Mills, R.; Durand, A.; Croal, M.; Rutledge, R. B.; Mason, L.

Selective serotonin reuptake inhibitors (SSRIs) are widely prescribed antidepressants, yet the mechanisms linking their rapid neurobiological effects to more gradual improvements in mood remain unclear. Computational accounts in which mood reflects the integration of reward outcome and reward prediction errors (RPEs) offer a compelling framework for examining SSRI effects, but reliance on laboratory tasks poses a major translational gap, both in timescale and ecological validity. We address this gap by examining the impact of SSRIs on affect-reward dynamics over a clinically relevant timescale using real-world rewards. Across 11 days, 66 healthy participants (n=37 receiving the SSRI citalopram) completed repeated ecological momentary assessments (EMA) of mood and goal-directed activities (up to 8 per day, totaling 3,542 surveys). By probing anticipated reward (pre-activity) and consummatory reward (post-activity), we derived real-world RPEs to test whether SSRIs amplify RPE impacts on momentary mood. Mixed-effects modelling showed that citalopram rapidly amplified the impact of reward outcome and RPEs on momentary mood, and predicted later increases in positive mood. Citalopram also selectively attenuated the impact of negative mood on expectations for future reward. These findings suggest that SSRIs modulate the bidirectional coupling between mood and reward processes in daily life, consistent with a positive feedback loop that could underpin the gradual mood improvement observed for antidepressant treatment. More broadly, our results highlight the utility of EMA as a highly sensitive real-world assay of early pharmacological effects on mood-reward dynamics.