This article was downloaded by: [Heriot-Watt University] On: 05 January 2015, At: 03:08 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
The Journal of General Psychology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/vgen20
Unique Contributions of Metacognition and Cognition to Depressive Symptoms a
b
c
Adviye Esin Yilmaz , Tülin Gençöz & Adrian Wells a
Dokuz Eylül University
b
Middle East Technical University
c
University of Manchester, Manchester Mental Health, Social Care NHS Trust, and NTNU Trondheim Published online: 24 Dec 2014.
Click for updates To cite this article: Adviye Esin Yilmaz, Tülin Gençöz & Adrian Wells (2015) Unique Contributions of Metacognition and Cognition to Depressive Symptoms, The Journal of General Psychology, 142:1, 23-33, DOI: 10.1080/00221309.2014.964658 To link to this article: http://dx.doi.org/10.1080/00221309.2014.964658
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages,
and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
The Journal of General Psychology, 2015, 142(1), 23–33 C 2015 Taylor & Francis Group, LLC Copyright
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Unique Contributions of Metacognition and Cognition to Depressive Symptoms ˙ ADVIYE ES˙IN YILMAZ Dokuz Eyl¨ul University ¨ ˙IN GENC¸OZ ¨ TUL Middle East Technical University ADRIAN WELLS University of Manchester Manchester Mental Health and Social Care NHS Trust NTNU Trondheim
ABSTRACT. This study attempts to examine the unique contributions of “cognitions” or “metacognitions” to depressive symptoms while controlling for their intercorrelations and comorbid anxiety. Two-hundred-and-fifty-one university students participated in the study. Two complementary hierarchical multiple regression analyses were performed, in which symptoms of depression were regressed on the dysfunctional attitudes (DAS-24 subscales) and metacognition scales (Negative Beliefs about Rumination Scale [NBRS] and Positive Beliefs about Rumination Scale [PBRS]). Results showed that both NBRS and PBRS individually explained a significant amount of variance in depressive symptoms above and beyond dysfunctional schemata while controlling for anxiety. Although dysfunctional attitudes as a set significantly predicted depressive symptoms after anxiety and metacognitions were controlled for, they were weaker than metacognitive variables and none of the DAS24 subscales contributed individually. Metacognitive beliefs about ruminations appeared to contribute more to depressive symptoms than dysfunctional beliefs in the “cognitive” domain. Keywords: cognition, depression, depressive symptoms, dysfunctional attitudes, metacognition, schema
A REVIEW OF THE LITERATURE DEMONSTRATES that there is a range of beliefs considered to be central in the development and maintenance of depressive symptoms. Cognitive theories of depression, such as schema theory (Beck, 1976), emphasize the importance of beliefs and dysfunctional attitudes. According to Address correspondence to Adviye Esin Yılmaz, PhD, Dokuz Eyl¨ul University, Department of Psychology, Izmir, 35260 Turkey; [email protected] (e-mail). 23
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
24
The Journal of General Psychology
this model, activation of dysfunctional schemata concerning achievement, dependency, and self-control leads to negative automatic thoughts and distortions in interpretations. So, for example, a belief such as: “I am a failure” once activated will lead to negative thoughts and emotion. Based on this formulation, the main aim of cognitive therapy is to identify, evaluate, and modify the content of negative thoughts and beliefs, and thereby, to change associated depressive mood and behavior. There is a large amount of empirical support for the clinical effectiveness of this approach to psychological disorder (e.g., Butler, Chapman, Forman, & Beck, 2006). More recently some of the limitations of the schemata construct as a basis for formulating cognitive styles in psychological disorder have been described (Wells, 2009; Wells & Matthews, 1994). In particular, Wells argues that the content of cognition (for instance the belief, “I am a failure”) is not the factor that controls processing. After all, an individual can choose to react to such cognitions in different ways, presumably with different emotional outcomes. Wells proposed that a separate level of higher-order cognition (i.e., metacognition) is required in order to model disorder effectively. Metacognition refers to the aspect of the information processing system that monitors, appraises, and controls thinking (Flavell, 1979). Wells’ metacognitive model asserts that psychological disorder results from the activation of a generic thinking style called the cognitive attentional syndrome (CAS), consisting of prolonged negative processing in the form of worry and rumination, maintaining attention on threat and ironic mental control strategies. In response to a negative thought or belief the CAS prolongs and intensifies negative ideas and negative affective experience. The CAS is seen as a self-regulatory strategy and is linked to metacognitive knowledge. Whilst this knowledge can be formulated and communicated propositionally as beliefs (e.g., “I must analyze all my failings in order to find an answer”), it is thought to include procedures or programmes for controlling thinking. The metacognitive model specifically identifies two subtypes of metacognitive beliefs involved in disorder: positive beliefs concerning the value of extended thinking and negative beliefs about the uncontrollability, importance, and meaning of cognition. Distinctions between the cognitive and metacognitive approaches are that the former links disorder to the content of thoughts and beliefs whilst the latter links disorder to extended thinking typically in the form or rumination/worry. Both approaches view higher order structures as involved in controlling cognition, but in cognitive theory these structures consist of beliefs about things, whilst in metacognitive theory they consist of a separate structure of beliefs about cognition itself. Empirical evidence for the role of metacognition in depression is steadily accumulating, supporting central predictions in the model. Papageorgiou and Wells’ studies (2001, 2003) conducted on both clinical and non-clinical samples, showed that positive beliefs about rumination were significantly and positively correlated with rumination and severity of depression. Negative metacognitive beliefs concerning rumination, as well as worry, were also demonstrated to be positively and
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Yılmaz et al.
25
significantly associated with depression and anxiety when the overlap between depression and anxiety was controlled (Papageorgiou & Wells, 1999), supporting the idea that depressive individuals negatively appraise their thinking processes, and this is independent of anxiety. An important question that remains to be addressed is whether metacognitive beliefs enhance the explanation of depression beyond cognitive theory constructs focusing on the content of schemas. Although a growing number of studies have supported aspects of the metacognitive model of depression, no studies so far have sought to examine the relative contributions of cognition and metacognition to depression. Thus, the aim of the present study was to undertake a preliminary investigation of the unique contributions of “cognitions” or “metacognitions” to depression while controlling for their intercorrelations and comorbid anxiety. The reason for controlling anxiety was to examine the association of cognitive and metacognitive variables with pure depressive symptomatology without contamination of comorbid anxiety symptoms. Following the metacognitive model of depression, it was hypothesized that metacognitive beliefs about rumination would be associated with depression independently of dysfunctional attitudes, after anxiety symptoms are controlled. Complementary to this hypothesis, it was also hypothesized that metacognitions would be more strongly associated with the depressive symptoms than dysfunctional attitudes because they were expected to have a more direct link to controlling thinking styles. Method Participants Two-hundred-and-fifty-one undergraduate and postgraduate students participated in the study. There were 164 females (65.3%) and 85 males (33.9%), with two participants not indicating their gender. The age of the sample ranged from 17 to 59 years, with a mean of 22.5 (SD = 5.0). Instruments Negative Beliefs About Rumination Scale (NBRS) The NBRS (Papageorgiou, Wells, & Meina, 2008) is a 13-item measure that assesses negative metacognitive beliefs about rumination. It consists of two subscales, NBRS1 and NBRS2. NBRS1 includes 8 items and assesses negative metacognitive beliefs about uncontrollability and harm associated with rumination (for instance, “I cannot stop myself from ruminating”; “Ruminating makes me physically ill”). NBRS2 which is composed of 5 items, assesses negative metacognitive beliefs about the interpersonal and social consequences of rumination (for instance, “People will reject me if I ruminate”; “Ruminating will turn me into a failure”). Respondents indicate their agreement with each item on a 4-piont rating scale ranging from do not agree (1) to agree very much (4). The internal consistency and test-retest reliability coefficients of the NBRS1 and NBRS2 were reported as .80 and .83, and .66 and .68, respectively (Papageorgiou & Wells,
26
The Journal of General Psychology
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
2003). In the present study, internal consistency of the whole scale was found to be .81. Positive Beliefs About Rumination Scale (PBRS) The PBRS (Papageorgiou & Wells, 2001) is a 9-item scale that assesses positive metacognitive beliefs about benefits and advantages of rumination. Example items include “Ruminating about the past helps me to prevent future mistakes and failures,” and “I need to ruminate about my problems to find answers to my depression.” Respondents indicate their agreement with each item on a 4-piont rating scale ranging from do not agree (1) to agree very much (4). The internal consistency and test-retest reliability coefficients of the scale were found to be .89 and .85, respectively (Papageorgiou & Wells, 2001). It has been shown that the PBRS is a psychometrically sound measure that possesses good reliability and validity in non-clinical and clinical populations (Papageorgiou & Wells, 2001). In the present study, Cronbach’s alpha coefficient was .94. Dysfunctional Attitude Scale-24 (DAS-24) The DAS-24 (Power, Katz, McGuffin, Duggan, Lam, & Beck, 1994) is a 24-item brief measure of depressogenic dysfunctional beliefs or assumptions. It was derived from the Forms A and B of the Dysfunctional Attitude Scale (DAS; Weissman & Beck, 1978; Weissman, 1979). As identical with the full versions of DAS-A and DAS-B, the short version of dysfunctional attitude scale has three subscales: (1) Achievement, (2) Dependency, and (3) Self-Control. Example items include “If I fail partly, it is as bad as being a complete failure”; and “My life is wasted unless I am a success” for achievement; “If others dislike you, you cannot be happy”; and “What other people think about me is very important” for dependency; and “I should be happy all the time”; and “I should always have complete control over my feelings” for self-control subscales. Acceptable internal consistency values for these subscales (.85, .74, and .68, respectively) were reported (Power et al., 1994). In the present study, internal consistency for the total scores was found to be .88, while the coefficients were .86, .81, and .69 for the achievement, dependency, and self-control subscales, respectively. Beck Depression Inventory (BDI) The BDI (Beck, Rush, Shaw, & Emery, 1979) is comprised of 21 items that assess the symptoms of depression. Each item is scored from 0 to 3, with higher scores indicating greater depressive symptoms. Total scores are obtained by summating all items and range from 0 to 63. The BDI has well-established reliability that its mean coefficient alpha across 25 years of studies was reported as .86 in psychiatric populations and .81 in non-psychiatric populations (Beck, Steer, & Garbin, 1988).
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Yılmaz et al.
27
Beck Anxiety Inventory (BAI) The BAI (Beck, Epstein, Brown, & Steer, 1988) consists of 21 items and is a 4-point Likert type measure of cognitive and somatic symptoms of anxiety. Scores range from 0 to 63, with higher scores indicating higher levels of anxiety. Good internal consistency and high short-term test-retest reliability has been demonstrated in mixed psychiatric samples and patients with anxiety disorders (Beck et al., 1988; de Beurs, Wilson, Chambless, Goldstein, & Feske, 1997), as well as non-clinical samples (e.g., Creamer, Foran, & Bell, 1995). As for concurrent and convergent validity, the BAI was found to be moderately correlated with anxiety (rs = .36 to .69) and depression (rs = .25 to .56) measures in psychiatric (Beck et al., 1988) and student samples (Osman, Kopper, Barrios, Osman, & Wade, 1997). Procedure Ethics approval for this research was granted by the ethics committee for research on humans. Participants were asked via the university mailing list whether they would like to be included in the present study. If they consented, they used a link to a Web site where they could find the information sheet, consent form, the questionnaires and demographic sheet. The necessary permissions to use the instruments online were obtained from the authorized individuals or institutes. The instruments were presented in a randomized order using a specific programming for the questionnaire link to eliminate the effect of sequencing. In other words, each participant filled in questionnaires in a different order defined by chance. The total administration time for the instruments was approximately 15 minutes. Results Screening for Data Prior to testing the main hypotheses of the study, variables were evaluated to determine whether assumptions of multivariate analyses were met. Exclusion of the univariate outliers from the data left 236 cases for the analyses, including 157 females (66.5%) and 78 males (33.1%). Descriptive Statistics and Correlational Analysis The intercorrelations, means and standard deviations of the NBRS, PBRS, BAI, BDI, DAS-24, and DAS-24 subscales are presented in Table 1. We also examined gender differences on these variables by means of independent samples t tests (see Table 1). Apart from anxiety, no significant differences between men and women on the study measures emerged as significant. The mean scores of women were significantly higher than that of men on the BAI. As can be seen from the correlation matrix, apart from the correlation between self-control and anxiety, all other correlations among variables were positive and statistically significant. Underscoring the necessity of controlling for anxiety in
.36∗∗ .28∗∗ —
.30∗∗ —
.36∗∗ .30∗∗ .89∗∗ —
4 .29∗∗ .20∗∗ .75∗∗ .51∗∗ —
5 .20∗∗ .14∗ .71∗∗ .57∗∗ .20∗∗ —
6 .55∗∗ .32∗∗ .35∗∗ .35∗∗ .31∗∗ .14∗ —
7 .43∗∗ .20∗∗ .24∗∗ .27∗∗ .17∗∗ .10 .56∗∗ —
8 18.95 (5.13) 18.50 (6.82) 89.50 (20.41) 28.92 (9.64) 30.29 (8.87) 30.28 (7.28) 7.07 (6.02) 8.91 (7.40)
Total (sd) 18.70 (4.97) 18.12 (6.60) 88.38 (20.02) 28.11 (9.38) 30.57 (8.86) 29.70 (6.95) 7.44 (5.85) 9.76 (7.67)
Female (sd)
19.45 (5.47) 19.28 (7.28) 92.06 (21.05) 30.76 (9.89) 29.88 (8.89) 31.43 (7.87) 6.42 (6.33) 7.23 (6.59)
Male (sd)
1.05 1.22 1.31 2.00 −0.57 1.72 −1.22 −2.49∗
t value
Note. NBRS = Negative Beliefs about Rumination Scale, PBRS = Positive Beliefs about Rumination Scale, DAS-24 = Dysfunctional Attitude Scale24, DAS-1 = Dysfunctional Attitude Scale-24-Achievement subscale, DAS-2 = Dysfunctional Attitude Scale-24-Dependency subscale, DAS-3 = Dysfunctional Attitude Scale-24-Self-control subscale, BDI = Beck Depression Inventory, BAI = Beck Anxiety Inventory. ∗∗ p < .01, ∗ p < .05
1. NBRS 2. PBRS 3. DAS 4. DAS-1 5. DAS-2 6. DAS-3 7. BDI 8. BAI
3
2
TABLE 1. Intercorrelations, Means, and Standard Deviations of the NBRS, PBRS, DAS-24, BDI and BAI (N = 236 for the Total Sample, N = 157 for Women, N = 78 for Men)
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
28 The Journal of General Psychology
Yılmaz et al.
29
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
exploring the cognitive and metacognitive predictors of depressive symptoms, anxiety was found to be significantly correlated with these factors and depressive symptoms. The relationships between metacognitive beliefs about rumination and dysfunctional attitudes were also significant. In addition, dysfunctional attitudes, namely achievement, dependency, and self-control, and metacognitive variables, namely positive and negative beliefs about rumination, were all significantly correlated with depressive symptomatology. Overview of Statistical Analysis for Testing Hypotheses Two complementary hierarchical multiple regression analyses were performed, in which depression (BDI) was regressed on the dysfunctional attitudes (DAS-24 subscales) and metacognitions (NBRS and PBRS) scales. In the first regression, comorbid anxiety (BAI) was entered on the first step, followed by the forced entry of the DAS-24 subscales (achievement, dependency, and self-control) as a set on step two. Metacognitive beliefs (NBRS and PBRS) were entered on step three. This regression was repeated by reversing steps 2 and 3 so that NBRS and PBRS were entered on step two to control for metacognitions and the DAS-24 subscales were entered together on the last step. In this way we could examine the additional variance explained by the block of DAS scales. Cognitions and Metacognitions in Predicting Depression In the first set with the BDI as the criterion variable, the BAI was entered on step 1 to control the effect of comorbid anxiety. Dysfunctional attitudes (DAS-24 subscales; achievement, dependency, and self-control) as a set were entered on step 2, followed by the forced entry of metacognitive beliefs (NBRS and PBRS) on step 3. As can be seen in Table 2, the full model accounted for 46% of the variance in depression (Multiple R = .68, F [6, 229] = 32.59, p < .001). On the first step, comorbid anxiety significantly contributed to the explained variance (R2 = .31, F [1, 234] = 104.03, p < .001). After controlling the overlap between depression and anxiety, dysfunctional attitudes explained significant additional variance in BDI score on step two (R2change = .06, F change [3, 231] = 7.78, p < .001). The individual contributions of DAS-dependency (β = .16, t [231] = 2.60, p < .05) and DAS-achievement (β = .16, t [231] = 2.13, p < .05) were significant on this step. On the last step, metacognitive beliefs about ruminations made a further significant contribution to the explained variance (R2change = .09, F change [2, 229] = 18.96, p < .001). Negative metacognitive beliefs about ruminations was the strongest predictor of the BDI (β = .30, t [229] = 5.25, p < .001), followed by positive metacognitive beliefs about ruminations (β = .12, t [229] = 2.21, p < .05), after the effects of comorbid anxiety and depressive schemata were partialled out. Moreover, the individual associations of DAS subscales with depressive symptoms were no longer significant, when metacognitions entered, indicating that only NBRS and PBRS were the individual predictors of BDI on this last step.
30
The Journal of General Psychology
TABLE 2. Statistics for the Regression Equations With BDI Regressed on BAI, DAS-24 Subscales (Achievement, Dependency, Self-Control), NBRS, and PBRS
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Variables Regression 1 Step 1: Anxiety BAI Step 2: Depressive Schemata Achievement Dependency Self-Control Step 3: Metacognitive Factors NBRS PBRS Regression 2 Step 1: Anxiety BAI Step 2: Metacognitive Factors NBRS PBRS Step 3: Depressive Schemata Achievement Dependency Self-Control Multiple R = .68∗∗∗
β (within set) t (within set)
.56 .16 .16 −.03
∗∗∗
10.20
2.13∗ 2.60∗ −.54
.30 .12
5.25∗∗∗ 2.21∗
.56
∗∗∗
.34 .14 .08 .11 −.04 Adjusted R2 = .45
10.20
6.05∗∗∗ 2.71∗∗ 1.10 1.94 −.71
df
Fchange
R2
1, 234 104.03∗∗∗ .31 234 3, 231 7.78∗∗∗ .37 231 231 231 2, 229 18.96∗∗∗ .46 229 229 1, 234 104.03∗∗∗ .31 234 2, 232 27.62∗∗∗ .44 232 232 .46 3, 229 2.79∗ 229 229 229
Note. BDI = Beck Depression Inventory, BAI = Beck Anxiety Inventory, DAS-24 = Dysfunctional Attitude Scale-24, NBRS = Negative Beliefs about Rumination Scale, PBRS = Positive Beliefs about Rumination Scale. ∗∗∗ p < .001, ∗∗ p < .01, ∗ p < .05
This regression was repeated by reversing steps 2 and 3 so that NBRS and PBRS were entered on step two to control for metacognitions and the DAS-24 subscales were entered together on the last step (see Table 2). On the second step, the metacognitive beliefs significantly increased the variance explained (R2change = .13, F change [2, 232] = 27.62, p < .001). On the third step, dysfunctional attitudes as a set made a significant but small contribution (R2change = .02, F change [3, 229] = 2.79, p < .05). Moreover, both NBRS and PBRS maintained their significance on this last step. This set of findings demonstrated that positive and negative metacognitions about rumination individually explained a significant amount of variance in depressive symptomatology above and beyond dysfunctional depressive schemata while controlling for anxiety. Although dysfunctional attitudes as a set significantly
Yılmaz et al.
31
predicted depression after anxiety and metacognitions were controlled, predictive values of this set of variables were weaker than that of metacognitive variables. Moreover, on the second regression analysis, none of the DAS-24 subscales contributed individually. In other words, when negative and positive metacognitions about ruminations were added to the regression equation dysfunctional schemas individually did not maintain their significance but the converse did apply.
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Discussion The aim of the study was to conduct a focused and theory driven investigation of the relationships between cognitive and metacognitive beliefs, and depressive symptoms. Specifically, this study aimed to focus on the additive and independent contribution of metacognitions to depressive symptomatology in relation to schemas. In so doing, the intercorrelations between anxiety and depression symptoms, and the predictor variables were controlled. In line with previous studies (e.g., Papageorgiou & Wells, 2003), the results of the correlational analyses demonstrated that higher levels of positive and negative metacognitions about rumination were associated with higher levels of depressive symptomatology. Similarly, the schemas of achievement, dependency, and selfcontrol were also positively correlated with depressive symptoms as found in previous studies (e.g., Beck et al., 1979; Power et al., 1994). Metacognitions were also correlated with achievement, dependency, and self-control. The results of the regression analyses showed that the block of dysfunctional schemas or metacognitions each predicted depressive symptoms on Step 2 when covariances with anxiety were controlled. The results of the final steps of the regressions demonstrated that higher levels of positive and negative metacognitions about ruminations were associated with higher levels of depressive symptoms irrespectively of the levels of achievement, dependency, and self-control. Although schemas as a set significantly predicted depression after anxiety and metacognitions were controlled, values of this set of variables were weaker than metacognitive variables. Moreover, none of the DAS-24 subscales contributed individually, indicating that schemas were not associated with depressive symptoms independently of metacognitions. Thus, positive and negative metacognitions about ruminations were determined as stronger predictors than dysfunctional depressive schemata. From these results it can be concluded that metacognitive beliefs may contribute more to depressive symptoms than do dysfunctional beliefs in the “cognitive” domain as measured by the DAS. In summary, the pattern of results obtained provide further support for the metacognitive model of depression and indicate that the metacognitive approach enhances our understanding of depression beyond cognitive theory focused on specific cognitive content. One of the differences between cognitive and metacognitive models of depression is the emphasis given to different types of beliefs. The present findings are consistent with metacognitive therapy of depression that
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
32
The Journal of General Psychology
focuses on the assessment and modification of positive and negative beliefs about rumination (Wells, Fisher, Myers, Wheatley, Patel, & Brewin, 2009, 2012). Some limitations of this study should be acknowledged. First, the use of different measurement devices for dysfunctional attitudes and depressive symptomatology may lead to different results. Thus, the relative contributions of cognitions and metacognitions should also be tested using different metacognitive, cognitive and depressive measurement tools. Second, this study was not conducted with a clinical sample, and it may be the case that relationships between depression and the predictor variables may be different in clinical samples. For instance, negative schemas may be better formed and exert a greater influence in patients. In addition, the low mean depression score of the present sample indicates that the range of scores was probably restricted. Although this is not surprising given the student profile of the sample, this would weaken any relationships with other variables. Thus, the study should be repeated in both clinical and non-clinical samples in order to produce a much wider range of scores, to compare the differences, and to evaluate the clinical generalizability of these relationships. AUTHOR NOTES Adviye Esin Yılmaz is assistant professor of Clinical Psychology at the Psychology Department of Dokuz Eyl¨ul University, Turkey. Her research focuses on cognitive theory and therapy of anxiety and mood disorders and scale adaptation. ¨ Genc¸o¨ z is professor of Clinical Psychology at the Psychology Department Tulin of Middle East Technical University, Turkey. She has published several articles in international journals about antecedents, concomitants, and consequences of psychological problems; personality traits and Big Five model of Personality; information processing approaches in clinical psychology; and scale adaptation. Adrian Wells is professor of Clinical and Experimental Psychopathology at University of Manchester (UK). He has published extensively in the areas of cognitive theory and therapy of anxiety and mood disorders. He is the originator of Metacognitive Therapy and his treatments have been included in NICE guidelines. FUNDING This research has been supported by the Scientific and Technological Research Council of Turkey (TUBITAK), in the framework of the 2214-Abroad Research Grant Program.
REFERENCES Beck, A. T. (1976). Cognitive therapy and emotional disorders. New York, NY: International Universities Press.
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Yılmaz et al.
33
Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety: Psychometric properties. Journal of Consulting and Clinical Psychology, 56, 893–897. Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy for depression. New York, NY: Guilford. Beck, A. T., Steer, R. A., & Garbin, M. A. (1988). Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review, 8, 77–100. Butler, A. C., Chapman, J. E., Forman, E. M., & Beck, A. T. (2006). The empirical status of cognitive-behavioral therapy: A review of meta-analyses. Clinical Psychology Review, 26, 17–31. Creamer, M., Foran, J., & Bell, R. (1995). The Beck Anxiety Inventory in a non-clinical sample. Behaviour Research and Therapy, 33, 477–485. de Beurs, E., Wilson, K. A., Chambless, D. L., Goldstein, A. J., & Feske, U. (1997). Convergent and divergent validity of the Beck Anxiety Inventory for patients with panic disorder and agoraphobia. Depression and Anxiety, 6, 140–146. Flavell, J. N. (1979). Metacognition and cognitive monitoring: A new area of cognitivedevelopmental inquiry. American Psychologist, 34, 906–911. Osman, A., Kopper, B. A., Barrios, F. X., Osman, J. R., & Wade, T. (1997). The Beck Anxiety Inventory: Reexamination of factor structure and psychometric properties. Journal of Clinical Psychology, 53, 7–14. Papageorgiou, C., & Wells, A. (1999). Process and meta-cognitive dimensions of depressive and anxious thoughts and relationships with emotional intensity. Clinical Psychology and Psychotherapy, 6, 156–162. Papageorgiou, C., & Wells, A. (2001). Positive beliefs about depressive rumination: Development and preliminary validation of a self-report scale. Behavior Therapy, 32, 13–26. Papageorgiou, C., & Wells, A. (2003). An empirical test of a clinical metacognitive model of rumination and depression. Cognitive Therapy and Research, 27, 261–273. Papageorgiou, C., Wells, A., & Meina, L. J. (2008) Development and preliminary validation of the Negative Beliefs about Rumination Scale. Manuscript in preparation. Power, M. J., Katz, R., McGuffin, P., Duggan, C. F., Lam, D., & Beck, A. T. (1994). The Dysfunctional Attitude Scale (DAS): A comparison of forms A and B and proposals for a new subscaled version. Journal of Research in Personality, 28, 263–276. Weissman, A. N. (1979). Assessing depressogenic attitudes: A validation study. Unpublished Thesis, University of Pennsylvania, Philadelphia, PA. Weissman, A. N. & Beck, A. T. (1978). Development and validation of the Dysfunctional Attitude Scale: A preliminary investigation. Paper presented at the meeting of the American Educational Research Association, Toronto, Ontario. Wells, A. (2009). Metacognitive therapy for anxiety and depression. New York, NY: Guilford. Wells, A. & Matthews, G. (1994). Attention and emotion: A clinical perspective. Hove UK: Erlbaum. Wells, A., Fisher, P., Myers, S., Wheatley, J., Patel, T., & Brewin, C. R. (2009). Metacognitive therapy in recurrent and persistent depression: A multiple-baseline study of a new treatment. Cognitive Therapy and Research, 33, 291–300. Wells, A., Fisher, P., Myers, S., Wheatley, J., Patel, T., & Brewin, C. R. (2012). Metacognitive therapy in treatment resistant depression: A platform trial. Behaviour Research and Therapy, 50, 367–373.
Original manuscript received February 13, 2014 Final version accepted September 8, 2014
The Journal of General Psychology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/vgen20
Unique Contributions of Metacognition and Cognition to Depressive Symptoms a
b
c
Adviye Esin Yilmaz , Tülin Gençöz & Adrian Wells a
Dokuz Eylül University
b
Middle East Technical University
c
University of Manchester, Manchester Mental Health, Social Care NHS Trust, and NTNU Trondheim Published online: 24 Dec 2014.
Click for updates To cite this article: Adviye Esin Yilmaz, Tülin Gençöz & Adrian Wells (2015) Unique Contributions of Metacognition and Cognition to Depressive Symptoms, The Journal of General Psychology, 142:1, 23-33, DOI: 10.1080/00221309.2014.964658 To link to this article: http://dx.doi.org/10.1080/00221309.2014.964658
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages,
and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
The Journal of General Psychology, 2015, 142(1), 23–33 C 2015 Taylor & Francis Group, LLC Copyright
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Unique Contributions of Metacognition and Cognition to Depressive Symptoms ˙ ADVIYE ES˙IN YILMAZ Dokuz Eyl¨ul University ¨ ˙IN GENC¸OZ ¨ TUL Middle East Technical University ADRIAN WELLS University of Manchester Manchester Mental Health and Social Care NHS Trust NTNU Trondheim
ABSTRACT. This study attempts to examine the unique contributions of “cognitions” or “metacognitions” to depressive symptoms while controlling for their intercorrelations and comorbid anxiety. Two-hundred-and-fifty-one university students participated in the study. Two complementary hierarchical multiple regression analyses were performed, in which symptoms of depression were regressed on the dysfunctional attitudes (DAS-24 subscales) and metacognition scales (Negative Beliefs about Rumination Scale [NBRS] and Positive Beliefs about Rumination Scale [PBRS]). Results showed that both NBRS and PBRS individually explained a significant amount of variance in depressive symptoms above and beyond dysfunctional schemata while controlling for anxiety. Although dysfunctional attitudes as a set significantly predicted depressive symptoms after anxiety and metacognitions were controlled for, they were weaker than metacognitive variables and none of the DAS24 subscales contributed individually. Metacognitive beliefs about ruminations appeared to contribute more to depressive symptoms than dysfunctional beliefs in the “cognitive” domain. Keywords: cognition, depression, depressive symptoms, dysfunctional attitudes, metacognition, schema
A REVIEW OF THE LITERATURE DEMONSTRATES that there is a range of beliefs considered to be central in the development and maintenance of depressive symptoms. Cognitive theories of depression, such as schema theory (Beck, 1976), emphasize the importance of beliefs and dysfunctional attitudes. According to Address correspondence to Adviye Esin Yılmaz, PhD, Dokuz Eyl¨ul University, Department of Psychology, Izmir, 35260 Turkey; [email protected] (e-mail). 23
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
24
The Journal of General Psychology
this model, activation of dysfunctional schemata concerning achievement, dependency, and self-control leads to negative automatic thoughts and distortions in interpretations. So, for example, a belief such as: “I am a failure” once activated will lead to negative thoughts and emotion. Based on this formulation, the main aim of cognitive therapy is to identify, evaluate, and modify the content of negative thoughts and beliefs, and thereby, to change associated depressive mood and behavior. There is a large amount of empirical support for the clinical effectiveness of this approach to psychological disorder (e.g., Butler, Chapman, Forman, & Beck, 2006). More recently some of the limitations of the schemata construct as a basis for formulating cognitive styles in psychological disorder have been described (Wells, 2009; Wells & Matthews, 1994). In particular, Wells argues that the content of cognition (for instance the belief, “I am a failure”) is not the factor that controls processing. After all, an individual can choose to react to such cognitions in different ways, presumably with different emotional outcomes. Wells proposed that a separate level of higher-order cognition (i.e., metacognition) is required in order to model disorder effectively. Metacognition refers to the aspect of the information processing system that monitors, appraises, and controls thinking (Flavell, 1979). Wells’ metacognitive model asserts that psychological disorder results from the activation of a generic thinking style called the cognitive attentional syndrome (CAS), consisting of prolonged negative processing in the form of worry and rumination, maintaining attention on threat and ironic mental control strategies. In response to a negative thought or belief the CAS prolongs and intensifies negative ideas and negative affective experience. The CAS is seen as a self-regulatory strategy and is linked to metacognitive knowledge. Whilst this knowledge can be formulated and communicated propositionally as beliefs (e.g., “I must analyze all my failings in order to find an answer”), it is thought to include procedures or programmes for controlling thinking. The metacognitive model specifically identifies two subtypes of metacognitive beliefs involved in disorder: positive beliefs concerning the value of extended thinking and negative beliefs about the uncontrollability, importance, and meaning of cognition. Distinctions between the cognitive and metacognitive approaches are that the former links disorder to the content of thoughts and beliefs whilst the latter links disorder to extended thinking typically in the form or rumination/worry. Both approaches view higher order structures as involved in controlling cognition, but in cognitive theory these structures consist of beliefs about things, whilst in metacognitive theory they consist of a separate structure of beliefs about cognition itself. Empirical evidence for the role of metacognition in depression is steadily accumulating, supporting central predictions in the model. Papageorgiou and Wells’ studies (2001, 2003) conducted on both clinical and non-clinical samples, showed that positive beliefs about rumination were significantly and positively correlated with rumination and severity of depression. Negative metacognitive beliefs concerning rumination, as well as worry, were also demonstrated to be positively and
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Yılmaz et al.
25
significantly associated with depression and anxiety when the overlap between depression and anxiety was controlled (Papageorgiou & Wells, 1999), supporting the idea that depressive individuals negatively appraise their thinking processes, and this is independent of anxiety. An important question that remains to be addressed is whether metacognitive beliefs enhance the explanation of depression beyond cognitive theory constructs focusing on the content of schemas. Although a growing number of studies have supported aspects of the metacognitive model of depression, no studies so far have sought to examine the relative contributions of cognition and metacognition to depression. Thus, the aim of the present study was to undertake a preliminary investigation of the unique contributions of “cognitions” or “metacognitions” to depression while controlling for their intercorrelations and comorbid anxiety. The reason for controlling anxiety was to examine the association of cognitive and metacognitive variables with pure depressive symptomatology without contamination of comorbid anxiety symptoms. Following the metacognitive model of depression, it was hypothesized that metacognitive beliefs about rumination would be associated with depression independently of dysfunctional attitudes, after anxiety symptoms are controlled. Complementary to this hypothesis, it was also hypothesized that metacognitions would be more strongly associated with the depressive symptoms than dysfunctional attitudes because they were expected to have a more direct link to controlling thinking styles. Method Participants Two-hundred-and-fifty-one undergraduate and postgraduate students participated in the study. There were 164 females (65.3%) and 85 males (33.9%), with two participants not indicating their gender. The age of the sample ranged from 17 to 59 years, with a mean of 22.5 (SD = 5.0). Instruments Negative Beliefs About Rumination Scale (NBRS) The NBRS (Papageorgiou, Wells, & Meina, 2008) is a 13-item measure that assesses negative metacognitive beliefs about rumination. It consists of two subscales, NBRS1 and NBRS2. NBRS1 includes 8 items and assesses negative metacognitive beliefs about uncontrollability and harm associated with rumination (for instance, “I cannot stop myself from ruminating”; “Ruminating makes me physically ill”). NBRS2 which is composed of 5 items, assesses negative metacognitive beliefs about the interpersonal and social consequences of rumination (for instance, “People will reject me if I ruminate”; “Ruminating will turn me into a failure”). Respondents indicate their agreement with each item on a 4-piont rating scale ranging from do not agree (1) to agree very much (4). The internal consistency and test-retest reliability coefficients of the NBRS1 and NBRS2 were reported as .80 and .83, and .66 and .68, respectively (Papageorgiou & Wells,
26
The Journal of General Psychology
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
2003). In the present study, internal consistency of the whole scale was found to be .81. Positive Beliefs About Rumination Scale (PBRS) The PBRS (Papageorgiou & Wells, 2001) is a 9-item scale that assesses positive metacognitive beliefs about benefits and advantages of rumination. Example items include “Ruminating about the past helps me to prevent future mistakes and failures,” and “I need to ruminate about my problems to find answers to my depression.” Respondents indicate their agreement with each item on a 4-piont rating scale ranging from do not agree (1) to agree very much (4). The internal consistency and test-retest reliability coefficients of the scale were found to be .89 and .85, respectively (Papageorgiou & Wells, 2001). It has been shown that the PBRS is a psychometrically sound measure that possesses good reliability and validity in non-clinical and clinical populations (Papageorgiou & Wells, 2001). In the present study, Cronbach’s alpha coefficient was .94. Dysfunctional Attitude Scale-24 (DAS-24) The DAS-24 (Power, Katz, McGuffin, Duggan, Lam, & Beck, 1994) is a 24-item brief measure of depressogenic dysfunctional beliefs or assumptions. It was derived from the Forms A and B of the Dysfunctional Attitude Scale (DAS; Weissman & Beck, 1978; Weissman, 1979). As identical with the full versions of DAS-A and DAS-B, the short version of dysfunctional attitude scale has three subscales: (1) Achievement, (2) Dependency, and (3) Self-Control. Example items include “If I fail partly, it is as bad as being a complete failure”; and “My life is wasted unless I am a success” for achievement; “If others dislike you, you cannot be happy”; and “What other people think about me is very important” for dependency; and “I should be happy all the time”; and “I should always have complete control over my feelings” for self-control subscales. Acceptable internal consistency values for these subscales (.85, .74, and .68, respectively) were reported (Power et al., 1994). In the present study, internal consistency for the total scores was found to be .88, while the coefficients were .86, .81, and .69 for the achievement, dependency, and self-control subscales, respectively. Beck Depression Inventory (BDI) The BDI (Beck, Rush, Shaw, & Emery, 1979) is comprised of 21 items that assess the symptoms of depression. Each item is scored from 0 to 3, with higher scores indicating greater depressive symptoms. Total scores are obtained by summating all items and range from 0 to 63. The BDI has well-established reliability that its mean coefficient alpha across 25 years of studies was reported as .86 in psychiatric populations and .81 in non-psychiatric populations (Beck, Steer, & Garbin, 1988).
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Yılmaz et al.
27
Beck Anxiety Inventory (BAI) The BAI (Beck, Epstein, Brown, & Steer, 1988) consists of 21 items and is a 4-point Likert type measure of cognitive and somatic symptoms of anxiety. Scores range from 0 to 63, with higher scores indicating higher levels of anxiety. Good internal consistency and high short-term test-retest reliability has been demonstrated in mixed psychiatric samples and patients with anxiety disorders (Beck et al., 1988; de Beurs, Wilson, Chambless, Goldstein, & Feske, 1997), as well as non-clinical samples (e.g., Creamer, Foran, & Bell, 1995). As for concurrent and convergent validity, the BAI was found to be moderately correlated with anxiety (rs = .36 to .69) and depression (rs = .25 to .56) measures in psychiatric (Beck et al., 1988) and student samples (Osman, Kopper, Barrios, Osman, & Wade, 1997). Procedure Ethics approval for this research was granted by the ethics committee for research on humans. Participants were asked via the university mailing list whether they would like to be included in the present study. If they consented, they used a link to a Web site where they could find the information sheet, consent form, the questionnaires and demographic sheet. The necessary permissions to use the instruments online were obtained from the authorized individuals or institutes. The instruments were presented in a randomized order using a specific programming for the questionnaire link to eliminate the effect of sequencing. In other words, each participant filled in questionnaires in a different order defined by chance. The total administration time for the instruments was approximately 15 minutes. Results Screening for Data Prior to testing the main hypotheses of the study, variables were evaluated to determine whether assumptions of multivariate analyses were met. Exclusion of the univariate outliers from the data left 236 cases for the analyses, including 157 females (66.5%) and 78 males (33.1%). Descriptive Statistics and Correlational Analysis The intercorrelations, means and standard deviations of the NBRS, PBRS, BAI, BDI, DAS-24, and DAS-24 subscales are presented in Table 1. We also examined gender differences on these variables by means of independent samples t tests (see Table 1). Apart from anxiety, no significant differences between men and women on the study measures emerged as significant. The mean scores of women were significantly higher than that of men on the BAI. As can be seen from the correlation matrix, apart from the correlation between self-control and anxiety, all other correlations among variables were positive and statistically significant. Underscoring the necessity of controlling for anxiety in
.36∗∗ .28∗∗ —
.30∗∗ —
.36∗∗ .30∗∗ .89∗∗ —
4 .29∗∗ .20∗∗ .75∗∗ .51∗∗ —
5 .20∗∗ .14∗ .71∗∗ .57∗∗ .20∗∗ —
6 .55∗∗ .32∗∗ .35∗∗ .35∗∗ .31∗∗ .14∗ —
7 .43∗∗ .20∗∗ .24∗∗ .27∗∗ .17∗∗ .10 .56∗∗ —
8 18.95 (5.13) 18.50 (6.82) 89.50 (20.41) 28.92 (9.64) 30.29 (8.87) 30.28 (7.28) 7.07 (6.02) 8.91 (7.40)
Total (sd) 18.70 (4.97) 18.12 (6.60) 88.38 (20.02) 28.11 (9.38) 30.57 (8.86) 29.70 (6.95) 7.44 (5.85) 9.76 (7.67)
Female (sd)
19.45 (5.47) 19.28 (7.28) 92.06 (21.05) 30.76 (9.89) 29.88 (8.89) 31.43 (7.87) 6.42 (6.33) 7.23 (6.59)
Male (sd)
1.05 1.22 1.31 2.00 −0.57 1.72 −1.22 −2.49∗
t value
Note. NBRS = Negative Beliefs about Rumination Scale, PBRS = Positive Beliefs about Rumination Scale, DAS-24 = Dysfunctional Attitude Scale24, DAS-1 = Dysfunctional Attitude Scale-24-Achievement subscale, DAS-2 = Dysfunctional Attitude Scale-24-Dependency subscale, DAS-3 = Dysfunctional Attitude Scale-24-Self-control subscale, BDI = Beck Depression Inventory, BAI = Beck Anxiety Inventory. ∗∗ p < .01, ∗ p < .05
1. NBRS 2. PBRS 3. DAS 4. DAS-1 5. DAS-2 6. DAS-3 7. BDI 8. BAI
3
2
TABLE 1. Intercorrelations, Means, and Standard Deviations of the NBRS, PBRS, DAS-24, BDI and BAI (N = 236 for the Total Sample, N = 157 for Women, N = 78 for Men)
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
28 The Journal of General Psychology
Yılmaz et al.
29
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
exploring the cognitive and metacognitive predictors of depressive symptoms, anxiety was found to be significantly correlated with these factors and depressive symptoms. The relationships between metacognitive beliefs about rumination and dysfunctional attitudes were also significant. In addition, dysfunctional attitudes, namely achievement, dependency, and self-control, and metacognitive variables, namely positive and negative beliefs about rumination, were all significantly correlated with depressive symptomatology. Overview of Statistical Analysis for Testing Hypotheses Two complementary hierarchical multiple regression analyses were performed, in which depression (BDI) was regressed on the dysfunctional attitudes (DAS-24 subscales) and metacognitions (NBRS and PBRS) scales. In the first regression, comorbid anxiety (BAI) was entered on the first step, followed by the forced entry of the DAS-24 subscales (achievement, dependency, and self-control) as a set on step two. Metacognitive beliefs (NBRS and PBRS) were entered on step three. This regression was repeated by reversing steps 2 and 3 so that NBRS and PBRS were entered on step two to control for metacognitions and the DAS-24 subscales were entered together on the last step. In this way we could examine the additional variance explained by the block of DAS scales. Cognitions and Metacognitions in Predicting Depression In the first set with the BDI as the criterion variable, the BAI was entered on step 1 to control the effect of comorbid anxiety. Dysfunctional attitudes (DAS-24 subscales; achievement, dependency, and self-control) as a set were entered on step 2, followed by the forced entry of metacognitive beliefs (NBRS and PBRS) on step 3. As can be seen in Table 2, the full model accounted for 46% of the variance in depression (Multiple R = .68, F [6, 229] = 32.59, p < .001). On the first step, comorbid anxiety significantly contributed to the explained variance (R2 = .31, F [1, 234] = 104.03, p < .001). After controlling the overlap between depression and anxiety, dysfunctional attitudes explained significant additional variance in BDI score on step two (R2change = .06, F change [3, 231] = 7.78, p < .001). The individual contributions of DAS-dependency (β = .16, t [231] = 2.60, p < .05) and DAS-achievement (β = .16, t [231] = 2.13, p < .05) were significant on this step. On the last step, metacognitive beliefs about ruminations made a further significant contribution to the explained variance (R2change = .09, F change [2, 229] = 18.96, p < .001). Negative metacognitive beliefs about ruminations was the strongest predictor of the BDI (β = .30, t [229] = 5.25, p < .001), followed by positive metacognitive beliefs about ruminations (β = .12, t [229] = 2.21, p < .05), after the effects of comorbid anxiety and depressive schemata were partialled out. Moreover, the individual associations of DAS subscales with depressive symptoms were no longer significant, when metacognitions entered, indicating that only NBRS and PBRS were the individual predictors of BDI on this last step.
30
The Journal of General Psychology
TABLE 2. Statistics for the Regression Equations With BDI Regressed on BAI, DAS-24 Subscales (Achievement, Dependency, Self-Control), NBRS, and PBRS
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Variables Regression 1 Step 1: Anxiety BAI Step 2: Depressive Schemata Achievement Dependency Self-Control Step 3: Metacognitive Factors NBRS PBRS Regression 2 Step 1: Anxiety BAI Step 2: Metacognitive Factors NBRS PBRS Step 3: Depressive Schemata Achievement Dependency Self-Control Multiple R = .68∗∗∗
β (within set) t (within set)
.56 .16 .16 −.03
∗∗∗
10.20
2.13∗ 2.60∗ −.54
.30 .12
5.25∗∗∗ 2.21∗
.56
∗∗∗
.34 .14 .08 .11 −.04 Adjusted R2 = .45
10.20
6.05∗∗∗ 2.71∗∗ 1.10 1.94 −.71
df
Fchange
R2
1, 234 104.03∗∗∗ .31 234 3, 231 7.78∗∗∗ .37 231 231 231 2, 229 18.96∗∗∗ .46 229 229 1, 234 104.03∗∗∗ .31 234 2, 232 27.62∗∗∗ .44 232 232 .46 3, 229 2.79∗ 229 229 229
Note. BDI = Beck Depression Inventory, BAI = Beck Anxiety Inventory, DAS-24 = Dysfunctional Attitude Scale-24, NBRS = Negative Beliefs about Rumination Scale, PBRS = Positive Beliefs about Rumination Scale. ∗∗∗ p < .001, ∗∗ p < .01, ∗ p < .05
This regression was repeated by reversing steps 2 and 3 so that NBRS and PBRS were entered on step two to control for metacognitions and the DAS-24 subscales were entered together on the last step (see Table 2). On the second step, the metacognitive beliefs significantly increased the variance explained (R2change = .13, F change [2, 232] = 27.62, p < .001). On the third step, dysfunctional attitudes as a set made a significant but small contribution (R2change = .02, F change [3, 229] = 2.79, p < .05). Moreover, both NBRS and PBRS maintained their significance on this last step. This set of findings demonstrated that positive and negative metacognitions about rumination individually explained a significant amount of variance in depressive symptomatology above and beyond dysfunctional depressive schemata while controlling for anxiety. Although dysfunctional attitudes as a set significantly
Yılmaz et al.
31
predicted depression after anxiety and metacognitions were controlled, predictive values of this set of variables were weaker than that of metacognitive variables. Moreover, on the second regression analysis, none of the DAS-24 subscales contributed individually. In other words, when negative and positive metacognitions about ruminations were added to the regression equation dysfunctional schemas individually did not maintain their significance but the converse did apply.
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Discussion The aim of the study was to conduct a focused and theory driven investigation of the relationships between cognitive and metacognitive beliefs, and depressive symptoms. Specifically, this study aimed to focus on the additive and independent contribution of metacognitions to depressive symptomatology in relation to schemas. In so doing, the intercorrelations between anxiety and depression symptoms, and the predictor variables were controlled. In line with previous studies (e.g., Papageorgiou & Wells, 2003), the results of the correlational analyses demonstrated that higher levels of positive and negative metacognitions about rumination were associated with higher levels of depressive symptomatology. Similarly, the schemas of achievement, dependency, and selfcontrol were also positively correlated with depressive symptoms as found in previous studies (e.g., Beck et al., 1979; Power et al., 1994). Metacognitions were also correlated with achievement, dependency, and self-control. The results of the regression analyses showed that the block of dysfunctional schemas or metacognitions each predicted depressive symptoms on Step 2 when covariances with anxiety were controlled. The results of the final steps of the regressions demonstrated that higher levels of positive and negative metacognitions about ruminations were associated with higher levels of depressive symptoms irrespectively of the levels of achievement, dependency, and self-control. Although schemas as a set significantly predicted depression after anxiety and metacognitions were controlled, values of this set of variables were weaker than metacognitive variables. Moreover, none of the DAS-24 subscales contributed individually, indicating that schemas were not associated with depressive symptoms independently of metacognitions. Thus, positive and negative metacognitions about ruminations were determined as stronger predictors than dysfunctional depressive schemata. From these results it can be concluded that metacognitive beliefs may contribute more to depressive symptoms than do dysfunctional beliefs in the “cognitive” domain as measured by the DAS. In summary, the pattern of results obtained provide further support for the metacognitive model of depression and indicate that the metacognitive approach enhances our understanding of depression beyond cognitive theory focused on specific cognitive content. One of the differences between cognitive and metacognitive models of depression is the emphasis given to different types of beliefs. The present findings are consistent with metacognitive therapy of depression that
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
32
The Journal of General Psychology
focuses on the assessment and modification of positive and negative beliefs about rumination (Wells, Fisher, Myers, Wheatley, Patel, & Brewin, 2009, 2012). Some limitations of this study should be acknowledged. First, the use of different measurement devices for dysfunctional attitudes and depressive symptomatology may lead to different results. Thus, the relative contributions of cognitions and metacognitions should also be tested using different metacognitive, cognitive and depressive measurement tools. Second, this study was not conducted with a clinical sample, and it may be the case that relationships between depression and the predictor variables may be different in clinical samples. For instance, negative schemas may be better formed and exert a greater influence in patients. In addition, the low mean depression score of the present sample indicates that the range of scores was probably restricted. Although this is not surprising given the student profile of the sample, this would weaken any relationships with other variables. Thus, the study should be repeated in both clinical and non-clinical samples in order to produce a much wider range of scores, to compare the differences, and to evaluate the clinical generalizability of these relationships. AUTHOR NOTES Adviye Esin Yılmaz is assistant professor of Clinical Psychology at the Psychology Department of Dokuz Eyl¨ul University, Turkey. Her research focuses on cognitive theory and therapy of anxiety and mood disorders and scale adaptation. ¨ Genc¸o¨ z is professor of Clinical Psychology at the Psychology Department Tulin of Middle East Technical University, Turkey. She has published several articles in international journals about antecedents, concomitants, and consequences of psychological problems; personality traits and Big Five model of Personality; information processing approaches in clinical psychology; and scale adaptation. Adrian Wells is professor of Clinical and Experimental Psychopathology at University of Manchester (UK). He has published extensively in the areas of cognitive theory and therapy of anxiety and mood disorders. He is the originator of Metacognitive Therapy and his treatments have been included in NICE guidelines. FUNDING This research has been supported by the Scientific and Technological Research Council of Turkey (TUBITAK), in the framework of the 2214-Abroad Research Grant Program.
REFERENCES Beck, A. T. (1976). Cognitive therapy and emotional disorders. New York, NY: International Universities Press.
Downloaded by [Heriot-Watt University] at 03:08 05 January 2015
Yılmaz et al.
33
Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety: Psychometric properties. Journal of Consulting and Clinical Psychology, 56, 893–897. Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy for depression. New York, NY: Guilford. Beck, A. T., Steer, R. A., & Garbin, M. A. (1988). Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review, 8, 77–100. Butler, A. C., Chapman, J. E., Forman, E. M., & Beck, A. T. (2006). The empirical status of cognitive-behavioral therapy: A review of meta-analyses. Clinical Psychology Review, 26, 17–31. Creamer, M., Foran, J., & Bell, R. (1995). The Beck Anxiety Inventory in a non-clinical sample. Behaviour Research and Therapy, 33, 477–485. de Beurs, E., Wilson, K. A., Chambless, D. L., Goldstein, A. J., & Feske, U. (1997). Convergent and divergent validity of the Beck Anxiety Inventory for patients with panic disorder and agoraphobia. Depression and Anxiety, 6, 140–146. Flavell, J. N. (1979). Metacognition and cognitive monitoring: A new area of cognitivedevelopmental inquiry. American Psychologist, 34, 906–911. Osman, A., Kopper, B. A., Barrios, F. X., Osman, J. R., & Wade, T. (1997). The Beck Anxiety Inventory: Reexamination of factor structure and psychometric properties. Journal of Clinical Psychology, 53, 7–14. Papageorgiou, C., & Wells, A. (1999). Process and meta-cognitive dimensions of depressive and anxious thoughts and relationships with emotional intensity. Clinical Psychology and Psychotherapy, 6, 156–162. Papageorgiou, C., & Wells, A. (2001). Positive beliefs about depressive rumination: Development and preliminary validation of a self-report scale. Behavior Therapy, 32, 13–26. Papageorgiou, C., & Wells, A. (2003). An empirical test of a clinical metacognitive model of rumination and depression. Cognitive Therapy and Research, 27, 261–273. Papageorgiou, C., Wells, A., & Meina, L. J. (2008) Development and preliminary validation of the Negative Beliefs about Rumination Scale. Manuscript in preparation. Power, M. J., Katz, R., McGuffin, P., Duggan, C. F., Lam, D., & Beck, A. T. (1994). The Dysfunctional Attitude Scale (DAS): A comparison of forms A and B and proposals for a new subscaled version. Journal of Research in Personality, 28, 263–276. Weissman, A. N. (1979). Assessing depressogenic attitudes: A validation study. Unpublished Thesis, University of Pennsylvania, Philadelphia, PA. Weissman, A. N. & Beck, A. T. (1978). Development and validation of the Dysfunctional Attitude Scale: A preliminary investigation. Paper presented at the meeting of the American Educational Research Association, Toronto, Ontario. Wells, A. (2009). Metacognitive therapy for anxiety and depression. New York, NY: Guilford. Wells, A. & Matthews, G. (1994). Attention and emotion: A clinical perspective. Hove UK: Erlbaum. Wells, A., Fisher, P., Myers, S., Wheatley, J., Patel, T., & Brewin, C. R. (2009). Metacognitive therapy in recurrent and persistent depression: A multiple-baseline study of a new treatment. Cognitive Therapy and Research, 33, 291–300. Wells, A., Fisher, P., Myers, S., Wheatley, J., Patel, T., & Brewin, C. R. (2012). Metacognitive therapy in treatment resistant depression: A platform trial. Behaviour Research and Therapy, 50, 367–373.
Original manuscript received February 13, 2014 Final version accepted September 8, 2014
