Do momentary changes in mood overly bias reward perception in bipolar disorder? A computational psychiatry approach
Year: 2025
Type: dissertation
Abstract: Aims: Bipolar disorder (BD) is characterised by cycles where mood and goal-directed behaviour escalate to extremes. A clear mechanism for what drives mood fluctuations is lacking, however, emerging neuro-computational models conceptualising mood as “momentum” are promising. The current study investigates whether momentary changes in mood excessively bias reward perception in BD by taking a computational approach. / Method: 48 participants with BD diagnoses and 43 matched controls completed a “mood bias” game multiple times per week for eight weeks. This game was a two- alternative forced-choice reward learning task where participants learned to select the more frequently rewarded stimulus. Learning was assessed before and after a mood manipulation. A subsequent test phase assessed mood bias; the extent that stimuli preferences were biased by the mood manipulation. We predicted that: mood bias would be stronger in the BD group than the control group; this would be modulated by residual manic and depressive symptoms; and that stronger mood bias would predict intensification of symptoms over the span of the study. / Results: The mood manipulation had comparable impact on mood across the BD and control groups. Within the BD group, higher levels of depression symptoms predicted higher sensitivity to variation in the value of the good and bad outcomes received in the mood manipulation. Crucially, the BD group displayed a stronger mood bias on their preferences, per unit of mood change, when accounting for manic and depressive symptoms. Depressive symptoms reduced the propensity for the mood manipulation to bias reward perception in the BD group. Intriguingly, a stronger mood bias predicted the intensification of depressive symptoms at the end of the experiment. / Conclusions: This is the first study to demonstrate that mood bias propensity significantly influences reward valuation in individuals with bipolar disorder and predicts their clinical trajectory. Consistent with neuro-computational models, these findings support the view that mood disturbances in BD are driven by a recursive propensity for mood to bias reward perception.
Author: Karamanci, Ceyla
Institution: University College London
Language: English
Subfield: Psychiatry and Mental health
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7128279086