Exploring Generative AI for Personalized Engagement in Therapy Adherence Contexts

Authors

  • Mirella M. Moser ZHAW School of Management and Law
  • Nina Lauria ZHAW School of Management and Law
  • Thomas Keller ZHAW School of Management and Law https://orcid.org/0000-0002-8100-6900

DOI:

https://doi.org/10.34190/icair.5.1.4211

Keywords:

Generative AI, Therapy Adherence, Personalization, Virtual Companion, Tamagotchi Effect

Abstract

This study explores the use of generative AI to create personalized virtual companions, referred to as "Tamagotchis," aimed at fostering emotional engagement and supporting therapy adherence. A prototype system was developed that generates individualized 2D image sequences representing symbolic objects or characters in different states, such as weakened, neutral, or improved. Through a mixed-methods evaluation involving a quantitative survey (n=78) and qualitative interviews (n=10), the study assessed the visual clarity, emotional resonance, and perceived relevance of the generated images. Results indicate that personalized objects were more emotionally engaging than predefined versions, and user involvement in the creation process significantly enhanced identification and motivation. These findings suggest that emotionally adaptive virtual companions may hold promise as a supportive element in long-term therapeutic interventions.

Author Biographies

Mirella M. Moser, ZHAW School of Management and Law

Research Associate at the Institute of Business Information Systems at ZHAW School of Management and Law

 

 

Nina Lauria, ZHAW School of Management and Law

Student at the Institute of Business Information Systems at ZHAW School of Management and Law

Thomas Keller, ZHAW School of Management and Law

Professor at the Institute of Business Information Systems at ZHAW School of Management and Law

 

 

Downloads

Published

2025-12-04