Personalized intelligent dialogue robot based on AIGC assists memoir writing for elderly people with cognitive impairment: a multi-method collaborative design framework (Preprint)

BACKGROUND Elderly individuals with cognitive impairment often face significant challenges in memoir writing, including memory fragmentation, emotional loneliness, and language expression barriers. Although AI-generated content (AIGC) technologies such as GPT-3.5 show potential in content creation, they often lack the personalization and adaptability required for users with dementia. OBJECTIVE This study aimed to design and evaluate a personalized AIGC-powered conversational robot to assist elderly people with cognitive impairment in memoir writing and emotional support. METHODS We developed a multi-method collaborative design framework integrating Kansei Engineering (KE), Quality Function Deployment (QFD), Axiomatic Design (AD), and the TOPSIS decision model. The system dynamically adapts interaction strategies based on users’ Mini-Mental State Examination (MMSE) scores. Evaluation included usability testing via the System Usability Scale (SUS), emotional congruence assessment, and memoir quality analysis across three time points with 20 participants. RESULTS The system achieved high usability (SUS score > 70) and significantly improved emotional congruence and memoir quality in the experimental group compared to controls. Participants with severe cognitive impairment (MMSE ≤ 18) benefited more from AI-driven narrative generation, while those with mild impairment (MMSE ≈ 22) preferred keyword-based prompts. CONCLUSIONS The personalized AIGC-based conversational robot effectively supports memoir writing and emotional well-being in elderly individuals with cognitive impairment. Its adaptive, multimodal design promotes meaningful human-AI collaboration and shows potential for enhancing cognitive rehabilitation and eldercare services.