"In my defense, only three hours on Instagram": Designing Toward Digital Self-Awareness and Wellbeing

Paper Title

"In my defense, only three hours on Instagram": Designing Toward Digital Self-Awareness and Wellbeing

Publication Info

• Topic area: Digital wellbeing and self-awareness in technology use.
• Keywords: Digital wellbeing, self-awareness, smartphone usage, estimated-actual gap, self-reflection, HCI, emotional wellbeing, technology probe, screen time, intentional engagement.

Background and Problem

• Problem / challenge: Existing digital wellbeing tools focus predominantly on restricting screen time, which oversimplifies the complex role of technology in users' lives and neglects its potential for meaningful engagement.
• Significance: Addressing digital wellbeing is crucial for balancing the pervasive role of technology in daily life, especially among heavy users such as college students, to mitigate negative effects like stress and reduced satisfaction.
• Motivation and related work: Prior research has explored screen time tracking, timers, and self-control mechanisms but often fails to promote lasting behavioral change or mindful engagement. The paper builds on the concept of self-awareness and the estimated–actual gap (E–A gap) to explore its potential in fostering intentional technology use.

Solution

• Proposed approach: WellScreen, a lightweight technology probe designed to scaffold daily self-reflection on smartphone usage by comparing estimated and actual use.
• Novelty:
  1. Empirical evidence demonstrating the promise of lightweight interventions in scaffolding self-reflections for digital wellbeing.
  2. Insights into how users develop a holistic understanding of wellbeing, including emotional and social dimensions, through reflective engagement.
  3. Design recommendations for technologies that enable digital self-awareness through estimation–reflection workflows.
• Procedure and key techniques:
  • Participants estimate smartphone usage at the start and end of the day.
  • Actual usage data is manually logged from device metrics.
  • Visualizations juxtapose estimated and actual usage, prompting reflection.
  • Daily reflections and surveys capture self-assessments of digital habits.

Results

• Concrete findings:
  • Participants underestimated productivity and social app usage by 16.77% and 9.49%, respectively, while overestimating entertainment app usage by 11.32%.
  • Positive affect increased by 10% (Cohen’s d=0.63, p<0.05) after using WellScreen.
  • Smaller E–A gaps were associated with higher satisfaction and self-control but also greater stress and lower goal adherence.
• Advantage over baselines: WellScreen enabled structured self-reflection, helping participants recalibrate expectations and recognize patterns in their smartphone usage, which traditional screen-time tracking tools fail to achieve.
• Experiments / evaluation:
  • Two-week deployment with 25 college students (20 completing).
  • Surveys measured personality traits, self-regulation, self-control, and emotional wellbeing.
  • Interviews provided qualitative insights into participants' reflections and experiences.
• Limitations and future work:
  • Small sample size and short-term deployment limit generalizability.
  • Observer effects and novelty may have influenced behavior.
  • Future work should explore longer-term studies, diverse populations, and refined prototypes.

Summary

This study introduced WellScreen, a lightweight intervention for fostering digital self-awareness through daily self-reflection on smartphone usage. Findings revealed significant discrepancies in participants' estimated and actual usage patterns, highlighting the limits of screen time as a proxy for wellbeing. WellScreen improved positive affect and encouraged participants to recalibrate expectations and engage more intentionally with technology. The study underscores the importance of designing digital wellbeing tools that prioritize self-awareness and meaningful engagement over restriction, offering design implications for future interventions that integrate prediction–reflection workflows, contextual journaling, and AI-driven co-interpretation of data.