حالت حافظه سازگار در بی قراری: نقش حالت احساسی و سبک شناختی

Biases in incidental memory for self-referent adjectives and intentional memory were compared across nondysphoric (ND; n=48n=48), experimentally induced dysphoric (EXP; n=49n=49), and naturally dysphoric (NAT; n=48n=48) individuals. Negative biases, “evenhandedness”, and positive biases were demonstrated among NAT, EXP and ND participants, respectively, in terms of incidental memory. Correlation analyses suggested that the effects of cognitive style (self-esteem, dysfunctional attitudes, and attributional style) are limited to negative stimuli. Memory for incidental positive stimuli was only predicted by state affect. Groups did not differ in performance on an intentional memory task. Implications for network and schema models of depression are explored.

Mood congruent recall has been demonstrated in a number of studies investigating memory processes in dysphoria. Whereas nondysphoric individuals tend to recall more positive than negative experimentally presented stimuli, dysphoric and depressed individuals tend to recall either approximately equal proportions of positive and negative stimuli, or more negative than positive stimuli (see Blaney, 1986; Matt, Vasquez, & Campbell, 1992 for reviews). This effect is typically more pronounced for self rated items (e.g., Does this word describe you?) versus semantically rated items (e.g., Does this word have a specific meaning or relate to a specific situation?). As an example, Gilboa, Roberts, and Gotlib (1997) found that mildly depressed individuals demonstrated an evenhanded recall of positive and negative valenced stimuli that were previously rated in terms of whether they described the participants’ personality, whereas nondepressed individuals demonstrated a positive recall bias following the same rating task. A number of other studies have reported similar findings with depressed and dysphoric individuals (Bradley & Mathews, 1983; Breslow, Kocsis, & Belkin, 1981; Mathews & Bradley, 1983). Nonetheless, findings have not been entirely consistent across the literature (for example see Bradley, Mogg & Williams, 1994; Ilsley, Moffoot, & O’Carroll, 1995 for failures to demonstrate mood congruent effects with measures of explicit memory). One recent review suggests that greater attention to both stable characteristics of the person and transient mood states might help explain variability in mood-congruency effects ( Rusting, 1998), and in this regard distinctions between the basic premises of Beck's schema model and Bower's network model are relevant. Beck (1967) and Beck (1976) suggests that negative schemata in depression bias all aspects of information processing, including memory. Schemata consist of organized elements of past reactions and experiences that form a relatively cohesive and persistent body of knowledge capable of guiding subsequent perceptions and appraisals ( Segal, 1988). In the case of a nondepressed person, schemata may consist of such constructs as competence, attractiveness, and mastery, whereas in the case of a depressed person schemata may consist of constructs such as failure, inadequacy and despair. These schemata would facilitate elaborative processing of thematically congruent stimuli, leading to biases in recall of positively valenced stimuli among nondepressed individuals and biases in recall of negatively valenced stimuli among depressed individuals. Presumably, the effects of these stable cognitive schemata would have a greater impact on memory for valenced stimuli compared to the effects of transient mood states on memory. Bower (1981) and Bower (1987) associative network model of mood and memory posits that there are pathways between mental nodes representing propositions, concepts and emotions. “Emotion nodes” can be linked to a number of related nodes involving semantic content. For example, an individual's sadness node might be linked to other nodes representing concepts such as “failure”, “loser” and “incompetent.” As a result of spreading activation, the experience of sadness would lead to increased activation of each of these related nodes. This heightened activation of concepts linked to the sadness node would lead to enhanced memory for mood congruent stimuli compared to mood incongruent stimuli. In other words, material that is congruent with an individual's mood, either at the time of storage or retrieval, will be better recalled than material that is incongruent with an individual's mood. Consequently, Bower's model suggests that both naturally occurring subclinical depressive symptoms and transient dysphoric mood states have similar effects on memory for valenced stimuli. Beyond the issue of the general impact of transient mood states on memory, a number of additional issues remain unresolved. First, while memory involves multiple components and processes, it is uncertain which of these are affected by depression and transient dysphoric mood states. The majority of research has investigated incidental recall in which participants are given a “surprise” memory test of words that were previously read. Studies that have focused on intentional recall have yielded less consistent results (e.g., Gilboa et al., 1997). It is possible that depressed individuals are able to override cognitive biases when they explicitly rehearse material for later testing. If so, it is likely that this type of override would be more probable among individuals with transient dysphoric mood states than among individuals with more long-standing and serious experiences with depressive symptoms. Second, it is unclear whether memory biases associated with depression are primarily the result of enhanced recall of negatively valenced stimuli or deficits in the recall of positively valenced stimuli or both. This issue has bearing on the theoretical question of whether depression is associated with the facilitated processing of negative information or the impaired processing of positive information (see Allen, Woolfolk, Gara, & Apter, 1996; Kuiper, Olinger, & Swallow, 1987; Woolfolk, Noalany, Gara, Allen, & Polino, 1995). It is also possible that transient mood states and more persistent depressotypic characteristics (e.g., negative schemata) have different effects on the processing of positive versus negative information. Finally, little work has investigated the potential contribution of negative cognitive style to mood congruent biases. Beck (1967) and Beck (1976) schema model suggests that these memory biases result from deeply ingrained dysfunctional beliefs. Consequently, this model predicts that memory biases should be associated with negative cognitive styles. Consistent with this hypothesis, Alloy, Abramson, Murray, Whitehouse, and Hogan (1997) found that nondepressed individuals who had been selected by virtue of having both negative attributional styles and high levels of dysfunctional attitudes exhibited negative memory biases relative to those with low scores on both of these measures. Likewise, a series of three recent studies suggests that lower self-competence and lower self-liking are associated with enhanced memory for failure-related and negative social content words ( Tafarodi, Marshall & Milne, 2003). The present study was designed to determine: (1) the extent to which memory disturbances resulting from experimentally manipulated dysphoric mood resemble those associated with naturally occurring depressive symptoms; (2) if dysphoria has greater effects on memory for positive versus negative stimuli; (3) if dysphoria is associated with disturbances in both incidental and intentional memory; and (4) the degree to which mood state, depressotypic cognitions, such as negative attributional style, dysfunctional attitudes, and low self-esteem, are associated with memory for positive and negative words.