ChatWise: AI-Powered Engaging Conversations for Enhancing Senior Cognitive Wellbeing

2025-02-19

arXiv.org

AI Summary

To address critical shortages in caregiving resources and the limited strategic guidance capability of existing AI chatbots—particularly their heavy reliance on large-scale annotated datasets—this study proposes a two-tiered reasoning dialogue system tailored for older adults with mild cognitive impairment (MCI). Built upon large language models (LLMs), the system integrates cognitive science–informed dialogue policies, fine-grained affective sensing, and multi-turn consistency modeling to transcend conventional LLM limitations in short-horizon interaction, enabling extended, emotion-adaptive, and cognitively stimulating multi-turn dialogues. Its core innovation is a novel, geriatric-optimized two-tiered reasoning framework that achieves deep, personalized interaction with minimal annotation cost. Empirical evaluation demonstrates statistically significant improvements in simulated older users’ attention, working memory, and affective states; intervention effects are especially pronounced for individuals with MCI.

Technology Category

• Cognitive Modeling & Cognitive Systems: Affective Computing
• Natural Language Processing: Conversational AI/Dialog Systems
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Cognitive health in older adults presents a growing challenge. While conversational interventions show feasibility in improving cognitive wellness, human caregiver resources remain overburdened. AI-based methods have shown promise in providing conversational support, yet existing work is limited to implicit strategy while lacking multi-turn support tailored to seniors. We improve prior art with an LLM-driven chatbot named ChatWise for older adults. It follows dual-level conversation reasoning at the inference phase to provide engaging companionship. ChatWise thrives in long-turn conversations, in contrast to conventional LLMs that primarily excel in short-turn exchanges. Grounded experiments show that ChatWise significantly enhances simulated users' cognitive and emotional status, including those with Mild Cognitive Impairment.

Problem

• Enhancing cognitive support for older adults using AI chatbots
• Reducing reliance on overloaded human caregiver resources
• Improving strategy-guided multi-turn dialogue generation

Innovation

• Strategy-guided AI chatbot for cognitive support
• Dual-level conversation reasoning framework
• Macro-level strategy and micro-level utterance integration