BrainPT, a Cognitive Mini-Game App for Short-Form Video Use in Young Adults: Neural and Behavioral Pilot Study

Background: Short-form video elicits a stimulus-driven, passive neural state, yet existing interventions target time of use alone and cannot show whether the mechanism of cognitive control has changed. Objective: This pilot study developed BrainPT, a mini-game-based cognitive intervention, and evaluated whether a single insertion alters the neural state during short-form viewing (Study 1) and whether repeated use improves behavioral self-regulation (Study 2). Methods: Study 1 used a within-subject design (n=9) in which electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) were recorded in the same session across short-form viewing, a single BrainPT set, and goal-directed tasks. Study 2 used a nonrandomized 2-week pretest-posttest design comparing an experimental group that used BrainPT daily (n=13) with a control group (n=8). Both visits followed the same in-laboratory protocol, and app logs and self-report indices were collected over the intervention period. Results: In Study 1, short-form viewing was accompanied by prefrontal deactivation: EEG power fell below baseline in all four bands and most in alpha (-1.72 dB, P=.01; q=.054 after correction). Prefrontal activation was opposite in sign between the goal-directed tasks and viewing (+0.180 vs -0.093, P=.002), and the viewing block immediately after a BrainPT insertion exceeded the block before it (+0.110 vs -0.093, P=.02). The same condition ordering appeared in an independent preintervention sample (n=18). In Study 2, daily short-form use fell from 143.5 to 50.9 min in the experimental group and rose from 104.8 to 132.9 min in the control group, a between-group difference in change of 120.7 min (P=.004). The Adult ADHD Self-Report Scale did not differ between groups (P=.39), and no neural index separated the groups after 2 weeks. Conclusions: A single insertion produced a measurable neural change: prefrontal state during the viewing that followed moved toward the goal-directed task. After 2 weeks of use that neural difference was not detectable between groups although behavior differed, so repeated use may act through the momentary state during viewing rather than by reorganizing the resting state. Both samples were small and nonrandomized, the primary outcome was self-reported, and the analyses were exploratory and not preregistered, so these estimates require confirmation in a preregistered randomized trial with objectively logged use.