Enhancing Cognitive Functions of Older Adults With Software Robot: Longitudinal Exploratory Field Study

The global prevalence of dementia continues to rise and demands scalable, nonpharmacological interventions. Digital cognitive training has expanded, but many older adults, who have limited digital literacy, struggle to sustain use. We designed an intervention that integrates social interaction, reward-based engagement, and an artificial intelligence (AI) conversational agent, which aims to reduce these barriers and support continuous participation.

This study examined whether a 12-week digital cognitive training program improves cognitive function in older adults. It also tested whether a group chat service, which enables interaction among participants and with an AI agent, increases engagement and social support.

We recruited 133 participants (mean age 64.75, SD 6.76; range 55-75 years) who had no diagnosis of dementia. All participants used the Care & Cure program for 12 weeks. The program includes an AI chatbot (Sammy Talk) and a group chat service (Our Town), which supports peer interaction. We measured cognitive function using the Korean Mini-Mental State Examination-Version 2 (K-MMSE-2). We also assessed degrees of social support (Medical Outcomes Study-Social Support Survey), depression (Short Form Geriatric Depression Scale-Korean Version), and engagement (Twente Engagement with eHealth Technologies Scale), and analyzed usage logs to examine participation patterns.

Participants showed improved cognitive function after the intervention (Hedges g=0.350, P<.001). Active users (n=67), who engaged more frequently with the program, showed greater improvement than nonactive users (n=66), especially among those who had lower baseline cognitive scores (Hedges g=0.523, P<.001). Social support increased, particularly emotional and informational support (t132=-6.509, P<.001). Participants reported higher engagement (t132=2.008, P<.05) and lower depression scores (t132=3.093, P<.01). Regression results showed that group chat participation, which promotes interaction with the AI agent, increased engagement in cognitive training (t131=12.395, P<.001). Increased engagement was associated with higher social support (t131=4.102, P<.001) and improved cognitive function (t131=2.467, P<.05). Cognitive training alone did not produce a significant effect. Participants showed low adherence, which indicates a need for strategies that sustain long-term use.

The Care & Cure program improved cognitive function and strengthened social support in older adults. Social interaction, which increases engagement, played a central role. These findings suggest that digital cognitive interventions should incorporate social mechanisms to achieve meaningful effects.