Depressive realism

Depressive realism is the hypothesis developed by Lauren Alloy and Lyn Yvonne Abramson that depressed individuals make more realistic inferences than do non-depressed individuals. Although depressed individuals are thought to have a negative cognitive bias that results in recurrent, negative automatic thoughts, maladaptive behaviors, and dysfunctional world beliefs, depressive realism argues not only that this negativity may reflect a more accurate appraisal of the world but also that non-depressed individuals' appraisals are positively biased. This theory remains very controversial, as it brings into question the mechanism of change that cognitive behavioral therapy for depression purports to target. While some of the evidence currently supports the plausibility of depressive realism, its effect may be restricted to a select few situations.

When participants were asked to press a button and rate the control they perceived they had over whether or not a light turned on, depressed individuals made more accurate ratings of control than non-depressed individuals. Among participants asked to complete a task and rate their performance without any feedback, depressed individuals made more accurate self-ratings than non-depressed individuals. For participants asked to complete a series of tasks, given feedback on their performance after each task, and who self-rated their overall performance after completing all the tasks, depressed individuals were again more likely to give an accurate self-rating than non-depressed individuals. When asked to evaluate their performance both immediately and some time after completing a task, depressed individuals made accurate appraisals both immediately before and after time had passed.

In a functional magnetic resonance imaging study of the brain, depressed patients were shown to be more accurate in their causal attributions of positive and negative social events than non-depressed participants who demonstrated a positive bias. This difference was also reflected in the differential activation of the fronto-temporal network, higher activation for non self-serving attributions in non-depressed participants and for self-serving attributions in depressed patients, and reduced coupling of the dorsomedial prefrontal cortex seed region and the limbic areas when depressed patients made self-serving attributions.

...
Wikipedia