Adaptive Narrative Interventions to Foster Psychosocial Wellbeing

Psychosocial wellbeing involves navigating interconnected mental, emotional, and social experiences across daily life. People manage negative thoughts, pursue goals, maintain relationships, and respond to responsibilities that shift over time. Digital tools such as text messaging systems and mobile apps offer accessible ways to support psychosocial wellbeing, yet many rely on static inputs or prescriptive suggestions that do not account for changing circumstances. When support fails to match current needs, it can feel superficial or misaligned, discouraging continued use. This dissertation examines a narrative-centered approach to address these limitations by inviting users to reflect on the context and meaning of their experiences. Narratives can help people interpret difficulties, identify recurring themes, and connect with shared experiences. By fostering reflection and personal meaning-making, such interventions can adapt to evolving needs and enhance the sense of being understood. The work investigates how narrative interventions can be designed and delivered in accessible formats such as text messaging and mobile apps. Studies using surveys, interviews, and deployments identify qualities of short stories for communicating psychological principles and practical considerations for length and timing. A large-scale evaluation assesses how large language models (LLMs) can tailor stories to emotional and situational contexts, showing that personalization can reduce belief in negative thoughts while preserving key narrative qualities. Additional studies explore structured activities that guide users in creating personal narratives about distressing experiences through brief prompts on mobile and web platforms. These activities demonstrate that reflection can occur with minimal effort while revealing trade-offs between timing and depth. Interactive LLM-powered tools are examined for their ability to help users revise and interpret their reflections, with feedback emphasizing flexibility and alignment with daily routines. Finally, the dissertation examines how social contexts such as responsibilities, obligations, and health-related challenges affect engagement with digital interventions and offers strategies for designing support that adapts to these circumstances. In summary, this dissertation advances a model of narrative-centered intervention that aligns content with lived experiences, adapts to social contexts, and enables collaboration between people and technology to support psychosocial wellbeing in flexible, responsive, and human-centered ways.

Department dc:contributor.department

Computer Science

Year dc:date.issued

2025

Author dc:creator

• Bhattacharjee, Ananya

Advisor dc:contributor.advisor

• Williams, Joseph Jay

• Adaptive Interventions
• Generative AI
• Human-AI Interaction
• Mental Health
• Reflection
• Wellbeing

Statement dc:rights

• Attribution 4.0 International

Licence dc:rights.uri

• http://creativecommons.org/licenses/by/4.0/

Handle dc:identifier.uri

https://hdl.handle.net/1807/150095

OAI identifier oai:identifier

oai:utoronto.scholaris.ca:1807/150095