RemiHaven: Integrating "In-Town" and "Out-of-Town" Peers to Provide Personalized Reminiscence Support for Older Drifters

TL;DR

RemiHaven addresses loneliness among older drifters in China by delivering a personalized reminiscence tool powered by Multimodal Large Language Models. It introduces two simulated peer modes, In-Town and Out-of-Town, to tailor conversations and generate multimodal memory materials that culminate in a shareable life storybook. Through a two-phase formative study and a subsequent evaluation with 10 participants, the approach demonstrates improved reminiscence experiences and positive emotional effects, while revealing challenges in content accuracy, topic control, and usability. The work offers design guidelines for AI-enhanced reminiscence interventions and highlights the potential of generative AI to support aging populations, with attention to cultural context and long-term impact.

Abstract

With increasing social mobility and an aging society, more older adults in China are migrating to new cities, known as "older drifters." Due to fewer social connections and cultural adaptation challenges, they face negative emotions such as loneliness and depression. While reminiscence-based interventions have been used to improve older adults' psychological well-being, challenges such as the lack of tangible materials and limited social resources constrain the feasibility of traditional reminiscence approaches for older drifters. To address this challenge, we designed RemiHaven, a personalized reminiscence support tool based on a two-phase formative study. It integrates "In-Town" and "Out-of-Town" peer agents to enhance personalization, engagement, and emotional resonance in the reminiscence process, powered by Multimodal Large Language Models (MLLMs). Our evaluations show RemiHaven's strengths in supporting reminiscence while identifying potential challenges. We conclude by offering insights for the future design of reminiscence support tools for older migrants.

Figures (11)

• Figure 1: The activities conducted during storyboard-based user studies.
• Figure 2: An example storyboard illustrating reminiscence scenarios tailored for older drifters.
• Figure 3: The architecture of RemiHaven.
• Figure 4: The RemiHaven system interfaces: (1) Users start by selecting reminiscence modes or accessing their Life Storybook. To avoid ambiguity and make it easier for users to understand, the two modes Conversation with "In-town" Peers and Conversation with "Out-of-Town" Peers are represented in the interface as "Guided Reminiscence" and "Free Reminiscence", respectively. Before system use, we comprehensively introduced the meanings and difference of these two concepts to older drifters. (2) The conversational interface, where the system engages users with questions to help them recall specific memories. (3-A) The initial image generated based on the user's recollections, (3-B) the image modification tool allowing users to redraw or adjust specific parts of the image, and (3-C) the finalized image ready for inclusion in the life storybook. (4) The system-generated textual description of the user's memory, which can be reviewed and edited. (5) Finally, users can export their memories as a PDF Life Storybook for saving or printing. The interface shown is translated from Chinese for clarity.
• Figure 5: The evaluation procedure.