Efficacy of Interactive Audio-based Mindfulness ChatGPT-Powered Intervention on Well-being

Brief mindfulness intervention is a short form is a form of short-duration practice, typically delivered for around 30 minutes or less (Howarth et al., 2019). It can be conducted in various formats, including a short breathing exercise or body scan (Palascha et al., 2021), and are increasingly offered through digital platforms, including app-based formats.

Empirical evidence has examined the efficacy of brief mindfulness intervention and yielded mixed results. For some studies, positive effects were found, while some studies found no supporting evidence. Such inconsistencies may be due to the several current limitations of brief mindfulness intervention. Firstly, the duration of the intervention is too short. Given this brevity, these short practices may not offer the same immersion and depth that traditional mindfulness may offer. Secondly, most brief mindfulness utilised audio-based formats, which tend to be passive in nature and may not fully engage participants. Lastly, for brief mindfulness delivered through app-based platforms, these applications adopt a self-help approach, which may lower engagement and user commitment. Moreover, these applications also relied on audio guides which lacks the student-facilitator engagement found in traditional classes.

To address the current limitations within the literature, a between-within-subjects experiment will be conducted where participants will be randomly assigned to one of the three conditions: (1) an interactive brief mindfulness breathing ChatGPT-powered intervention, (2) an interactive mind-wandering ChatGPT-powered control, or (3) a mindfulness breathing audio control. We hypothesise that the brief mindfulness breathing ChatGPT-powered intervention group will have significant improvements in well-being outcomes as compared to the other two control groups. By exploring the potential of an interactive ChatGPT-powered intervention, this study aims to understand its efficacy in improving well-being.

Study Overview

• Status: Recruiting
• Conditions: Perceived Stress; State Anxiety; Positive and Negative Affect; Self Control; State Mindfulness Attention
• Intervention / Treatment: Other: Interactive Mind Wandering ChatGPT-Powered Intervention; Other: Interactive Mindfulness Breathing ChatGPT-Powered Intervention; Other: Mindfulness Breathing Audio-only

• Study Type: Interventional
• Enrollment (Estimated): 210
• Phase: Not Applicable

Contacts and Locations

• Study Contact: Name: Xun Ci Soh; Phone Number: +65 83994980; Email: xunci.soh.2024@msps.smu.edu.sg

Study Locations

• Singapore
  • Singapore, Singapore
    • Recruiting
    • Singapore Management University
    • Contact: Xun Ci Soh; Phone Number: +65 83994980; Email: xunci.soh.2024@msps.smu.edu.sg

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria:

• Participants who are aged 18 to 30 (inclusive)

Exclusion Criteria:

• Not applicable

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Basic Science
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

3

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Active Comparator: Interactive Mind Wandering (ChatGPT); The intervention is a ChatGPT-powered mind wandering bot configured using a 15-minute guided exercise adapted from Hafenbrack et al. (2014), Clinton et al. (2018), and Arch and Craske (2006). The session is divided into one-minute segments: the first minute provides settling instructions (for example, inviting participants to adopt a comfortable position), followed by step-by-step guidance of the mindfulness breathing exercise. To foster reflection and encourage active engagement, the chatbot was trained to interact with the participants using interactive prompts. This includes asking questions, such as "When you are ready, please share...What feeling or memory stood out most clearly as your mind wandered?", "When you are ready, please share...where did your attention drift... to just now?".	Other: Interactive Mind Wandering ChatGPT-Powered Intervention; The intervention is a ChatGPT-powered mind wandering bot configured using a 15-minute guided exercise adapted from Hafenbrack et al. (2014), Clinton et al. (2018), and Arch and Craske (2006). The session is divided into one-minute segments: the first minute provides settling instructions (for example, inviting participants to adopt a comfortable position), followed by step-by-step guidance of the mindfulness breathing exercise. To foster reflection and encourage active engagement, the chatbot was trained to interact with the participants using interactive prompts. This includes asking questions, such as "When you are ready, please share...What feeling or memory stood out most clearly as your mind wandered?", "When you are ready, please share...where did your attention drift... to just now?".
Experimental: Interactive Mindfulness Breathing (ChatGPT); The intervention is a ChatGPT-powered mindfulness breathing bot configured using a 15-minute guided exercise adapted from Hafenbrack et al. (2014), Clinton et al. (2018), and Arch and Craske (2006). The session is divided into one-minute segments: the first minute provides settling instructions (for example, inviting participants to adopt a comfortable position), followed by step-by-step guidance of the mindfulness breathing exercise. To foster reflection and encourage active engagement, the chatbot was trained to interact with the participants using interactive prompts. This includes asking questions, such as " Share what this present moments feel like for you", "What sensations do you notice in your abdomen as you place your hand there?", "Share what stood out to you as you noticed your thoughts, feelings, or sensations just now.".	Other: Interactive Mindfulness Breathing ChatGPT-Powered Intervention; The intervention is a ChatGPT-powered mindfulness breathing bot configured using a 15-minute guided exercise adapted from Hafenbrack et al. (2014), Clinton et al. (2018), and Arch and Craske (2006). The session is divided into one-minute segments: the first minute provides settling instructions (for example, inviting participants to adopt a comfortable position), followed by step-by-step guidance of the mindfulness breathing exercise. To foster reflection and encourage active engagement, the chatbot was trained to interact with the participants using interactive prompts. This includes asking questions, such as " Share what this present moments feel like for you", "What sensations do you notice in your abdomen as you place your hand there?", "Share what stood out to you as you noticed your thoughts, feelings, or sensations just now.".
Active Comparator: Mindfulness Breathing Audio; This intervention consists of listening to a 15-minute mindfulness breathing audio-only exercise, adapted from Hafenbrack et al. (2014), Clinton et al. (2018), and Arch and Craske (2006). There will be no verbal interactions in this intervention.	Other: Mindfulness Breathing Audio-only; This intervention consists of listening to a 15-minute mindfulness breathing audio-only exercise, adapted from Hafenbrack et al. (2014), Clinton et al. (2018), and Arch and Craske (2006). There will be no verbal interactions in this intervention.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Stress	Participants' current stress levels will be measured with a single item ("How stressed do you feel right now?"), rated on an 11-point scale (0 = No stress, 10 = Extreme stress).	Change from pre-intervention (baseline, before intervention) to immediately post-intervention (after the 15-minute session)
Positive and Negative Affect	Participants' affective states will be measured using the 18-item Circumplex Model of Affect Scale (Russell, 1980), which assesses emotional states across two separate dimensions: positive affect (PA) and negative affect (NA). Participants are required to indicate the extent to which they currently experience each emotion on a 5-point Likert scale (1 = Not at all, 5 = Extremely), in response to the question, "Overall, how do you feel right now?".	Change from pre-intervention (baseline) to immediately post-intervention (after the 15-minute session)
State Anxiety	Participants' state anxiety levels will be measured with the 6-item Spielberger State-Trait Anxiety Inventory (STAI; Marteau & Bekker, 1992), rated on a 4-point Likert scale (1 = Not at all, 4 = Very much). Participants are required to rate how they generally feel in response to statements presented (e.g., "I feel calm", "I am tense", "I feel upset").	Change from pre-intervention (baseline) to immediately post-intervention (after the 15-minute session).
Self-control Capacity	Participants' self-control capacity will be measured with the 5-item Brief State Self-Control Capacity Scale (SMS-5; Ciarocco et al., 2009; Linder et al., 2015), rated on a 7-point Likert scale (1 = Very untrue of me, 7 = Very true of me). Participants are required to rate how they feel in response to the statements presented (e.g., "I feel drained now", "I feel lazy now", "I feel like my willpower is gone now").	Change from pre-intervention (baseline) to immediately post-intervention (after the 15-minute session).
State Mindfulness Attention	Mindful Attention State. Participants' state mindfulness attention levels will be measured with the 21-item State Mindfulness Scale (SMS; Tanya & Bernstein, 2013; Ruimi et al., 2022), on a 5-point Likert scale (1 = Not at all, 5 = Very well). Participants are required to rate how they feel in response to the statements presented (e.g., "I was aware of different emotions that arose in me", "I noticed many small details of my experience", "I felt that I was experiencing the present moment fully").	Change from pre-intervention (baseline) to immediately post-intervention (after the 15-minute session).

Collaborators and Investigators

• Sponsor: Singapore Management University
• Collaborators: Nanyang Technological University

Study record dates

Study Major Dates

• Study Start (Actual): October 30, 2025
• Primary Completion (Estimated): May 1, 2026
• Study Completion (Estimated): May 1, 2026

Study Registration Dates

• First Submitted: December 3, 2025
• First Submitted That Met QC Criteria: December 30, 2025
• First Posted (Estimated): January 12, 2026

Study Record Updates

• Last Update Posted (Estimated): January 12, 2026
• Last Update Submitted That Met QC Criteria: December 30, 2025
• Last Verified: October 1, 2025

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Mental Disorders; Behavior; Social Behavior; Anxiety Disorders; Self-Control
• Other Study ID Numbers: IRB-25-212-A172-M2(1125)

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No