Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Designing KRIYA: An AI Companion for Wellbeing Self-Reflection

Shanshan Zhu, Wenxuan Song, Jiayue Melissa Shi, Dong Whi Yoo, Karthik S. Bhat, Koustuv Saha

TL;DR

The paper addresses the problem that wellbeing apps focus on dashboards and targets, often inducing performance pressure rather than meaningful interpretation. It presents KRIYA, an AI companion designed to enable co-interpretation of personal health data through Morning Forecast, Evening Debrief, What-If Planning, and a conversational chat space, tested with 18 college students using hypothetical data. Findings show that co-interpretive dialogue reframes reflection as curiosity-driven sensemaking, compassionate framing reduces judgment, and transparency about uncertainty builds trust, with planning kept as low-stakes exploration. The work expands the design space for digital wellbeing by proposing AI companions that share interpretive labor, support contextual storytelling, and adapt to emotional and cognitive needs, offering practical implications for future deployments and longitudinal studies.

Abstract

Most personal wellbeing apps present summative dashboards of health and physical activity metrics, yet many users struggle to translate this information into meaningful understanding. These apps commonly support engagement through goals, reminders, and structured targets, which can reinforce comparison, judgment, and performance anxiety. To explore a complementary approach that prioritizes self-reflection, we design KRIYA, an AI wellbeing companion that supports co-interpretive engagement with personal wellbeing data. KRIYA aims to collaborate with users to explore questions, explanations, and future scenarios through features such as Comfort Zone, Detective Mode, and What-If Planning. We conducted semi-structured interviews with 18 college students interacting with a KRIYA prototype using hypothetical data. Our findings show that through KRIYA interaction, users framed engaging with wellbeing data as interpretation rather than performance, experienced reflection as supportive or pressuring depending on emotional framing, and developed trust through transparency. We discuss design implications for AI companions that support curiosity, self-compassion, and reflective sensemaking of personal health data.

Designing KRIYA: An AI Companion for Wellbeing Self-Reflection

TL;DR

The paper addresses the problem that wellbeing apps focus on dashboards and targets, often inducing performance pressure rather than meaningful interpretation. It presents KRIYA, an AI companion designed to enable co-interpretation of personal health data through Morning Forecast, Evening Debrief, What-If Planning, and a conversational chat space, tested with 18 college students using hypothetical data. Findings show that co-interpretive dialogue reframes reflection as curiosity-driven sensemaking, compassionate framing reduces judgment, and transparency about uncertainty builds trust, with planning kept as low-stakes exploration. The work expands the design space for digital wellbeing by proposing AI companions that share interpretive labor, support contextual storytelling, and adapt to emotional and cognitive needs, offering practical implications for future deployments and longitudinal studies.

Abstract

Most personal wellbeing apps present summative dashboards of health and physical activity metrics, yet many users struggle to translate this information into meaningful understanding. These apps commonly support engagement through goals, reminders, and structured targets, which can reinforce comparison, judgment, and performance anxiety. To explore a complementary approach that prioritizes self-reflection, we design KRIYA, an AI wellbeing companion that supports co-interpretive engagement with personal wellbeing data. KRIYA aims to collaborate with users to explore questions, explanations, and future scenarios through features such as Comfort Zone, Detective Mode, and What-If Planning. We conducted semi-structured interviews with 18 college students interacting with a KRIYA prototype using hypothetical data. Our findings show that through KRIYA interaction, users framed engaging with wellbeing data as interpretation rather than performance, experienced reflection as supportive or pressuring depending on emotional framing, and developed trust through transparency. We discuss design implications for AI companions that support curiosity, self-compassion, and reflective sensemaking of personal health data.
Paper Structure (34 sections, 5 figures, 2 tables)

This paper contains 34 sections, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Personal Health Informatics and Self-Reflection
  4. Emotional Attunement and Compassionate Design in Digital Wellbeing
  5. Human-centered AI and Conversational Agents for Wellbeing
  6. Study Design and Methods
  7. Designing a Prototype AI Companion: KRIYA
  8. Participants and Recruitment
  9. Study Procedure
  10. Segment A: Baseline Usage and Mental Models.
  11. Segment B: Checking Current Health App
  12. Segment C: KRIYA Walkthrough and Post-Interaction Reflection
  13. Data Analysis
  14. Privacy, Ethics, and Positionality
  15. Results
  16. ...and 19 more sections

Figures (5)

  • Figure 1: Interaction workflow of KRIYA (an AI companion prototype) during the interview study. The participants engaged with hypothetical data through a sequence of interactions, including 1) Morning Forecast (contextualized overview of anticipated patterns for the day), 2) Comfort Zone selection (participants set more realistic boundaries for the day based on schedule and comfort), 3) conversational reflection (dialogue-based sensemaking around recent data patterns), 4) Evening Debrief (retrospective reflection on the day's experiences), 5) What-If Planning (exploration of hypothetical changes and future scenarios), and 6) final reflections (participants' meta-level evaluation of the experience in the session).
  • Figure 2: Screenshots of the AI companion prototype, KRIYA, for the (a) dashboard and (b) What-If Planning for speculative, low-stakes exploration of future scenarios.
  • Figure 3: Distribution plots based on the exit surveys (the dotted lines represent the median values).
  • Figure A1: Screenshots of the KRIYA. Modules include: (a) Morning Forecast where users set a Comfort Zone (a realistic range rather than a single goal) (b) a conversational interface for asking questions about steps and sleep.
  • Figure A2: Screenshots of the KRIYA prototype in Evening Debrief and Detective Mode that support explanation-oriented reflection.