mPulse Mobile Sensing Model for Passive Detection of Impulsive Behavior: Exploratory Prediction Study.

Area of Science:

• Digital Health
• Behavioral Science
• Mobile Sensing Technology

Background:

• Mobile health technology enables continuous data collection on patient activity and cognition.
• Passive mobile metrics like battery life and screen time can offer insights into mental health.
• Impulsivity is a key factor in many health issues, yet understanding it via mobile data is underexplored.

Purpose of the Study:

• To assess the feasibility of using mobile sensor data to passively detect and monitor self-reported state impulsivity and impulsive behaviors.
• To explore the relationship between passive mobile sensing and various facets of impulsivity.

Main Methods:

• A 21-day continuous mobile sensing study involving 26 participants on iOS and Android platforms.
• Utilized the mPulse mobile sensing system to collect data on call logs, battery charging, and screen checking.
• Validated models using mobile sensing features to predict impulsivity traits, behavioral measures, and ecological momentary assessments (EMA).

Main Results:

• Passive mobile phone usage data (call logs, battery, screen time) significantly predicted sensation seeking, planning, and perseverance traits.
• Daily passive sensing successfully predicted objective behavioral measures like present bias, attention errors, and risk-taking task performance.
• Models predicted daily EMA responses on affect, stress, and productivity, but not direct measures of previous-day impulsivity.

Conclusions:

• Everyday smartphone sensors hold potential for developing impulsivity phenotypes and detecting impulsive behavior.
• Passive sensing offers a novel approach to understanding and monitoring impulsivity in real-world settings.
• Further research is needed to refine passive sensing models for enhanced precision in detecting impulsive behaviors.