Daily Affect Fluctuations in Phone Screen Content Predict Anxiety and Depressive Symptoms

2026-03-13

AI Summary

The relationship between digital media use and mental health remains unclear, particularly due to a lack of fine-grained, real-world tracking of individuals’ dynamic digital behaviors. This study employs an intensive longitudinal design over one year, capturing smartphone screenshots every five seconds and pairing them with biweekly psychological assessments. Using deep learning models, the emotional valence and arousal of screen content are quantified. For the first time at the within-person level, the study reveals that greater exposure on a given day to low-arousal negative content predicts higher subsequent levels of anxiety and depression. Moreover, within-person fluctuations in emotional exposure significantly outperform between-person differences in explaining psychological symptoms. These findings transcend the limitations of traditional cross-sectional or between-subject comparisons and underscore the highly individualized nature of digital behavior.

Technology Category

• Application Domains: Humanities & Computational Social Science
• Humans and AI: Emotional Intelligence
• Cognitive Modeling & Cognitive Systems: Affective Computing

Application Category

• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

The relationship between digital media use and mental health remains poorly understood, in part because real-world digital behavior is rarely captured at scale. This intensive longitudinal study tracked participants' complete natural smartphone interactions over one year. We collected screenshots every 5 seconds from 145 adults (yielding 111 million screenshots), alongside biweekly assessments of anxiety and depression (mean = 24 surveys). The valence and arousal of each screenshot were assessed using a deep learning affect model. Individuals showed highly idiosyncratic media patterns, with substantially more variance in anxiety and depression accounted for within-person than between-person. Day-to-day fluctuations in the valence and arousal of a person's screen content predicted subsequent changes in depression and anxiety, whereas between-person differences did not. Specifically, greater exposure to low-arousal negative content was associated with higher depression and anxiety. These findings underscore the dynamic, idiosyncratic nature of digital consumption and the need for targeted measurement and intervention.

Problem

• digital media use
• mental health
• affect fluctuations
• anxiety
• depression

Innovation

• intensive longitudinal study
• smartphone screenshot tracking
• deep learning affect model
• within-person dynamics
• digital phenotyping