Executive control and emotional processing biases in depressive patients

Abstract

Depressed patients show cognitive deficits along with mood disturbances. Growing evidence suggests impairment at the level of executive control, which might account in part for patients' difficulties in everyday activities and cognitive performance. Furthermore, there is evidence that depressive patients show information processing biases for emotional information which are thought to play a role in the etiology and maintenance of the disorder. Attentional bias occurs in an early stage of information processing, while memory bias occurs in a later stage of processing (strategic elaboration). The goal of this study was to investigate executive control (the Stroop test) and information processing biases for emotional information in an early stage of processing (the emotional Stroop test) and in a later stage of processing (memory recognition test) in healthy subjects and depressive patients. A further objective of this study was to compare the performance of melancholic and non-melancholic depressive patients in the Stroop test, in the emotional Stroop test and in the memory recognition test. Last, we wanted to investigate the relationship between the performance in an executive control task (the Stroop effect) and information processing bias measures for emotional information. This study is the first to investigate the Stroop test, the emotional Stroop test and the memory recognition test in the same healthy subjects and depressed patients. Furthermore, this is the first study investigating information processing biases for emotional information in the melancholic and non-melancholic patients. Twenty-three depressive patients and 27 healthy subjects performed computerized mixed trial Stroop and emotional Stroop tests. Afterwards, the subjects performed the memory recognition task. Depressive patients were divided according to DSM-IV diagnosis into melancholic and non-melancholic subgroups. Furthermore the level of anxiety and depression was assessed in all subjects. Results of the Stroop task showed that when the depressed patients were analyzed as a whole group, they showed only a trend toward a larger Stroop effect at the beginning of the task. When the analysis was performed with the melancholic and non-melancholic subgroups, contrary to the expectations, only the non-melancholic patients were impaired in the Stroop task compared to the melancholic patients and healthy subjects. Furthermore, we failed to find evidence for an attentional bias in the depressed patients in the emotional Stroop task measured as longer RTs to the emotional compared to neutral stimuli. However, both groups committed more errors in the negative compared to the neutral and positive condition. Memory bias was examined with the memory recognition test since it allowed us to study both “pure” memory and response bias. Response accuracy d’ and response bias beta were calculated according to the signal-detection model. We failed to find evidence for a memory bias in depressed patients measured as discrimination accuracy d’. Considering the response bias measure beta, the analysis showed that the healthy subjects had a more conservative response bias toward positive stimuli. This means that healthy subjects were less likely to answer “yes” to the positive stimuli than to other stimuli. The patients on the other hand had a more conservative response bias toward both negative and positive stimuli compared to neutral stimuli. Contrary to the expectations, there were no differences between the melancholic and non-melancholic patients. The results of the correlational analysis provide evidence that the executive control and emotional information processing are connected phenomena in the healthy subjects but not in the depressed patients. The healthy subjects with poor executive control are paying more attention to the negative stimuli compared to neutral stimuli. This was not the case in the depressed patients. We suggest that the unexpected result of melancholic patients performing better than non-melancholic ones in the Stroop task may be due to their more pronounced rigidity, which makes them more resistant to distraction. Hence, more detailed psychopathological assessment is desirable for future investigations of the melancholic patients. Furthermore, since we failed to find attentional bias in the depressed patients toward the emotional stimuli in the emotional Stroop test, we are concluding that besides methodological issues there are more important clinical factors than diagnosis (i.e. trait anxiety). We are suggesting that memory bias is impossible or difficult to demonstrate in the depressed patients when stimulus exposure occurs under sets that are explicitly antithetical to self-referencing. The relationship found between the Stroop effect and the emotional Stroop effect in the healthy subjects is suggesting that healthy individuals with lower levels of executive control may be more vulnerable to depression.