• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

A Test of Integration of the Activation Hypothesis and the Diathesis- Stress Component of the Hopelessness Theory of Depression

Posted on: Sunday, 11 July 2004, 06:00 CDT

Objectives. This prospective study tested the integration of the diathesis-stress component of the hopelessness theory of depression (Abramson, Metalsky, & Alloy, 1989) and Persons and Miranda's (1992) activation hypothesis (i.e. depressogenic inferential styles are typically latent cognitive processes that must be primed in order to be accurately assessed).

Design. In order to test the diathesis-stress component of the hopelessness theory, we used a short-term longitudinal design. In order to test the activation hypothesis, inferential styles were assessed both before and after a negative cognitive priming questionnaire.

Methods. A group of 165 university students completed measures of inferential styles about the self, consequences, and causes before and after completing a negative cognitive priming questionnaire (Time 1). Participants also completed measures of depressive symptoms prior to completing the cognitive priming questionnaire and 5 weeks later (Time 2). Finally, negative events occurring between Time 1 and Time 2 were assessed.

Results. Contrary to the diathesis-stress component of the hopelessness theory, none of the unprimed inferential styles interacted with negative events to predict increases in depressive symptoms. In line with the integration of the hopelessness theory and the activation hypothesis, however, each of the primed inferential styles interacted with negative events to predict increases in depressive symptoms even after controlling for the proportion of variance in depressive symptoms accounted for by the unprimed inferential style stress interactions.

Conclusion. Individuals with depressogenic inferential styles are likely to show increases in depressive symptoms following the occurrence of negative events. At the same time, these depressogenic inferential styles are typically latent cognitive processes that must be primed in order to be accurately assessed.

The hopelessness theory of depression (Abramson et al., 1989) is a cognitive diathesis-stress theory that proposes that certain inferential styles - habitual ways of perceiving the causes and meanings of events - predispose individuals to hopelessness depression when they encounter negative events. Three different depressogenic inferential styles are delineated by the theory: inferential styles about the self, consequences, and causes. The inferential style about the self refers to the tendency to draw negative inferences about the self (i.e. one's "worth, desirability, abilities, etc.) given that a negative event has occurred. The inferential style about consequences refers to the tendency to catastrophize the consequences of negative events and to view these consequences as important, unlikely to change, and as affecting many areas of one's life. The inferential style about causes refers to the tendency to attribute negative events to global and stable causes. Individuals with depressogenic inferential styles are more likely than individuals without such styles to make depressogenic inferences for negative events they encounter. Making depressogenic inferences increases the likelihood that hopelessness will develop. Once hopelessness develops, hopelessness depression is inevitable, since hopelessness is viewed by the theory as a sufficient cause of depression.

Since its proposal, the hopelessness theory has generated a vast amount of research examining its etiological component. Results have been mixed. Although some studies have provided full support for the diathesis-stress component of the theory (e.g. Alloy & Clements, 1998; Alloy, Just, & Panzerella, 1997; Metalsky & Joiner, 1992, 1997), other studies have provided only partial support (e.g. Abela, 2002; Abela & Seligman, 2000; Metalsky, Halberstadt, & Abramson, 1987; Metalsky, Joiner, Hardin, & Abramson, 1993) or no support (e.g. Follette & Jacobson, 1987; Hunsley, 1989; Ralph & Mineka, 1998; Swendsen, 1997). Differences between studies have been attributed to many factors, including inappropriate research designs, inadequate conceptualizations of attributional style, differences in the timings of assessments of depressive symptoms, insufficient statistical power, and lack of assessments of the congruency between the domain of individuals' inferential styles (interpersonal vs. achievement) and the type of negative events encountered (Abramson, Alloy, & Metalsky, 1995; Robins & Hayes, 1995).

In addition, Persons and Miranda (1992) have postulated that inconsistency in results is likely to be the result of inadequate assessments of inferential styles. According to Persons and Miranda's activation hypothesis, cognitive vulnerability factors, such as depressogenic inferential styles, are typically latent and must be activated in order to be accurately assessed. Therefore, studies that do not activate or prime inferential styles before assessing them are likely to inadequately assess the true propensity of participants' depressogenic thinking. These studies, consequently, are less likely to yield results that are supportive of hopelessness theory.

Persons and Miranda's (1992) activation hypothesis is in line with hypotheses made by many other cognitive theorists. For example, Beck (1967, 1983) has suggested that depressogenic cognitive schemata are typically latent and must be activated by relevant Stressors in order to be detected. Similarly, Riskind and Rholes (1984) have suggested that although depressogenic thought patterns are enduring characteristics of depression-prone individuals, they may not always be activated or cognitively accessible. When primed and accessible, however, such patterns are hypothesized to produce the characteristic cognitive distortions and symptoms of depression. Last, according to Teasdale's (1983) differential activation hypothesis, individual differences exist in the type of cognitive structures and processes that become activated when an individual experiences mild depressed mood. The more negative the cognitive structures and processes are that are activated by initial depressed mood, the more likely the individual will develop severe and enduring symptoms of depression. Although each of the above mentioned theorists uses different terminology, each suggests that our ability to predict the occurrence of depressive symptoms using measures of negative thinking patterns depends on whether or not these patterns have been cognitively primed or activated before they are assessed.

Several studies conducted within the framework of Beck's (1967, 1983) cognitive theory of depression support this line of reasoning by showing that dysfunctional thinking appears in vulnerable individuals (as defined by past history of depressive disorder) but not in non-vulnerable individuals when activating stimuli (mood or life events) have primed otherwise latent cognitions. For example, Miranda and Persons (1988) found that participants with a history of depression endorsed more dysfunctional attitudes than did participants without such a history. This effect, however, occurred only for participants who were in a current negative mood state. In a replication of this study, Roberts and Kassel (1996) found that dysfunctional attitudes and negative automatic thoughts correlated with negative affect more strongly in remitted dysphorics than in never-dysphoric individuals. Miranda, Persons, and Byers (1990) found that dysfunctional beliefs varied with spontaneous diurnal mood fluctuations in depressed psychiatric patients but not in participants who had never been depressed. Miranda, Gross, Persons, and Hahn (1998) found that increases in depressed mood following a film-based negative mood induction led to increased endorsement of dysfunctional attitudes in formerly depressed participants but not in participants with no prior history of depression. Finally, Miranda (1992) found that nondepressed participants showed elevated levels of dysfunctional attitudes if they had both (1) previously experienced a depressive episode and (2) recently experienced a negative event.

Studies using priming procedures to assess depressogenic schematic processing have obtained similar results. For example, Ingram, Bernet, and McLaughlin (1994) found that non-depressed college students with a history of depression were excessively distracted by negative words in a dichotic listening task if, and only if, the task was preceded by a negative mood induction procedure. In contrast, participants without a history of depression were not distracted by such words regardless of whether or not the task was preceded by a negative mood induction. In a related series of studies, Hedlund and Rude (1995) and Teasdale and Dent (1987) had participants read positive and negative trait adjectives and indicate which ones described themselves. A few moments later, they were asked to recall the adjectives they had endorsed. Individuals with and without a history of depression did not differ on this incidental recall task if the task was not preceded by a negative mood induction procedure (Teasdale & Dent, 1987) or induced self- focus (Hedlund & Rude, 1995). The two groups did differ, however, on the task if it was preceded by such procedures. Individuals with a history of depression recalled more depressive self-refere\nt traits than those without such a history. Williams (1988, cited in Ingram, Miranda, & Segal, 1998) had participants recall positive and negative self-referent adjectives under two conditions: neutral mood and induced sad mood. Differential recall (positive vs. negative adjectives) under neutral mood did not predict depression 1 year later. Differential recall under induced sad mood, however, did predict depression at the 1-year follow-up. Participants who recalled more negative than positive self-descriptive adjectives under these conditions were more likely to become depressed at follow-up. Finally, Segal, Gemar, and Williams (1999) found that following cognitive-behavioural or psychopharmacological treatment for depression, recovered depressives who showed a significant increase in dysfunctional attitudes following a negative mood induction were more likely to relapse by follow-up than recovered depressives who did not show such an increase.

Although Persons and Miranda's (1992) activation hypothesis has received a considerable degree of support from studies conducted within the framework of Beck's (1967, 1983) cognitive theory of depression, this hypothesis has not yet been examined within the context of the hopelessness theory. At the same time, discovering that depressogenic inferential styles are typically latent cognitive processes that must be primed in order to be accurately assessed would have important theoretical and practical implications. Theoretically, it would suggest that the hopelessness theory be modified in order to incorporate Persons and Miranda's activation hypothesis into the model. Practically, it would provide researchers with an explanation for past inconsistencies in the literature examining the diathesis-stress component of the hopelessness theory. In other words, given that past studies have not explicitly attempted to activate depressogenic inferential styles before assessing them, it is possible that several studies failed to provide support for the diathesis-stress component of the theory because depressogenic inferential styles were not currently activated in a significant proportion of participants when they were assessed.

The goal of the current study was to test the integration of the diathesis-stress component of the hopelessness theory of depression and Persons and Miranda's (1992) activation hypothesis. Participants completed measures of inferential styles about the self, consequences, and causes before and after completing a cognitive priming questionnaire (Time 1) in which they were asked to recall a recent negative event in vivid detail. Participants also completed measures of depressive symptoms prior to completing the cognitive priming questionnaire and 5 weeks later (Time 2). Finally, negative events occurring between Time 1 and Time 2 were assessed. In order to provide a stringent test of the integration of the diathesis- stress component of the hopelessness theory and Persons and Miranda's (1992) activation hypothesis, we examined the incremental validity of the activation procedure. In other words, we hypothesized that participants' primed inferential styles about the self, consequences and causes would interact with the occurrence of negative events to predict increases in depressive symptoms between Time 1 and Time 2 even after controlling for variance in depression scores accounted for by the unprimed inferential style stress interactions.

Method

Participants

A sample of 165 undergraduates at the University of Pennsylvania (57 men and 108 women) served as participants in this study. They were recruited from the Psychology Department's subject pool and received extra credit in their introductory psychology course for participating. Two participants (one man and one woman) did not fill out all of the questionnaires at Time 1. Subsequently, their data were not used in any of the analyses.

Measures

Extended Attributional Style Questionnaire (EASQ-1; Metalsky et al., 1987)

The EASQ-1 consists of a total of 12 hypothetical negative events. Six of the events are of an interpersonal nature (i.e. an important romantic relationship you are involved in breaks up because the other person no longer wants a relationship with you). The other six events are of an achievement nature (i.e. you take an examination and receive a low grade). As with the original Attributional Style Questionnaire (Peterson et al., 1982), participants are asked to write down the one major cause of each event. They are then asked to rate the cause on a 1-7 scale for each of the following dimensions: internality, globality and stability. The higher the scores are on these dimensions, the more internal, global and stable the attributional style. Consistent with the hopelessness theory (i.e. Abramson et al., 1989), we used the generality subscale (average of globality and stability; ASQ-GEN) to test our predictions. Scores were averaged across the 12 negative life events, yielding one score ranging from 1 to 7, with higher scores corresponding to more depressogenic attributional styles.

Expanded Attributional Style Questionnaire (EASQ-2; Peterson & Villanova, 1988)

The EASQ-2 consists of a total of 24 hypothetical negative events. It is identical in format to the EASQ-1. Participants are asked to write down the one major cause of each event. They are then asked to rate the cause on a 1-7 scale for internality, globality and stability. The higher the scores are on these dimensions, the more internal, global and stable the attributional style. In order to make the EASQ-2 similar to the EASQ-1 in terms of length and content, we used a modified version that consisted of 12 rather than 24 events: six interpersonal and six achievement. As with the EASQ- 1, we used the generality subscale (average of globality and stability; ASQ-GEN) to test our predictions.

Cognitive Style Questionnaire (CSQ-1 ; Abramson & Metalsky, 1986)

The CSQ-1 assesses cognitive styles about consequences and the self. To assess the cognitive style about consequences, participants are asked the following question for each of the 12 negative life events in the EASQ-1: 'How likely is it that the [negative event] will lead to other negative things happening to you?' They are asked to respond on a 1-7 scale, with 1 meaning 'It is not at all likely that the [negative event] will lead to other negative things happening to me' and 7 meaning 'It is extremely likely that the [negative event] will lead to other negative things happening to me'. To assess the cognitive style about self, participants are asked the following question for each of the 12 negative life events in the EASQ-1: 'To what degree does the [negative event] mean that you are flawed in some way?' Participants are asked to respond on a 1-7 scale, with 1 meaning 'The [negative event] definitely does not mean I am flawed in some way' and 7 meaning 'The [negative event] definitely does mean I am flawed in some way'. Scores were averaged across the 12 negative life events yielding subscales for consequences (CSQ-CONS) and the self (CSQ-SELF). Scores on each subscale can range from 1 to 7, with higher scores corresponding to more depressogenic inferential styles.

Cognitive Style Questionnaire (CSQ-2; Abramson & Metalsky, 1986)

The CSQ-2 is identical in every respect to the CSQ-1 except that the questions are asked in reference to the 12 hypothetical negative events on the EASQ-2 rather than to those on the EASQ-1. Scores were averaged across the 12 negative life events yielding subscales for consequences (CSQ-CONS) and the self (CSQ-SELF). Scores on each subscale can range from 1 to 7, with higher scores corresponding to more depressogenic inferential styles.

Cognitive Priming Questionnaire (CPQ; Abela, 1995)

The CPQ is a three-part questionnaire that was designed to use the recall of a past upsetting episode to activate latent cognitive styles. In Part 1, participants are asked to close their eyes and to recall as vividly as possible the time when they were most upset during the last year. They are then asked to write as detailed a description as possible of this time, including a description of the precipitating event(s) and their reactions (i.e. thoughts and feelings) to it. In Part 2, participants are presented with a list of 41 depressogenic cognitions (i.e. Tm worthless'). The list was derived from cognitions that appear on the Automatic Thoughts Questionnaire (Hollon & Kendall, 1980) and the depression subscale of the Cognition Checklist (Beck, Brown, Steer, Eidelson, & Riskind, 1987). These thoughts have been shown to discriminate depressed patients from non-depressed patients and control groups (Hollon, Kendall, & Lumry, 1986; Steer, Beck, Clark, & Beck, 1994). Participants are asked to read each of these cognitions out loud. They are then asked to concentrate on the meaning of each statement and to try to recall if they had a similar thought during the period they previously described in Part 1. Last, they are asked to place an 'x' next to each statement that accurately describes their thoughts during this period. In Part 3, participants are presented with a list of 45 adjectives that describe emotions experienced during a depressed state. Participants are asked to focus on their feelings during the previously described period. They are then asked to place an 'x' next to each of the adjectives that accurately describes their feelings during this time.

Beck Depression Inventory (BDI: Beck, Ward, Mendelson, Mock, & Erbaugh 1961)

The BDI is a 21-item self-report questionnaire that assesses the severity of depressive symptoms present within the past 2 weeks. Each item is rated on a 0-3 scale, with higher numbers corresponding to more severe symptoms. Scores range from 0 to 63.

Negative Life Events Questionnaire (NLEQ: Saxe & Abramson, 1987)

The NLEQ is a 134-item self-report questionnaire that assesses the occurrence of negative eventsand stressful situations. The items on the NLEQ are a comprehensive list of negative life events and stressful situations experienced by college students. For each event or situation, participants are asked to rate whether or not they experienced it within the past 5 weeks. Participants are next asked to indicate the number of times it occurred (for events) or the number of days it persisted (for situations). Scores for each event can range from 0 (i.e. did not occur) to 5 (i.e. occurred five or more times). Scores for each situation can range from O (i.e. did not occur) to 5 (i.e. persisted for 5 days or more). The total score is the sum of all of the items. Scores range from 0 to 670, with higher numbers corresponding to the occurrence of a greater number of negative events.

Procedure

Participants were told that they were participating in a study about students' perceptions of stressful life events. At Time 1, participants met with a researcher and filled out a consent form, a subject identification form, and each of the following questionnaires: BDI, EASQ-1, CSQ-1, CPQ, EASQ-2 and CSQ-2. The order of the cognitive style questionnaires (CSQ-1 and CSQ-2) and the attributional style questionnaires (EASQ-1 and EASQ-2) was counterbalanced. The order of the questionnaires in subject packets alternated so that every other subject received the same order.

Five weeks later at Time 2, participants returned to the laboratory and met with a researcher. At this time, they filled out the BDI to assess increases in depressive symptoms and the NLEQ to assess experience of negative life events. After filling out these questionnaires, participants were debriefed. In addition, any questions about the study were answered.

Results

Descriptive data

The mean score on the NLEQ was 60.20 with a standard deviation of 32.59. Although BDI scores were higher at Time 1 (M = 7.38, SD = 6.78) than at Time 2 (M = 6.58, SD = 6.77), this difference was not significant, (t(162) = 1.43, ns). In addition, this difference is not contrary to our hypotheses, as we do not expect every participant's level of depressive symptoms to increase over the 5- week period. According to the hopelessness theory, only participants who both possess depressogenic inferential styles and experience negative life events should show increases in depressive symptoms between Time 1 and Time 2. Furthermore, other studies have demonstrated that averaged BDI scores tend to decrease after multiple administrations (Metalsky et al., 1993; Sharpe & Gilbert, 1998).

Participants' primed inferential styles about the self (P-CSQ- SELF; M = 3.45, SD = 1.37) were slightly more depressogenic than their unprimed (U-CSQ-SELF) scores (M = 3.40, SD = 1.25), although not significantly so (t(162) = 0.93, ns). Participants' primed inferential styles about consequences (P-CSQ-CONS; M = 3.59, SD = 1.21) were significantly more depressogenic (t(162) = 2.82, p < .01) than their unprimed (U-CSQ-CONS) scores (M = 3.43, SD = 1.07). Similarly, participants' primed inferential styles about causes (P- ASQ-GEN; M = 3.94, SD = 0.92) were significantly more depressogenic (t(162) = 2.50, p < .05) than their unprimed (U-ASQ-GEN) scores (M = 3.82, SD = 0.82). According to the priming hypothesis, not all participants' scores are expected to increase following priming. Only those participants who already possess depressogenic inferential styles are expected to show increases in their EASQ and CPQ scores.

Overview of statistical analysis

In order to test the diathesis-stress component of the hopelessness theory, we performed hierarchical multiple regressions (Cohen & Cohen, 1983). In all cases, the dependent variable was the Time 2 BDI score. First, the Time 1 BDI score (the covariate) was entered into the equation. By entering this score first, we are controlling for Time 1 depressive symptoms by essentially using our independent variables to predict residual change in BDI scores from Time 1 to Time 2. Second, the stress variable (NLEQ) was entered into the equation. Third, the unprimed inferential style score were entered. Fourth, the interaction between unprimed inferential style and stress was entered. This step provided a test of the diathesis- stress portion of the theory using unprimed assessments of each inferential style. Fifth, we entered the primed inferential style score. Last, and most important, we entered the interaction between primed inferential style and stress. The final two steps provided a stringent test of our cognitive priming hypothesis.

Results

Results pertaining to the inferential style about the self are presented in Table 1. The occurrence of negative life events between Time 1 and Time 2 (NLEQ) was a significant predictor of residual change in BDI scores from Time 1 to Time 2, F(1, 160) = 18.17, p < .001. In contrast, U-CSQ-SELF by itself, was not a significant predictor of residual change scores, F(1, 159) = .53, ns. Contrary to the hopelessness theory, the U-CSQ-SELF NLEQ interaction was not a significant predictor of residual change scores, F(1, 158) = 0.13, ns. In addition, P-CSQ-SELF, by itself, was not a significant predictor of residual change scores, F(157, 1) = 1.87, ns. Last, and most crucial to our cognitive priming hypothesis, the P-CSQ-SELF NLEQ interaction uniquely accounted for a significant 4.8% of the variance in residual change in BDI scores from Time 1 to Time 2, F(156, 1) = 7.97, p < .01. The sign of the partial correlation (pr = .22) reveals that the interaction is in the predicted direction - a depressogenic inferential style about the self coupled with high levels of stress predicted increases in depressive symptoms.

Results pertaining to the inferential styles about consequences and causes were similar and are presented in Table 2 and 3, respectively. As with the analyses concerning an inferential style about the self, neither unprimed inferential styles about consequences nor causes (U-CSQ-CONS and U-ASQ-GEN) nor their interactions with stress (U-CSQ-CONS NLEQ and U-ASQ-GEN NLEQ) were significant predictors of residual change in BDI scores. Similarly, neither P-CSQ-CONS nor P-ASQ-GEN were significant predictors of changes in BDI scores. However, both the interaction of primed inferential styles about consequences with stress (P-CSQ-CONS NLEQ) and the interaction of primed inferential styles about causes with stress (P-ASQ-GEN NLEQ) uniquely accoxmted for a significant 6.8% and 7.6% of the variance, respectively, in residual change in BDI scores from Time 1 to Time 2. The signs of the partial correlations (pr = .26 and pr = .28, respectively) reveal that the interactions are in the predicted direction - depressogenic inferential styles about consequences and causes predicted increases in depressive symptoms following the occurrence of high levels of stress.

Table 1. Depressogenic inferential styles about the self: Prediction of residual change in BDI scores from Time 1 to Time 2

Table 2. Depressogenic inferential styles about consequences: Prediction of residual change in BDI scores from Time 1 to Time 2

Table 3. Depressogenic inferential styles about causes: Prediction of residual change in BDI scores from Time 1 to Time 2

In order to depict the interactions, we regressed Time 2 BDI scores on Time 1 BDI scores, primed inferential style NLEQ interactions. We then followed the procedure recommended by Cohen and Cohen (1983) to compute residual change scores by inserting specific values for predictor variables (i.e. 1.5 SD above and below the mean) into the regression equations coinciding with analyses reported in Table 1, 2 and 3. Figure 1 shows the results of these calculations. As can be seen in the top graph, students with low P- CSQ-SELF scores showed little change in BDI scores from Time 1 to Time 2 regardless of whether they experienced low or high levels of stress. In contrast, students with high P-CSQ-SELF scores showed increase in BDI scores following high levels of stress, but not following low levels of stress. A similar pattern of results was obtained for the P-CSQ-CONS NLEQ and P-ASQ-GEN NLEQ interactions (see the middle and bottom graphs of Fig. 1 respectively).

Because the three inferential styles were highly correlated with each other, we again examined the effect of each inferential style ? stress interaction while controlling for the other two. Time 2 BDI scores were entered first. NLEQ scores were entered second. P-CSQ- SELF, P-CSQ-CONS, and P-ASQ-GEN were entered third. Last, we entered each of the three inferential style interaction terms. In this analysis, the P-CSQ-SELF NLEQ interaction did not have a unique effect (pr = - .10 ns), nor did the P-CSQ-CONS NLEQ interaction (pr = .13, ns) or the P-ASQ-GEN NLEQ interaction (pr = .06, ns).

Discussion

When the results were examined exclusively within the framework of the hopelessness theory of depression, the results failed to support the theory. In other words, contrary to the diathesis- stress component of the theory, unprimed depressogenic inferential styles about the self, consequences and causes did not interact with the occurrence of negative events to predict increases in depressive symptoms. At the same time, when the results were examined within the framework of the integration of the hopelessness theory and Persons and Miranda's (1992) activation hypothesis, the results fully supported our hypotheses. In other words, primed depressogenic inferential styles about the self, consequences and causes interacted with the occurrence of negative events to predict increases in depressive symptoms. It is important to note that these styles predicted such increases even after controlling for the proportion of variance in depressive symptoms accounted for by the unprimed inferential style stress interactions.

In line with the reasoning of Persons and Miranda (1992) and Riskind and Rholes (1984), the results of the current study sug\gest that depressogenic inferential styles about the self, consequences and causes are enduring characteristics of individuals vulnerable to hopelessness depression. At the same time, depressogenic inferential styles may not always be more dominant than, or as cognitively accessible as, other neutral or even positively valenced styles of thinking. Further, to the extent that they exist, depressogenic inferential styles are likely to be cognitively primed by stressful situations. The capacity of a situation to elicit depressogenic inferential styles through priming may be a function of the characteristic level of accessibility of such negativistic thinking patterns in the individual. Last, when primed and thus accessible, depressogenic inferential styles are likely to produce the depressogenic inferences that are hypothesized to lead to the symptoms of hopelessness depression. This framework therefore suggests that our capacity to predict the likelihood of occurrences of symptoms of hopelessness depression from measures of depressogenic inferential styles depends on whether or not these styles have been cognitively primed.

The results of the current study suggest that the hopelessness theory of depression should incorporate Persons and Miranda's (1992) activation hypothesis into the theory so as to provide a more adequate operationalization of depressogenic inferential styles. In addition, the current results provide one possible explanation for inconsistencies in results of past studies examining the diathesis- stress component of the hopelessness theory. Given that no past studies have utilized priming procedures when assessing inferential styles, we cannot be certain that they assessed the true propensity for depressogenic thinking individuals vulnerable to hopelessness depression. Future researchers should incorporate such priming techniques into their assessment procedures in order to increase the accuracy of their assessments of these inferential styles.

Figure I. Regression equations plotting residual changes in Beck Depression Inventory (BDI) scores from Time 1 to Time 2 as a function of negative life events and the primed inferential style about the self (top panel), primed inferential style about consequences (middle panel), and primed inferential style about causes (bottom panel).

At the same time, it is important to note that other studies using designs similar to the current study have obtained support for the diathesis-stress component of the hopelessness theory without utilizing priming procedures when assessing inferential styles (e.g. Alloy & Clements, 1998; Metalsky & Joiner, 1992, 1997). Such findings cannot be used as support against the activation hypothesis, as participants' current levels of either environmental stress or depressed mood may have served as natural primes. At the same time, given that these studies have provided support for the hopelessness theory without using activation procedures when assessing inferential styles, one might argue that priming is not truly necessary. We would argue, however, that taking such a stance is likely to lead to continued inconsistency between studies in terms of degree of support obtained for the hopelessness theory. According to the activation hypothesis, the use of priming procedures in the assessment of inferential styles is likely to decrease measurement error, as a greater proportion of participants' true propensities towards depressogenic thinking will be accurately assessed. In any given study, a certain proportion of participants are likely to complete Time 1 assessments of inferential style with their inferential styles already primed as a result of either naturally occurring Stressors or elevated levels of depressed mood. The proportion of participants who are naturally primed is likely to vary from study to study depending on factors such as sample composition, the manner in which questions were administered, and the timing of when questionnaires were administered. For support to be obtained for the hopelessness theory in the absence of priming procedures, the proportion of participants naturally primed at Time 1 who experience high levels of stress during the follow-up interval and who show increases in depressive symptoms must be significantly greater than the proportion of participants not naturally primed at Time 1 who experience high levels of stress during the follow-up interval and who show increases in depressive symptoms. As some studies will meet this requirement, some studies will provide support for the diathesis-stress component without using activation procedures when assessing inferential styles. At the same time, as other studies will not meet this requirement, failure to incorporate activation procedures into their assessment protocol will lead to a failure to provide support for the diathesis-stress component of the hopelessness theory. Thus, if the activation hypothesis is accurate, as the results of the current study suggest, failure to utilize activation procedures when assessing inferential styles is likely to lead to increased measurement error in the assessment of inferential styles and thus is also likely to lead to continued inconsistency in the findings of diathesis-stress studies.

Factors other than the activation hypothesis are also likely to account for inconsistencies among studies. For example, the symptom component of the hopelessness theory states that hopelessness depression is characterized by a unique symptoms profile. Consequently, depressogenic inferential styles should interact with negative events to predict increases in hopelessness depression symptoms (e.g. sadness, motivational symptoms, suicidal ideation) but not non-hopelessness depression symptoms (e.g. anhedonia, appetite disturbance, hypersomnia). Several studies have provided support for this hypothesis (e.g. Abela & Sarin, 2002; Alloy & Clements, 1998; Alloy et al., 1997; Hankin, Abramson, & Siler, 2001; Metalsky & Joiner, 1997). Given that support has accumulated for the symptom component of the hopelessness theory, measurement error is also likely to be introduced into studies that utilize measures of depression, such as the Beck Depression Inventory (BDI), that were not specifically designed to assess hopelessness depression symptoms. Such error is likely to result from such measures both failing to assess all of the hypothesized symptoms of hopelessness depression and assessing many symptoms not hypothesized to be part of hopelessness depression. When we compare the results of our analyses using unprimed assessments of inferential styles to the results from studies that have utilized a similar design (e.g. Alloy & Clements, 1998; Metalsky & Joiner, 1992, 1997), all studies are likely to include some degree of measurement error in the assessment of inferential styles, as inferential styles were not assessed utilizing activation procedures. At the same time, given that the majority of these studies, unlike the current study, assessed hopelessness depression symptoms specifically (e.g. Alloy & Clements, 1998; Metalsky & Joiner, 1997), these studies are likely to include a smaller degree of measurement error in the assessment of depressive symptoms. A smaller degree of measurement error in the assessment of hopelessness depression symptoms may have increased the likelihood of obtaining findings that support the hopelessness theory. Interestingly, Metalsky and Joiner (1997) examined whether unprimed depressogenic attributional styles interacted with negative events to predict increases in both Hopelessness Depression Symptom Questionnaire (HDSQ) scores and BDI scores. In line with the results of the current study, an unprimed depressogenic attributional style did not interact with negative events to predict increases in BDI scores. At the same time, an unprimed depressogenic attributional style did interact with negative events to predict increases in HDSQ scores. Given that measurement error can be introdviced in the assessment of both depressogenic inferential styles and depressive symptoms, future researchers should address both the activation hypothesis and the symptom component to decrease measurement error in assessing both of these variables. Doing so is likely to lead to even greater consistency in findings across studies.

It is also important to note that although inferential styles about the self, consequences and causes each interacted with high levels of stress to predict increases in depressive symptoms, none of them accounted for unique variance in depressive symptoms. In other words, the results of the current study are best viewed as one effect (i.e. depressogenic inferential styles interacted with negative events to predict increases in depressive symptoms) rather than as three separate effects (i.e. depressogenic inferential styles about the self, consequences and causes each interacted with negative events to predict increases in depressive symptoms). Similar results have been obtained in other studies with adult populations (e.g. Abela, 2002; Abela & Seligman, 2000; Metalsky & Joiner, 1992). Clearly inferential styles about the self, consequences and causes are logically distinguishable. Such styles, however, do not appear to be empirically distinguishable in these data. One possible interpretation of these findings is that all three inferential styles are schema-driven cognitive processes. Similar depressogenic schemata may underlie each of the three inferential styles. At the same time, inferential styles about the self, consequences and causes have been found to be empirically distinguishable in child and preadolescent samples (Abela, 2001; Abela & Sarin, 2002). Future research should therefore continue to explore the empirical distinguishability of these inferential styles in adult populations.

Finally, an examination of Fig. 1 reveals that the magnitude of change in BDI \scores predicted by the interaction of primed depressognic inferential styles with high levels of stress was relatively small. One could possibly argue that although such an increase is statistically significant, it is not necessarily clinically meaningful. It is important to note that the hopelessness theory (Abramson et al., 1989) posits that hopelessness depression exists on a continuum of severity ranging from subclinical depressive mood reactions to enduring clinically significant depressive symptoms. Severity of hopelessness depression is hypothesized to vary as a function of (1) the severity of depressogenic inferential styles, (2) the severity of negative events, and (3) the generality of hopelessness. Therefore, according to the theory, depressogenic inferential styles can serve as vulnerability factors to either subclinical depressive mood reactions or clinically significant levels of depressive symptoms, depending on the severity of Stressors encountered and the generality of hopelessness that develops. Given that the theory states that a less severe analogue to clinical depression exists when Stressors are not extreme and hopelessness is not generalized, researchers have examined the hopelessness theory from multiple perspectives ranging from predicting mild depressive mood reactions following the occurrence of stressful event (e.g. Abela, 2002; Abela & Seligman, 2000; Metalsky et al., 1987) to predicting the development of clinically significant symptoms in high-risk populations (e.g. Alloy et al., 2000). As the current study utilized a sample of university students in which the base rates of both depressogenic inferential styles and severe negative events were relatively low, it is not surprising that change in BDI scores was relatively small. At the same time, future research should examine the integration of the hopelessness theory and Persons and Miranda's (1992) activation hypothesis in a sample where a significant number of participants possess depressogenic inferential styles and are likely to experience severe negative events in order to examine whether the current findings extend to the onset of clinically significant depressive episodes.

The present study advances research on the hopelessness theory for several reasons. First, until recently, most researchers have utilized cross-sectional designs to assess the etiological component of the hopelessness theory. Given that the hopelessness theory is a diathesis-stress theory, the use of such designs is inadequate for examining the theory's validity (for detailed critiques of such methods see Abramson et al, 1988; Alloy, Hartlage, & Abramson, 1988). Therefore, by examining the interaction of each of the cognitive diatheses with negative life events in predicting depression, this study is among the minority of studies to appropriately examine the validity of the diathesisstress component of the theory. second, of the studies that have used a prospective design to examine the diathesis-stress component of the hopelessness theory in adult samples, only three (Abela, 2002; Abela & Seligman, 2000; Metalsky & Joiner, 1992) have examined the inferential styles about the self and consequences. The remaining studies have only examined the inferential style about causes. Third, of the studies that have used a prospective design to examine the diathesis-stress component of the hopelessness theory, this is the first to utilize priming procedures in the assessment of inferential styles. The results revealed that, as predicted, depressogenic inferential styles are typically latent in individuals vulnerable to hopelessness depression and need to be primed in order to be accurately assessed. Future research examining the etiological component of hopelessness theory should therefore incorporate priming procedures into their assessment protocol.

At the same time, several limitations of this study should be noted. First, the current study utilized self-report measures to assess depressive symptoms. Although the BDI possesses high degrees of reliability and validity, it is difficult to draw conclusions about clinical diagnoses of depressive disorders based on self- report measures. Future studies should utilize clinical interviews to see whether the current findings extend to the development of clinical depression. At the same time, when clinical diagnoses are used, diagnostic categories should not be restricted to Diagnostic and statistical manual of mental disorders (4th ed.), major depression (DSM-IV; American Psychiatric Association, 1994) as hopelessness depression is hypothesized to cut across various diagnostic categories (e.g. major depressive disorder, dysthymia, anxiety disorders, and borderline personality disorder). Second, the current study utilized the BDI to assess depressive symptoms. Given that there is a growing body of literature supporting the symptom component of the hopelessness theory (i.e. Abela & D'Alessandro, 2001; Abela & Sarin, 2002; Alloy & Clements, 1998; Alloy et al., 1997; Hankin et al., 2001; Metalsky & Joiner, 1997), future research should utilize instruments such as the Hopelessness Depression Symptom Questionnaire (Metalksy & Joiner, 1997) that allow for a more thorough assessment of hopelessness depression symptoms. Third, the current study utilized selfreport measures to assess stress. Although measures of stress that require participants solely to indicate the frequency and duration of negative events are less likely to be influenced by informant bias than those that ask participants to rate the subjective impact of each event, more sophisticated methods, such as the interviewing procedures used by Brown and Harris (1978) to assess contextual threat, may provide more precise measurement of stress levels. Fourth, the present study included only one follow-up assessment after a 5-week time lag. Therefore, participants who experienced high levels of stress early in the 5-week interval may have experienced temporary increases in depressive symptoms that were not detected at the follow-up assessment. Future research should make more frequent assessments of depressive symptoms or utilize assessment techniques that cover the entire interval in order to provide a more powerful test of the theory (i.e. for an example of such techniques, see Abela, 2001). Finally, the specific vulnerability hypothesis of the hopelessness theory states that some people tend to make depressogenic inferences only after negative interpersonal events, whereas others tend to do so only after negative achievement events. Only when there is a match between the content of an individual's inferential style and the domain of the negative event will depression occur. Future studies examining the diathesis-stress component of the hopelessness theory should incorporate a test of this hypothesis.

In conclusion, when results were examined exclusively within the framework of the hopelessness theory of depression, the results from the present study failed to provide support for the theory. In other words, contrary to the diathesis-stress component of the theory, unprimed depressogenic inferential styles did not interact with the occurrence of negative events to predict increases in depressive symptoms. At the same time, the results of the current study provided support for the integration of Persons and Miranda's (1992) activation hypothesis and the hopelessness theory. In other words, in line with the integrative theory, primed depressogenic inferential styles interacted with the occurrence of negative events to predict increases in depressive symptoms. Such results suggest that depressogenic inferential styles are typically latent and must be activated in order to be assessed accurately. Future research examining the diathesis-stress component of the hopelessness theory is therefore likely to benefit from incorporating Persons and Miranda's (1992) activation hypothesis into the hopelessness model. Such research is also likely to benefit from using more sophisticated assessments of stress and depressive symptoms in order to provide us with a deeper understanding of the cognitive processes and mechanisms underlying the aetiology of hopelessness depression.

Acknowledgements

Parts of this article were presented at the meeting of the Association for the Advancement of

Behavior Therapy, Washington DC, November 1998.

We thank Victoria Barbadoro, Elana Bornstein, Asaf Cohen, Joel Correia, Dana Farrell, Jennifer Friedman, Sarika Garg, Scott Glassman, Jessica Grisham, Lynne Havsy, Joel Herzig, Batya Kaufman, Bryan Kostelnik, Diana Leibert, Erina Lin, Andrea Nass, Suzanne Pineles, Jessica Schreck, Josephine Shih, Cara Weinstein and Amiel Weinstock for recruiting participants and administering and scoring subject questionnaires. We also thank Andrew MacLeod, David Paul, and two anonymous reviewers for comments on earlier versions of this paper.

References

Abela, J. R. Z. (1995). The Cognitive Priming Questionnaire. Unpublished questionnaire.

Abela, J. R. Z. (2001). The hopelessness theory of depression: A test of the diathesis-stress and causal mediation components in third and seventh grade children. Journal of Abnormal Child Psychology, 29, 241-254.

Abela, J. R. Z. (2002). Depressive mood reactions to failure in the achievement domain: A test of the integration of the hopelessness and self-esteem theories of depression. Cognitive Therapy and Research, 26, 531-552.

Abela, J. R. Z., & D'Alessandro, D. U. (2001). An examination of the symptom component of the hopelessness theory of depression in a sample of depressed schoolchildren. Journal of Cognitive Psychotherapy, 15, 33-48.

Abela, J. R. Z., & Sarin, S. (2002). Cognitive vulnerability to hopelessness depression: A chain is only as strong as its weakest link. Cognitive Therapy and Research, 26, 811-829.

Abela, J. R. Z., & Seligman, M. E. P. (2000). The hopelessness theory of depressio\n: A test of the diathesis-stress component in the interpersonal and achievement domains. Cognitive Therapy and Research, 24, 361-378.

Abramson, L. Y., Alloy, L. B., Metalsky, G. I. (1995). Hopelessness depression. In G. M. Buchanan & M. E. P. Seligman (Eds.), Explanatory style (pp. 113-134). Hillsdale, NJ: Erlbaum.

Abramson, L. Y., & Metalsky, G. I. (1986). The Cognitive Style Questionnaire: Measurement of negative cognitive styles about self and consequences. Unpublished manuscript.

Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness depression: A theory based subtype of depression. Psychological Review, 96, 358-372.

Abramson, L. Y., Alloy, L. B., & Metalsky, G. I. (1988). The cognitive diathesis-stress theories of depression: Toward an adequate evaluation of the theories' validates. In L. B. Alloy (Ed.). Cognitive processes in depression (pp. 3-30). New York: Guilford.

Alloy, L. B., Abramson, L. Y., Hogan, M. E., Whitehouse, W. G., Rose, D. T., Robinson, M. S., Kim, R. S., & Lapkin, J. B. (2000). The Temple-Wisconsin Vulnerability to Depression Project Lifetime history of axis I psychopathology in individuals at high and low cognitive risk for depression. Journal of Abnormal Psychology, 109, 403-418.

Alloy, L. B., & Clements, C. M. (1998). Hopelessness theory of depression: Tests of the symptom component. Cognitive Therapy and Research, 22, 303-335.

Alloy, L. B., Hartlage, S., & Abramson, L. Y. (1988). Testing the cognitive diathesis-stress theories of depression: Issues in research design, conceptualization, and assessment. In L. B. Alloy (Ed.), Cognitive processes in depression (pp. 31-73). New York: Guilford.

Alloy, L. B., Just, N., & Panzarella, C. (1997). Attributional style, daily life events, and hopelessness depression: Subtype validation by prospective variability and specificity of symptoms. Cogntive Therapy and Research, 21, 221-244.

American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: APA.

Beck, A. T. (1907). Depression: Clinical, experimental, and theoretical aspects. New York: Harper and Row.

Beck, A. T. (1983). Cognitive therapy of depression: New perspectives. In P. J. Clayton & J. E. Barrett (Eds.), Treatment of depression: Old controversies and new approaches (pp. 265-290). New York: Raven.

Beck, A. T., Brown, G., Steer, R. A., Eidelson, J. L., & Riskind, J. H. (1987). Differentiating anxiety and depression: A test of the cognitive content specificity hypothesis. Journal of Abnormal Psychology, 96, 179-183.

Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561-571.

Brown, G. W., & Harris, T. O. (1978). Social origins of depression: A study of psychiatric disorder in women. New York: Free Press.

Cohen, J., & Cohen, P. (1983). Applied multiple regression/ correlation analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

Follette, V. M., & Jacobson, N. S. (1987). Importance of attributions as a predictor of how people cope with failure. Journal of Personality and Social Psychology, 52, 1205-1211.

Hankin, B. L, Abramson, L. Y., Siler, M. (2001). A prospective test of the hopelessness theory of depression in adolescence. Cognitive Therapy and Research, 25, 607-632.

Hedlund, S., & Rude, S. S. (1995). Evidence of latent depressive schemas in formerly depressed individuals. Journal of Abnormal Psychology, 104, 517-525.

Hollon, S. D., & Kendall, P. C. (1980). Cognitive self- statements in depression: Development of an automatic thoughts questionnaire. Cognitive Therapy and Research, 4, 383-395.

Hollon, S. D., Kendall, P. C., Lumry, A. (1986). Specificity of depressotypic cognitions in clinical depression. Journal of Abnormal Psychology, 95, 52-59.

Hunsley, J. (1989). Vulnerability to depressive mood: An examination of the temporal consistency of the reformulated learned helplessness model. Cognitive Therapy and Research, 13, 599-608.

Ingram, R. E., Bernet, C. Z., & McLaughlin, S. C. (1994). Attention allocation processes in individuals. Cognitive Therapy and Research 18, 317-332.

Ingram, R. E., Miranda, J., & Segal, Z. (1998). Cognitive vulnerability to depression. New York: Guilford.

Metalsky, G. L, Halberstadt, L. J., & Abramson, L. Y. (1987). Vulnerability to depressive mood reactions: Toward a more powerful test of the diathesis-stress and causal mediation components of the reformulated theory of depression. Journal of Personality and Social Psychology, 52, 386-393.

Metalsky, G. I., & Joiner, T. E., Jr. (1992). Vulnerability to depressive symptomatology: A prospective test of the diathesis- stress and causal mediation components of the hopelessness theory of depression. Journal of Personality and Social Psychology, 63, 667- 675.

Metalsky, G. L, & Joiner, T. E., Jr. (1997). The Hopelessness Depression Symptom Questionnaire. Cognitive Therapy and Research, 21, 359-384.

Metalsky, G. L, Joiner, T. E., Jr., Hardin, T. S., & Abramson, L. Y. (1993). Depressive reactions to failure in a naturalistic setting: A test of the hopelessness and self-esteem theories of depression. Journal of Abnormal Psychology, 102, 101-109.

Miranda, J. (1992). Dysfunctional thinking is activated by stressful life events. Cognitive Therapy and Research, 16, 473-483.

Miranda, J., Gross, J. J., Persons, J. B., & Hahn, J. (1998). Mood matters: Negative mood induction activates dysfunctional attitudes in women vulnerable to depression. Cognitive Therapy and Research, 1998, 363-376.

Miranda, J., & Persons, J. B. (1998). Dysfunctional attitudes are mood state dependent. Journal of Abnormal Psychology, 97, 76-79.

Miranda, J., Persons, J. B., & Byers, C. N. (1990). Endorsement of dysfunctional beliefs depends on current mood state. Journal of Abnormal Psychology, 99, 237-241.

Persons, J. B., Miranda, J. (1992). Cognitive theories of vulnerability to depression: Reconciling negative evidence. Cognitive Therapy and Research, 16, 485-502.

Peterson, C. R., Semmel, A., von Baeyer, C., Abramson, L. Y., Metalsky, G. L, & Seligman, M. E. P. (1982). The attributional Style Questionnaire. Cognitive Therapy and Research, 6, 287-299.

Peterson, C., & Villanova, P. (1988). An Expanded Attributional Style Questionnaire. Journal of Abnormal Psychology, 97, 87-89.

Ralph J. A., & Mineka, S. (1998). Attributional style and self- esteem: The prediction of emotional distress following a midterm exam. Journal of Abnormal Psychology, 107, 203-215.

Riskind, J. H., & Rholes, W. S. (1984). Cognitive accessibility and the capacity of cognitions to predict future depression: A theoretical note. Cognitive Therapy and Research, 8, 1-12.

Roberts, J. E., & Kassel, J. D. (1996). Mood-state dependence in cognitive vulnerability to depression: The roles of positive and negative affect. Cognitive Therapy and Research, 20, 1-12.

Robins, C. J., & Hayes, A. M. (1995). The role of causal attributions in the prediction of depression. In G. M. Buchanan & M. E. P. Seligman (Eds.), Explanatory style (pp. 71-98). Hillsdale, NJ: Erlbaum.

Saxe, L. L., & Abramson, L. Y. (1987). The Negative Life Events Questionnaire: Reliability and validity. Unpublished manuscript.

Segal, Z. V., Gemar, M., & Williams, S. (1999). Differential cognitive response to a mood induction following successful cognitive therapy and pharmacotherapy for depression. Journal of Abnormal Psychology, 108, 3-10.

Sharpe, J. P., & Gilbert, D. G. (1998). Effects of repeated administration of the Beck Depression Inventory and other measures of negative mood states. Personality and Individual Differences, 24, 457-463.

Steer, R. A., Beck, A. T., Clark, D. A., & Beck, J. S. (1994). Psychometric properties of the cognition checklist with psychiatric outpatients and university students. Psychological Assessment, 6, 67- 70.

Swendsen, J. D. (1997). Anxiety, depression, and their comorbiclity: An experience sampling test of the helplessness- hopelessness theory. Cognitive Therapy and Research, 21, 97-114.

Teasdale, J. D. (1983). Negative thinking in depression: Cause, effect, or reciprocal relationship? Advances in Behavior Research and Therapy, 5, 3-25.

Teasdale, J. D., & Dent, J. (1987). Cognitive vulnerability to depression: An investigation of two hypotheses. British Journal of Clinical Psychology, 26, 113-126.

Williams, R. M. (1988) Individual differences in the effects of mood on cognition. Unpublished doctoral dissertation, University of Oxford.

Received 10 May 2002; revised version received 11 December 2002

John R. Z. Abela1, Karen Brozina1 and Martin E. P. Seligman

1 Department of Psychology, McGill University, Canada

2 Department of Psychology, University of Pennsylvania, USA

* Correspondence should be addressed to John R. Z. Abela, Deportment of Psychology, McGill University, 1205 Dr. Penfield Avenue, Montreal, Quebec, Canada, H3A IBI (e-mail: abela@ego.psych.megill.ca).

Copyright British Psychological Society Jun 2004

More News in this Category