Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Currently submitted to: Journal of Medical Internet Research

Date Submitted: Jan 27, 2026
Open Peer Review Period: Jan 28, 2026 - Mar 25, 2026
(currently open for review)

Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.

A Personalized Exercise Assistant using Reinforcement Learning (PEARL): Results from a four-arm Randomized-controlled Trial

  • Amy Armento Lee; 
  • Narayan Hegde; 
  • Nina Deliu; 
  • Emily Rosenzweig; 
  • Arun Suggala; 
  • Sriram Lakshminarasimhan; 
  • Qian He; 
  • John Hernandez; 
  • Martin Seneviratne; 
  • Rahul Singh; 
  • Pradnesh Kalkar; 
  • Karthikeyan Shanmugam; 
  • Aravindan Raghuveer; 
  • Abhimanyu Singh; 
  • Hariharan Manoharan; 
  • My Nguyen; 
  • James Taylor; 
  • Jatin Alla; 
  • Sofia S. Villar; 
  • Hulya Emir-Farinas

ABSTRACT

Background:

Consistent physical inactivity among adults and adolescents poses a major global health challenge. Mobile health (mHealth) interventions, particularly Just-in-Time Adaptive Interventions (JITAIs), offer a promising avenue for scalable and personalized physical activity promotion. However, developing and evaluating such adaptive interventions at scale, while integrating robust behavioral science, presents methodological hurdles.

Objective:

The PEARL study aimed to assess the feasibility and effectiveness of a reinforcement learning (RL) algorithm, informed by health behavior change theory (COM-B), to personalize the content and timing of physical activity nudges via the Fitbit app compared to fixed and random nudging strategies, and to a control group with no nudges.

Methods:

We conducted a large-scale, four-arm randomized controlled trial (RCT) enrolling 13,463 Fitbit users. Participants were randomized to: (1) Control (no nudges); (2) Random (random content/timing); (3) Fixed (logic based on baseline COM-B survey); and (4) RL (adaptive algorithm). The primary outcome was the change in average daily step count from baseline to 2 months. Secondary outcomes included user engagement and survey responses regarding capability, opportunity, and motivation.

Results:

7,711 participants were included in the primary analysis (mean age 42.1 years; 86.3% female). At 1 month, the RL group showed a significant increase in daily steps compared to Control (+296 steps, P<.001), Random (+218 steps, P=.005), and Fixed (+238 steps, P=.002) groups. At 2 months, the RL group sustained a significant increase against the Control (+210 steps, P=.01). Generalized estimating equation (GEE) models confirmed a sustained significant increase in the RL group (+208 steps, P=.002). In exit surveys, the RL group reported higher favorable responses regarding nudge customization (37%) compared to other groups.

Conclusions:

This study demonstrates the feasibility and early efficacy of using RL to personalize digital health nudges at scale. While long-term retention remains a challenge, the adaptive approach outperformed static behavioral rules, showcasing the promise of dynamic personalization in a real-world mHealth setting. Clinical Trial: doi: 10.17605/OSF.IO/TW7UP


 Citation

Please cite as:

Armento Lee A, Hegde N, Deliu N, Rosenzweig E, Suggala A, Lakshminarasimhan S, He Q, Hernandez J, Seneviratne M, Singh R, Kalkar P, Shanmugam K, Raghuveer A, Singh A, Manoharan H, Nguyen M, Taylor J, Alla J, Villar SS, Emir-Farinas H

A Personalized Exercise Assistant using Reinforcement Learning (PEARL): Results from a four-arm Randomized-controlled Trial

JMIR Preprints. 27/01/2026:91156

DOI: 10.2196/preprints.91156

URL: https://preprints.jmir.org/preprint/91156

Download PDF


Request queued. Please wait while the file is being generated. It may take some time.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.