Print Email Download Reference This Send to Kindle Reddit This
submit to reddit

Attributional style and depression

Attributional Style And Depression: Does Experiential Avoidance Function As A Mediator?

        Depression is a complex illness, with many factors thought to be involved in the development and maintenance of the disorder. One possible risk factor for depression is the attributional style which an individual has (Abramson, Seligman, & Teasdale, 1978); or in other words, the way a person explains why the events in his or her life occur. Although numerous studies have found that negative attributional style is linked to depression, other studies have shown there to be a weaker relationship. Therefore, there may be additional factors involved in mediating the relationship between attributional style and depression. This study hypothesized that experiential avoidance would function as one of these mediators.

Attributional Style and Depression

        The concept of attributional style was first developed by Abramson et al.ญญ (1978) in their reformulation of the Learned Helplessness theory of depression1 (Seligman & Maier, 1967). Attributional style refers to the habitual way that individuals explain the events which occur in their lives, and their expectancies for future events. Abramson et al. stated three dimensions which can be used to indicate a person's attributional style. The first is a locus of control measure. This is where attributions may be made to internal causes, with the individual believing the cause of the event is within oneself; or to external causes, where the cause is outside of oneself. The second dimension is about how stable or unstable the individual believes the cause of the event is. Finally the cause may be described as global, where it also affects other areas of life; or specific, where the cause only affects the given situation. These three dimensions can be measured on scales and be combined to yield a perception of an individual's attributional style. Those who explain negative events in terms of internal, stable and global causes are said to have a pessimistic or negative attributional style, rendering them 'helpless', whilst those who explain negative events to external, unstable and specific causes have an optimistic attributional style (Seligman, 1990). Attribution theory postulates that those individuals who have negative attributional styles are more prone to depression.

This model has received a lot of support over the years. Sweeney, Anderson and Bailey (1986) concluded in a meta-analysis of over 100 studies that attributional style is highly predictive of depression, therefore providing support for the theory. However, a study by Munton (1985) found that the link between attributional style and depression was fairly weak; he claimed that the association between the two factors is overstated, and that there are probably other explanations for the existence of this relationship. Therefore, this link is likely to be affected by other factors. Past research has highlighted the role of experiential avoidance in depression (Tull, Gratz, Salters, & Roemer, 2004), and so it is possible that this cognitive process mediates the relationship between attributional style and depression.

Experiential Avoidance

Experiential avoidance can be defined as an unwillingness to experience negatively evaluated private thoughts, sensations, and emotions. The individual makes deliberate attempts to escape from or to consciously suppress these internal events (Hayes, 1994). Long term suppression of these private events, whilst serving a self protective function in the short term, may cause problems over an extended period of time. Experiential avoidance becomes an issue when avoidance behaviours are applied strictly and inflexibly, to the extent that great effort goes into suppressing these unwanted thoughts (Kashdan, Barrios, Forsyth, & Steger, 2006). Hayes, Wilson, Gifford, Follette and Strosahl (1996) state that a process known as cognitive fusion is at the root of experiential avoidance. This fusion causes negatively evaluated thoughts to be defined as unbearable, and this leads to the splitting of the self away from the internal thoughts as a coping mechanism against negative feelings.

Studies have shown that experiential avoidance is involved in the development and maintenance of many psychological problems, including depression (Tull et al., 2004). Gross and John (2003) argued that this occurs because individuals who are constantly disconnected with their internal events are not able to fully immerse themselves in enjoyable experiences; this results in less frequent positive life events, and hence, more negative emotions. Furthermore, thought suppression has been found to have negative rebound effects, in a way that suppressed thoughts are likely to be encountered more frequently following attempts to eliminate them from awareness (Wegner, Schneider, Carter, & White, 1987). It may be via these mechanisms that experiential avoidance engagement could cause depressive symptoms to develop, or be exacerbated.

The link between experiential avoidance and depression can be strengthened by considering the impact of Acceptance and Commitment Therapy (ACT) (Hayes, Strosahl, & Wilson, 1999) upon patients with depression. ACT is a 'third wave' cognitive therapy, which acts upon the premise that psychological problems are caused by experiential avoidance and cognitive fusion. Unlike traditional Cognitive Behavioural Therapy (CBT), ACT does not try to change maladaptive thought patterns; instead, it teaches individuals to notice and accept private events by encouraging cognitive defusion of thoughts, and greater psychological flexibility in regards to their internal events. These methods are designed to help the individual to deal with painful thoughts more effectively. Evidence for this therapy has shown that it is highly effective in treating depression (e.g. Zettle & Hayes, 1986; Forman, Herbert, Moitra, Yeomans, & Geller, 2007).

Experiential Avoidance: A Mediator?

        As previously stated, both attributional style and experiential avoidance have been linked to depression. This research project aims to study these associations in greater detail by considering the function that experiential avoidance might play in the relationship between attributional style and depression. It is theoretically sound to examine the role of experiential avoidance in this context for many reasons. Firstly, studies on the relationship between attributional style and depression have yielded inconsistent results (Coyne & Gotlib, 1983). This suggests that there may be one or more other variables involved in this relationship, which vary from study to study; one of which could be individual differences in experiential avoidance. For example, in individuals with negative attributional styles, high levels of experiential avoidance are likely to contribute to these individuals developing depression. On the other hand, some individuals with a negative attributional style may demonstrate low levels of experiential avoidance, and instead of suppressing feelings are able to recognize and accept private events. In this instance, high psychological flexibility (or low levels of experiential avoidance) may function as a protective factor against depression, despite the individuals having negative attributional styles. This example illustrates how individual differences in experiential avoidance levels could explain why some studies have shown strong links between attributional style and depression, whilst others have not.

Another reason to consider whether experiential avoidance is involved in the relationship between attributional style and depression is because it has been noted as a mediating factor in other relationships. A study by Merwin, Rosenthal and Coffey (2009) found that the relationship between sexual abuse and depression was mediated by experiential avoidance. Interestingly, studies have shown that individuals with a history of sexual abuse are likely to have negative attributional styles (e.g. Gold, 1986; Runyon & Kenny, 2002). Taken together, these two studies illustrate that it is appropriate to consider experiential avoidance as an important concept in determining the link between attributional style and depression.

On the basis of past research, this study hypothesized that experiential avoidance is a mediator of the relationship between attributional style and depression. As depression is a highly complex illness with numerous contributing factors, it was expected that experiential avoidance would only function as a partial mediator, rather than a full one. In order to determine whether these hypotheses were accurate, the three variables of attributional style, experiential avoidance and depression were measured by three questionnaires, and the results were statistically analyzed in line with Baron and Kenny's (1986) four-step approach for determining mediation.

Method

Design

        In a within-subjects design, three continuous variables were measured by the use of three online questionnaires, with no manipulation of variables. The criterion variable was the individual's depression levels as measured by the Beck Depression Inventory-II2 (Beck, Steer, & Brown, 1996), and the predictor variable was the individual's attributional styles, which were measured by an adapted version of the Attributional Style Questionnaire (Peterson et al., 1982). Experiential avoidance was used as a mediator, and was measured by the Acceptance and Action Questionnaire-II (Bond et al., 2010). Results were analysed with simple and multiple regressions, and an online Sobel test was utilized to assess the possible meditational effects of experiential avoidance.

Participants

        Seventy-four psychology undergraduate students (54 females, 20 males) from the University of Southampton, England, served as participants in this research in exchange for course credits. Participants were aged between 18 and 30 years old, with a mean age of 20.01 (SD = 1.79).

Measures

        Beck depression inventory-II. The Beck Depression Inventory (BDI-II) contains 21 self report questions which aim to determine an individual's depressive symptoms. Possible scores on this questionnaire range from 0 to 63, with a higher score indicating a higher level of depression. Used widely, the BDI-II has strong psychometric qualities, including a high level of internal consistency with college students (a = .93) and a high test-retest reliability (r = .93) (Beck et al., 1996).

        Adapted attributional style questionnaire. The Attributional Style Questionnaire (ASQ) is a self report measure which determines how individuals explain the events which occur in their lives, and their expectancies for future events. The original version includes 12 hypothetical events; six of these being negative, and six being positive. Within these two categories, three questions are based on academic events and three on interpersonal events. Individuals respond to these theoretical experiences by stating the one main cause of the event if it was to happen to them. Then participants rate the cause of the event on internality, stability, and globality scales, and they also indicate how important the event would be if it happened to them3. The scales range from one to seven, with one indicating attributions being made to external, unstable causes, with a lack of globality; and a rating of seven indicating that attributions are made to internal, stable and global causes. Higher ratings on the importance question indicate that the individual perceives the event as very important.

        The original ASQ was adapted to fit the requirements of the present study. This is because the ASQ was used alongside the Acceptance and Action Questionnaire, for which the focus is on how much individuals avoid or suppress negative thoughts or events, rather than positive ones. Therefore, the positive events were omitted from the ASQ, and only the negative events were used to assess attributional style. This adapted version is therefore being referred to in the present study as the Adapted Attributional Style Questionnaire (AASQ). The authors of the ASQ stated that as the two groups of events (good and bad) are independent, it does not affect the validity of the questionnaire to divide the sections and to only impose one group. The six bad events from the questionnaire were found to have an acceptable test-retest reliability (r = .64) (Peterson et al., 1982).

        Acceptance and action questionnaire-II. The Acceptance and Action Questionnaire (AAQ-II) consists of 10 self-report questions which determine how much emotional avoidance and thought suppression a person engages in, as well as whether or not the person takes action when negatively evaluated thoughts are encountered. The individuals respond to each of the questions on a seven point scale, where one represents 'never true' and seven represents 'always true'. Therefore, the possible scores on the AAQ-II range from 0-70. Items 2, 3, 4, 5, 7, 8, and 9 were reverse scored so that, for all questions, a higher score represented higher psychological flexibility, and lower levels of experiential avoidance. The 10-item scale was selected because it has been found to have greater psychometric qualities than older versions of the AAQ; it has a good reliability of between .81 and .87 (Bond et al., 2010) and a high level of internal consistency (a = .89) (McCracken & Zhao-O'Brien, 2010).

Procedure

        Participants read an online consent form which contained information on the procedure of the experiment. As there was no deception involved in this research, participants were able to give fully informed consent after reading this form. If participants were happy to take part then they indicated this by ticking a box at the bottom of the webpage. Demographic details on age and gender were then taken. Participants were presented with three questionnaires simultaneously; these were the BDI-II, AASQ, and the AAQ-II. Upon completion, student identification numbers were reported so that participants could be rewarded with course credits. Finally, an online debriefing form was shown to participants, which detailed the aims and hypotheses of the experiment, and also provided contact details of the researcher if they had any questions. Participants were then free to close the online study page.

Results

Initial Analyses

        Before analyzing the data, statistical tests were conducted to assess whether the data fitted the parametric assumptions. One sample Kolmogorow-Smirov tests were conducted on the BDI-II, AASQ, and the AAQ-II scores to compare the data to a normal distribution. All three variables were normally distributed. Therefore, no items needed to be excluded from the data set. Furthermore, both age and gender had insignificant correlations with attributional style, experiential avoidance, and depression, indicating that the data did not need to be analyzed separately for males and females, or for participants of different ages. Furthermore as the AASQ was altered slightly, a reliability analysis was conducted for this measure; the results showed an adequate level of internal consistency for the questionnaire (a = .72).

        In multiple regression analysis, multicollinearity is sometimes an issue. If the two predictor variables are highly correlated, then it is difficult to identify the individual contribution of one variable separately to the other variable, for their effects on the criterion variable. Therefore this needed to be tested between the two predictor values; AASQ and AAQ-II. Statistical tests showed that multicollinearity did not exist between the two variables.

Correlations

        Significant correlations were found between all combinations of the predictor variable (AASQ), mediator variable (AAQ-II) and the criterion variable (BDI-II). Therefore it was reasonable to assume that experiential avoidance may have had a meditational effect on the relationship between attributional style and depression.

Regression Analysis

        Firstly, simple regression analyses were conducted to determine whether experiential avoidance and attributional style were predictive of depression levels. The results from these regression models are presented in Tables 2 and 3 (respectively), which show that both variables were significant predictors of depression. Therefore, as attributional style becomes more negative, or experiential avoidance levels increase, depressive symptoms worsen. In step 1 of a multiple regression model (see Table 3) AASQ explained a total of 31% of the variance in depression levels. This value increased to 54% in step 2, meaning that adding AAQ-II into the model accounted for an extra 23% of the variance in depression levels.

Analysis for Mediational Effects

        The present study aimed to address the question of whether experiential avoidance acts as a mediator of the relationship between attributional style and depression. Baron and Kenny (1986) suggest a four step approach to determine mediation. Firstly, the predictor variable needs to be predictive of the criterion variable. As illustrated by Table 3, regression analysis showed that this condition was met, with attributional style being a significant predictor of depression (p <.001). This indicated that there was an effect to be mediated. The second of Baron and Kenny's steps is establishing that the predictor variable predicts the mediator. Regression analysis also showed this to be true (Table 2), with attributional style being a unique predictor of experiential avoidance (p <.001). The third step is confirming that the mediator affects the criterion variable when including both the predictor variable and the mediator in the analysis. A multiple regression model, presented in Table 3, showed this to be true, with experiential avoidance being predictive of depression independent of attributional style (p <.001). Finally, Baron and Kenny state that if complete mediation is present, the relationship between the predictor variable and the criterion variable should decrease to zero when controlling for the mediator. A partial mediation is present if the relationship between the predictor and criterion variables decreases when taking into account the mediator, but remains significant. This final step required the testing of the indirect effect; or in other words, examining the relationship between attributional style and depression whilst taking experiential avoidance into account.

        The Sobel test (Sobel, 1982) was used to test the indirect effect. There are many online versions of this test, but the one used in the present study was Medgraph (Jose, 2003), which is designed to graphically illustrate meditational effects. The relevant regression coefficients and standard errors were entered into this program which determined whether experiential avoidance acted as a mediator. The Medgraph program reported a significant Sobel test value for the indirect route (z = 3.56, p <.001). The coefficient for the direct route (attributional style predicting depression) decreased from ฿ = .56 to ฿ = .31 when the indirect route was tested. The results indicate that a partial mediation is present.

Discussion

        This study aimed to examine the role of experiential avoidance in the relationship between attributional style and depression. It was hypothesized that experiential avoidance would function as a mediator in this context. For example, in those individuals with negative attributional styles, their likelihood of developing depression would depend on how much experiential avoidance they engaged in. It was therefore expected that when taking experiential avoidance into account, the relationship between attributional style and depression would become insignificant, or it would decrease in magnitude but remain significant. The results indicated that the latter option was true; experiential avoidance functioned as a partial mediator in the relationship; when controlling for experiential avoidance, the relationship between attributional style and depression fell by 45%. This indicates that the hypothesis of the study was accurate. These results have many implications in terms of theoretical insight, treatments for depression, and future research ideas.

Implications

        This study provides further evidence that experiential avoidance plays an important role in depression, and has shown yet another area in which the construct acts as a mediator. The results of this study suggest that high psychological flexibility may function as a protective factor against the development of depression, despite an individual having other cognitive disturbances which are predictive of depression; such as a negative attributional style, in this case. This finding ties in with other studies which have looked at experiential avoidance as a mediator. For example, having low levels of experiential avoidance has also been noted as a protective factor against other clinical problems such as anxiety (Berman, Wheaton, McGrath, & Abramowitz, 2010).

        As experiential avoidance acts as a partial mediator between attributional style and depression, this may impact on the success of depression treatments. An individual who engages in experiential avoidance is likely to be at a disadvantage when it comes to therapy, as he or she might find it difficult to connect with the therapist in a reciprocal relationship (Horvath & Bedi, 2002), or to engage in the behavioural strategies which are used to invalidate maladaptive thought patterns (J. S. Beck, 1995). The results from the present study help to illustrate this problem. For example, Cognitive Behavioural Therapy (CBT) is effective in treating depression (Hollon, 1990), and studies have shown that CBT can significantly improve an individual's attributional style (Seligman et al., 1988). However, because experiential avoidance may act as a barrier to recovery, as implied by past research (Berking, Neacsiu, Comtois, & Linehan, 2009), this might limit the success of the treatment because traditional CBT does not target experiential avoidance. The findings from this study therefore highlight the value of therapists paying special attention to a patient's experiential avoidance engagement when diagnosing and treating depression; and they also underscore the importance of 'third wave' cognitive therapies such as ACT.

        Additionally, the results from this study help us to gain insight into the complicated mechanisms of depression. As experiential avoidance was found to be a partial mediator in the relationship between attributional style and depression, this indicates that other factors may be functioning in this relationship as well. Experiential avoidance is only one mediator which has been found to be involved. Past research has focused on many other factors in relation to depression, such as psychological rumination (Nolen-Hoeksema, 1991) and social problem solving ability (Henry, 2005), and so by combining the findings from the present study with previous research, it seems highly likely that depression levels are mediated by multiple factors, in a multifaceted relationship. Although it would be difficult to take into consideration all of the possible risk factors for depression in just one experiment, the results from this study add to the growing repertoire of information regarding the nature of depression.

Limitations

        Despite significant results, there are many drawbacks of the design used in the study at hand. To begin with, the questionnaires adopted for this research have associated weaknesses. Beck et al. (1986) warn against the use of the BDI-II for diagnostic purposes as many of the symptoms observed through the questionnaire are also symptoms of other disorders; the co-morbidity between depression and other disorders means that, in some instances, it may not have been depression which was being measured, but other disorders instead. Furthermore, the attributional style questionnaire was shortened to fit with the requirements of the study at hand by only using the six negative events; therefore, it could have benefitted from having a larger range of negative events on which to measure people's attributions. All three of the questionnaires were also self-report, which leaves the measurements vulnerable to weaknesses such as interpretation errors, or social desirability.

        In addition to these weaknesses, cause and effect relationships cannot be determined through the quantitative research used in this study. Although a significant effect of experiential avoidance was found, this by no means implies that experiential avoidance has a causal effect in the relationship between attributional style and depression. As the present study researched a topic which has not been investigated thus far in the literature, the results here require clarification in order to gain more scientific value.

Future Research

        Due to the weaknesses of the current study, there are some improvements which could be made in future research. It would be useful to expand on the findings shown in this experiment by replicating the study with a larger sample size, and by utilizing a different measure of attributional style; such as the Expanded Attributional Style Questionnaire (EASQ) (Peterson & Villanova, 1988) which consists of 18 negative events. By doing this it would help to verify these preliminary findings into the topic of attributional style, experiential avoidance, and depression. Furthermore, to understand the findings of this study better, and to perhaps develop cause and effect relationships, qualitative research could be conducted with depressed individuals to help develop a more in depth understanding of how this partial meditational relationship might function.

        Finally, as experiential avoidance functions as a partial mediator, rather than a full mediator, it suggests that other mechanisms are at work in the relationship between attributional style and depression. Perhaps future research could determine which other factors mediate the link; for example, whether social problems solving ability or rumination, as previously mentioned, are also involved. Conducting research into areas such as these may help to reveal other factors which mediate the relationship in question. This kind of research would have positive implications for the further development of effective treatments for depression.

Conclusion

        The results from this experiment confirmed the hypothesis of the study; that experiential avoidance acts as a partial mediator in the relationship between attributional style and depression. Experiential avoidance appears to be an important contributing factor to depression levels within the context of attributional style, which therapists should be attentive to when treating depression.

Doing your resits? We can help!