Objectives. There are dysfunctional cognitions that may be characteristic of bipolar disorder (BD), and which may be mood-state dependent. However, it has been found that such cognitions may be resilient to minor positive mood increase. The aim of this study is to investigate whether positive mood-induction procedure has the effect of altering the availability of dysfunctional schemas in a group of individuals with BD. Design. The sentence completion task was designed to assess the content of dysfunctional schemas: it was modified for use with positive mood-induction procedure in the current study so that more dysfunctional schematic models would lead to completion of sentence stems by negative constructs, whereas functional schematic models led to completion of sentence stems by positive constructs. Methods. Using the modified sentence completion task, 30 participants with remitted bipolar I disorder were compared with 30 individuals with no history of affective disorder. Results. At baseline the bipolar group inserted significantly more dysfunctional completions than the control group. Following mood induction, the number of dysfunctional completions was reduced for both groups. The bipolar group still inserted significantly more dysfunctional completions relating to autonomy than the control group following mood induction. However, no significant group by time interactions were identified. Conclusions. As predicted, participants' reports of dysfunctional attitudes reduced following the mood-induction procedure, although no difference was identified between the groups in terms of the size of this reduction. The bipolar group continued to insert significantly more dysfunctional completions for the factor of autonomy, suggesting that this group has more access to autonomy schemas, regardless of change in mood. This finding may have implications in terms of focus of therapy and relapse prevention work.