Session: Government Agency Student Posters

Paper Number: 173503

Adaptive Vr-Based Mindfulness Intervention Using Ai-Powered Metahumans and Biosensing for Oral Nicotine Pouch Use in Young Adults 

Introduction

The use of oral nicotine pouches (ONPs)—a relatively new class of smokeless tobacco products—is rapidly increasing among young adults, presenting new challenges for nicotine dependence prevention and treatment. Despite early marketing efforts to position ONPs as harm reduction tools, emerging research suggests that they may reinforce nicotine dependence and contribute to sustained usage patterns. Current intervention strategies often fail to address the specific behavioral and psychological profiles of ONP users, who may differ from traditional smokers or vapers in terms of use context, motivations, and perceived risk. 

We are developing iPAL MIND, an immersive, AI-enhanced virtual reality (VR) platform designed to deliver real-time, mindfulness-based interventions tailored to individual users. The goal is to promote emotional regulation, reduce craving, and support long-term abstinence among young adult ONP users.

Methods

The iPAL MIND system uses a multi-modal sensing and feedback architecture. College students between 18-25 who currently use ONPs are equipped with biometric and neuroimaging sensors including functional near-infrared spectroscopy (fNIRS) to monitor cortical activity associated with attention and emotional regulation, and heart rate sensors to detect physiological arousal and stress. Participants are randomly assigned to guided mindfulness sessions facilitated by either a human instructor or a lifelike MetaHuman avatar. For the avatar lead group, a large language model (LLM) continuously analyzes incoming physiological data and adjusts the avatar’s instructions in near real-time to infer emotional and cognitive states enabling personalized and adaptive guidance from the MetaHuman. The data collection, state inference, and adaptive feedback structure is designed to maximize participant engagement and therapeutic benefit in real time. Results between groups will be evaluated for efficacy. The potential reach of both treatment options will be evaluated to determine how well each approach scales to weigh the potential therapeutic impact for the target population.

Results

While participant trials are pending, initial system testing and simulation studies demonstrate that iPAL MIND can respond meaningfully to biosignal changes in stress and attentional states. The system successfully modifies verbal prompts and pacing based on real-time signal fluctuations, confirming its technical feasibility and responsiveness. Data collection will include craving assessments, neurophysiological changes, engagement metrics and saliva cotinine tests to evaluate nicotine exposure across multiple sessions. We hypothesize that human-lead interventions will perform better than MetaHuman lead sessions, and MetaHuman-lead interventions will provide effective results that scale better, thus, providing a larger net benefit through increased reach to at-risk populations.

Conclusion

IPAL MIND represents a novel approach to behavioral health intervention by integrating real-time biometric sensing, immersive VR environments, and artificial intelligence. This approach allows for scalable, personalized mindfulness training that adapts in the moment to individual user needs. Particularly in the context of novel nicotine products like ONPs, such interventions may fill a critical gap in prevention and cessation efforts.

Future work includes formal testing in randomized trials, integration with mobile support tools, and expansion to other populations and substance use behaviors. By tailoring interventions to the psychophysiological state of the individual, iPAL MIND exemplifies the potential of intelligent systems to support behavior change through responsive, human-centered technology.

Acknowledgment

This work is supported by the National Science Foundation and National Institutes of Health Award #2013651.

Presenting Author: Jeff Simpson Montana State University

Presenting Author Biography: Jeff Simpson is a postbaccalaureate computer science major at Montana State University researching technology-enabled behavioral health interventions. As a student researcher in the Human Interaction Lab, his work focuses on extended reality (XR) systems—including VR, AR, and MR—to support behavioral change therapies for substance use disorders. He integrates AI-powered MetaHumans, biosensing tools (e.g., fNIRS, heart rate monitors), and large language models (LLMs) to deliver adaptive, personalized experiences. Jeff’s research spans human–computer interaction (HCI), user experience design, and trust in human–robot interaction.

Authors:

Jeff Simpson Montana State University
Nicholas Addotey Montana State University
Bridger Foran Montana State University
Laura Stanley Montana State University