The neurocognitive cost of media multitasking: fragmented screens, fragmented minds!

As “digital natives,” Generation Z university students rely heavily on mobile devices across all life domains. Diverging from the literature’s historical focus on total screen time, this study aimed to objectively characterize the entire digital footprint (including duration, fragmentation, and content type) and investigate how these patterns are associated with cognitive control. We conducted a cross-sectional digital phenotyping study with 206 health sciences students. Objective 7-day smartphone logs were integrated with a custom mobile Stroop task completed on participants’ own devices to ensure high ecological validity. A clinical battery assessed sleep quality, Attention Deficit Hyperactivity Disorder (ADHD) symptoms, anxiety, headache, neck disability, and computer vision syndrome. The Cumulative Digital Load Score was developed and compared with objective device metrics. Theory-driven family-based correlation analyses (13 FDR-corrected tests) examined screen pattern–Stroop associations, and trial-level Linear Mixed-Effects Models (LMMs) isolated momentary predictor effects across 12,035 trials. Objective data revealed a highly intensive and fragmented digital profile: a median of 6.6 h of daily screen time, 142 pickups, an average session duration of 2.9 min, and a 40% passive consumption ratio. Family-based analyses demonstrated that fragmentation metrics were significantly associated with conflict resolution costs (FDR q < 0.05). LMMs revealed that raw total screen time did not predict cognitive accuracy. The models identified two distinct impairment profiles: high digital attention fragmentation was associated with selective weakening of proactive conflict control (increasing errors specifically on incongruent trials), whereas clinical/somatic symptom burden and ADHD symptomatology were associated with a broader accuracy cost that, particularly for somatic pain burdens, extended even to highly automated congruent trials. Convergence between the self-reported CDLS and the objective device logs was weak (ρ ≈ 0.14–0.18), indicating that the two capture related but distinct aspects of digital behaviour rather than interchangeable estimates of the same construct. Digital exposure does not cause a single, homogenous cognitive deficit. Highly fragmented use, characterized by constant interruptions and short session durations, is associated with poorer proactive cognitive control and a specific accuracy cost. Passive consumption is associated with block-dependent processing delays without affecting highly automatic cognitive tasks. Because the two blocks were administered in a fixed order, this contrast is confounded with task order and practice. These findings suggest that cognitive health in the digital age may benefit from targeted “digital hygiene” strategies prioritizing the minimization of attention fragmentation, rather than solely focusing on screen time reduction.