Introduction:
Social media delivers frequent micro-rewards (likes, notifications) that may alter dopamine-based reward processing. Heavy use has been linked to anhedonia, apathy, and sleep disruption, yet neuropsychological mechanisms remain poorly understood, particularly in non-neuroimaging, low-resource settings. Integrating computational modeling with digital phenotyping provides a cost-effective pathway to study these effects in young adults.


Purpose:
This pilot study investigates whether high social-media users show altered reward prediction error (RPE) signals, reduced social reward responsivity, diminished effort-based motivation, and whether these effects are moderated by sleep disruption.


Method:
A cross-sectional, between-groups design (N = 60 university students, 18-30 years) compared high vs low social-media users, identified via the Social Media Use Integration Scale. Baseline testing included: (a) a probabilistic reinforcement-learning task (computational RPE estimates), (b) a simulated social reward task (Instagram-like approvals), and (c) the Effort Expenditure for Rewards Task (EEfRT). Self-report measures captured anhedonia (TEPS), apathy (AES), and mood (PHQ-9, GAD-7). Participants installed a smartphone-based digital phenotyping app (Beiwe/Ethica/MindLAMP) and completed 14-day sleep diaries to track passive data (screen time, app sessions, circadian misalignment). Data analyses test between-group differences in RPE, mediation of social reward responsivity, and moderation by sleep disruption.


Results:
Preliminary analysis shows that high users show (i) reduced RPE magnitude, (ii) blunted responsivity to social approval, (iii) fewer high-effort choices, and (iv) stronger effects in those with greater social jetlag.


Conclusions:
This study proposes an integrative framework linking social-media overuse with dopaminergic reward desensitization, motivational deficits, and sleep misalignment. By combining computational modeling, behavioral tasks, and smartphone-based phenotyping, it demonstrates a scalable, low-cost approach to studying the psychological impacts of digital ecosystems in emerging economies.