Nolen-Hoeksema, 1995, Lyubomirsky et al., 1999 and Mellings and Alden, 2000). Avoidance models have been applied to many types of psychopathology. For example, Hayes, Wilson, Gifford, Follette, and Strosahl (1996) proposed that experiential avoidance, a tendency to avoid contact with certain aspects of private experiences, is characteristic of many forms of psychopathology, including substance abuse, obsessive-compulsive disorder, panic disorder, and borderline personality disorder. Experiential avoidance is presumed to be harmful because it prevents individuals from effectively responding to emotional stimuli and often has the paradoxical effect of increasing avoided material (Hayes et al., 2004 and Wenzlaff and Wegner, 2000). Avoidance has also been described as a fundamental process in generalized anxiety disorder (GAD) in Borkovec’s (1994) model, which postulates that worry serves to distract individuals from deeper, more emotional, topics. This avoidance of emotionally arousing material through worry then provides a false, and reinforcing, sense of control over emotions, which ultimately prevents the individual from effectively processing their negative emotion. Given the high rate of comorbidity of GAD and depression (60%; Brown, Campbell, Lehman, Grisham, & Mancill, 2001), and the high correlations of measures of rumination and worry (r = .66; Beck & Perkins, 2001), it may be that rumination, like worry, serves an avoidance function. More specifically, it may be that ruminators avoid the experience of sadness through recursive cognition. Consistent with an avoidance conceptualization of rumination, studies have reported a direct link between self-reported depressive rumination and experiential avoidance (Cribb et al., 2006, Smith et al., 2005 and Smith et al., 2007). In addition, several studies have demonstrated a link between rumination and avoidant behaviors. For example, Lyubomirsky, Kasri, Chang, and Chung (2006) reported that, among women who had previously received a diagnosis of breast cancer, ruminating women waited 2 months longer than non-ruminators to report their initial symptoms to their health care professionals. The authors concluded that rumination contributed to a delayed response to symptoms, which supports the hypothesis that high ruminating individuals may avoid dealing with emotionally threatening material. Similarly, Smith et al. (2007) reported that high ruminating individuals reported significantly more cutting behavior than low ruminators, which has been construed as a way for individuals to get relief from aversive emotional experiences (Linehan, 1993). High ruminating individuals also reported using substance use as a coping strategy significantly more than low ruminators, which also suggests a preference for escape from negative emotions (Nolen-Hoeksema, 2007 and Smith et al., 2007). These studies all point to a link between depressive rumination and purposeful avoidance of negative emotions. The goal of the current study was to explore whether an experiential avoidance conceptualization explains the negative effects of depressive rumination. Other investigations of experiential avoidance have used a variety of methodologies to examine the avoidance construct. For example, a widely used self-report measure of experiential avoidance is the Acceptance and Action Questionnaire (AAQ; Hayes et al., 2004) and preliminary studies have found an association between self-reported rumination and high scores on the AAQ (Cribb et al., 2006 and Smith et al., 2005). A dichotic listening task has also been used as a behavioral measure of avoidance in worry (Laguna, Ham, Hope, & Bell, 2004). In this task, high worrying individuals demonstrated greater recall of non-threatening words, even when instructed to attend to threatening content presented simultaneously. This indicates a shift in attention away from anxiety provoking content, and thereby suggests that these individuals were demonstrating experiential avoidance. Similarly, it was expected that high ruminating individuals would also shift their attention away from depressive content in our modified dichotic listening task. Another indicator of avoidance lies in physiological response to aversive stimuli. Specifically, experiential avoidance has been linked to discordance between self-reported mood and arousal and physiological indices of arousal. For example, in a classic study of worry in GAD, Borkovec, Lyonfields, Wiser, and Deihl (1993) pre-selected individuals who were afraid of public speaking and had them worry or relax prior to imagining giving a public speech (an exposure to a feared stimulus). They found that individuals in the worry condition showed less heart rate response to this exposure, whereas individuals who relaxed showed greater heart rate response. These findings suggest that individuals who worried suppressed their heart rate response, and thereby avoided the physiological reaction to the feared stimulus. Similarly, other studies have also linked discrepancies between self-reported affective response and physiological indicators of response to experiential avoidance (Feldner et al., 2003 and Sloan, 2004). As a whole, these studies indicate that worry, avoidance of emotional state, and overall tendency to avoid experiences are related to a specific pattern of response to emotional stimuli: higher levels of self-reported negative mood and arousal and lower levels of physiological arousal. It was expected that rumination would also produce this pattern of high self-reported negative affect and attenuated physiological arousal during a sad mood induction. Given the variety of methods available to assess avoidance, and the preliminary support for a link between rumination and avoidance, this study aimed to build upon this work by investigating an avoidance conceptualization of rumination in 3 ways: via 1) self-report questionnaire measures of avoidance, 2) a behavioral index of avoidance (shift in attention away from depressive content during a modified dichotic listening task), and 3) a discrepancy between self-reported emotional response and physiological response to a sad mood induction. The third index of avoidance was explored by replicating the paradigm of Borkovec et al. (1993) in their examination of avoidance and worry. Given that worry and rumination frequently co-occur, and that worry and rumination are both repetitive cognitive processes, this seemed like an appropriate methodology to test avoidance in rumination. Therefore, their paradigm was modified in this study such that high ruminating (HR) and low ruminating (LR) participants were instructed to either ruminate or relax prior to completing a sad mood induction task while heart rate response was recorded. In addition to examining an avoidance model of rumination, this study also aimed to address some of the limitations of previous research in this area. For example, concerns have been raised regarding the face validity of the AAQ, the wide range of constructs measured by the AAQ (e.g. beliefs about emotions, avoidant behaviors, fears about the consequences of emotions, cognitive activities related to avoidance), and evidence that the AAQ actually may be measuring negative affectivity as opposed to avoidance. Therefore, the current study included an additional self-report measure of avoidance, the Affective Control Scale, which quantifies fear of emotional responses (ACS; Williams, Chambless, & Ahrens, 1997). It was hoped that the ACS would be a more pure index of avoidance given its focus primarily on metacognitive beliefs about emotion that drive avoidance, and its relationship to other avoidant processes such as worry (Roemer, Salters, Raffa, & Orsillo, 2005). However, because the AAQ is a widely used measure of experiential avoidance and has previously been used to relate to rumination, it was also included. In addition, this study aimed to address limitations of the original Borkovec et al. (1993) paradigm. One criticism of Borkovec et al.’s (1993) study is that it did not control for baseline levels of physiological activation in worriers; therefore, the blunted physiological response observed in worriers may have been due to increased physiological response to worry itself (Peasely-Miklus & Vrana, 2000). In support of this blunting hypothesis, Peasely-Miklus and Vrana (2000) reported that high worrying individuals showed a significant physiological response from baseline to a worry induction, and they did not exhibit an additional physiological response to the subsequent fear induction (Peasely-Miklus & Vrana, 2000). The authors concluded that although the heart rate suppression effect from Borkovec et al.’s (1993) study was replicated (i.e. there was no change in heart rate from the worry to fear inductions), it may be because participants were already activated from the worry induction. To control for this possibility, the present study obtained baseline measures of physiological response prior to the rumination induction and the blunting effect was specifically tested. In addition, this study included a measure of analytical thinking, which is hypothesized to be characteristic of rumination, to expressly test whether or not the rumination induction was effective in eliciting more ruminative thought, and, conversely, whether the relaxation induction was successful in circumventing rumination in HR individuals. In sum, this study examined an experiential avoidance conceptualization of depressive rumination using 3 methodologies. It was expected that 1) individuals who reported high levels of rumination would score higher on self-report measures of experiential avoidance than LR, 2) HR individuals would remember a greater number of neutral words (indicating a shift in attention away from depressive material) than LR during a dichotic listening task, and 3) HR individuals would report greater emotional response to a sad mood induction than LR, but demonstrate less of a physiological response to the sad mood induction. In addition, it was expected that the blunting hypothesis would not be supported by these data (i.e. heart rate would not increase in response to the rumination induction). This is the first study to directly test this conceptualization of rumination, and it adds to the current literature by examining multiple indices of avoidance (self-report, behavioral, physiological), and exploring an alternative hypothesis.