A multi-agent system called MindfulAgents uses large language models to generate guided meditation scripts, encourage reflection on emotional states, and personalize meditation in real time. In a lab study with 13 participants, it improved in-session engagement, self-awareness, and reduced momentary stress. A four-week deployment with 62 participants showed increased long-term engagement and level of mindfulness. Participants reported that the system offered more relevant meditation sessions tailored to individual needs, supporting sustained practice. The findings highlight the potential of LLM-driven personalization for digital mindfulness interventions.
Respiration biosignal feedback from a smartphone's built-in accelerometer can improve the usability of mindfulness apps and help track skill development. A respiration tracking algorithm, tested on 261 meditation sessions in controlled and real-world settings, accurately captures slow breathing patterns typical of mindfulness, achieving a mean error of 1.6 breaths per minute. A novel framework estimates three mindfulness skills—concentration, sensory clarity, and equanimity—from respiration data, with F1 scores of 80-84% for tracking skill progression. A user study comparing an experimental group receiving biosignal feedback with a control group using a standard app shows that respiration feedback enhances system usability, suggesting smartphone sensors can enhance digital mindfulness training without additional wearables.