AI-optimized trial shows mental health app can ease university student distress

Credit: Unsplash/CC0 Public Domain

Researchers from Deakin University, Black Dog Institute and UNSW have completed an AI-optimized trial testing digital mental health interventions for university students experiencing psychological distress.

Published on 1 November in JAMA Network Open, the breakthrough study tested three self-guided, smartphone programs—physical activity, mindfulness and sleep hygiene—in more than 1,200 university students who reported elevated distress.

Using a AI-enhanced adaptive trial design, the research team tested multiple interventions and rapidly identified which delivered the greatest benefits for individual students—cutting months off traditional clinical trial timelines.

Professor Sunil Gupta from Deakin’s Applied Artificial Intelligence Initiative said the study demonstrates the potential to personalize mental health care at scale.

“Our algorithm learns as it goes, sending each person to the treatments that show the most promise for them. This helps researchers quickly figure out which interventions will work best for that individual.

“This novel trial methodology reduces the need for large control groups, meaning more participants received personalized interventions when they needed them.”

Professor Jill Newby from the Black Dog Institute and UNSW said that this is the first time AI has been used to match digital mental health tools to students’ individual needs.

“We were able to learn, in real time, which brief programs worked best for different levels of distress—so we can deliver faster, more effective and personalized support.”

The trial has been delivered entirely through the Vibe Up smartphone app. Dr. Leonard Hoon from Deakin Applied AI said the study was supported on the Conductor platform, a novel platform designed to enable AI-enhanced adaptive trials.

Participants—each currently enrolled in tertiary education—received remote access to brief, psychoeducational interventions designed to fit into their daily routines. The study ran a series of AI-powered mini-trials, each lasting up to one month, to identify which treatments worked best for different levels of distress.

The study found that mindfulness and physical activity worked best for students with severe distress, while sleep hygiene and physical activity were most effective for those with mild distress. For moderate distress, no single approach stood out, highlighting the need for further tailored options.

Traditional clinical trials often take years and require large control groups. In contrast, this adaptive design used near real-time data to direct participants to the most promising interventions as results emerged. Researchers estimate a conventional trial would require 25% more participants to detect the same effects—demonstrating the power of AI to make mental health research faster and more efficient.

What’s next?

The research team are now building on these findings in two new directions:

Expanding the adaptive trial method: testing whether this AI-driven approach can compare different digital treatments for depression, including cognitive behavioral therapy (CBT) and lifestyle-based interventions.
Using digital markers for prediction: testing whether smartphone-tracked digital markers can predict how well a person responds to or benefits from the Vibe Up interventions.

More information:
Jill Newby et al, Brief Digital Interventions for Psychological Distress, JAMA Network Open (2025). DOI: 10.1001/jamanetworkopen.2025.40502

Provided by
Deakin University