Negative self-referential processing has fruitfully been studied in unipolar depressed patients, but remarkably less in patients with bipolar disorder (BD). This exploratory study examines the relation between task-based self-referential processing and depressive symptoms in BD and their possible importance to the working mechanism of mindfulness-based cognitive therapy (MBCT) for BD. The study population consisted of a subsample of patients with BD (n = 49) participating in an RCT of MBCT for BD, who were assigned to MBCT + TAU (n = 23) or treatment as usual (TAU) (n = 26). Patients performed the self-referential encoding task (SRET), which measures (1) positive and (2) negative attributions to oneself as well as (3) negative self-referential memory bias, before and after MBCT + TAU or TAU. At baseline, all three SRET measures were significantly related to depressive symptoms in patients with BD. Moreover, repeated measures analyses of variance revealed that negative self-referential memory bias diminished over time in the MBCT + TAU group, compared with the TAU group. Given the preliminary nature of our findings, future research should explore the possibly mediating role of reducing negative self-referential memory bias in preventing and treating depressive symptoms in BD through MBCT.

Background: Negative self-referential processing has fruitfully been studied in unipolar depressed patients, but remarkably less in patients with bipolar disorder (BD). This study examines the relation between task-based self-referential processing and depressive symptoms in BD and their possible importance to the working mechanism of mindfulness-based cognitive therapy (MBCT) for BD.Methods: The study population consisted of a subsample of patients with BD (n = 49) participating in an RCT of MBCT for BD, who were assigned to MBCT+TAU (n = 23) or treatment as usual (TAU) (n = 26). Patients performed the self-referential encoding task (SRET), which measures (1) positive and (2) negative attributions to oneself as well as (3) negative self-referential memory bias, before and after MBCT. Results: At baseline all three SRET measures were significantly related to depressive symptoms in patients with BD. Moreover, repeated measures analyses of variance revealed that negative self-referential memory bias diminished over time in the MBCT+TAU group, compared with the TAU group. Conclusions: Task-based self-referential processing is related to depressive symptoms in BD. Moreover, the negative self-referential memory bias was specifically amenable to change through MBCT. Future research should explore the possibly mediating role of reduction of negative self-referential memory bias in preventing and treating depressive symptoms in BD through MBCT.