Factors Influencing Conversational Engagement in Robot-Delivered Individual Cognitive Stimulation Therapy (iCST) for Dementia in Home Settings

Jul 16, 2026

Social robots can support dementia care by guiding structured conversation and therapeutic activities, yet the conversational dynamics of robot-delivered cognitive stimulation therapy (CST) remain insufficiently studied. This work analyzed interaction patterns from robot-delivered individual CST (iCST) sessions conducted at home. Using the Co-STAR system deployed for one week with eight people with dementia, conversational metrics measured engagement. Results show personalised prompts improve engagement, conversational behaviour shifts across phases, and early metrics relate to long-term participation.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages7

Reading Time18 min

Introduction

Study focus and contributions

Hypotheses and research questions

Factors Influencing Conversational Engagement in Robot-Delivered Individual Cognitive Stimulation Therapy (iCST) for Dementia in Home Settings Emmanuel Akinrintoyo and Nicole Salomons arXiv :2607 .07998v 1 [ cs .RO] 9 Jul 2026 Abstract—Social robots offer a promising means of supporting cognitive therapies for dementia care by guiding structured conversation and therapeutic activities. However, little is known about the conversational dynamics that emerge during robotdelivered cognitive stimulation therapy (CST) sessions. This study analysed the interaction patterns from robot-delivered individual CST (iCST) sessions conducted with people living with dementia in home settings. Our Co-STAR (Cognitive Stimulation Therapy by an Autonomous Robot) system was deployed in the homes of eight PwDs for one week, who completed 30-minute sessions. Conversational metrics, including words per turn, speech production rate, response duration, response latency, and self-referential language, were analysed to examine how conversational engagement is shaped by prompt personalisation, interaction phase, and participant characteristics. The findings highlight three key interactional properties of robotdelivered iCST. First, personalised prompts significantly increase response duration, self-referential language, and overall engagement compared to generic prompts. Second, conversational behaviour changes within sessions, with a reduction in the verbal output and autobiographical engagement observed during later interaction phases, which suggests cognitive fatigue. Third, first-session conversational metrics were associated with long-term participation, while living situation influenced conversational engagement patterns. These findings provide empirical insights into the factors that shape conversational engagement in robot-delivered iCST. They inform the design of adaptive conversational robots for dementia therapy. I. INTRODUCTION Dementia is a neurocognitive syndrome that manifestsas a progressive decline in cognitive and communicative abilities [1] . Persons living with dementia (PwDs) experience conversational difficulties. These include word-finding problems, reduced initiation, breakdowns in turn-taking, and greater disfluencies, such as filler words [2], [3], [4] . Conversations, such as in therapy sessions, support memory activation, language production, social connectedness and identity expression for PwDs [5], [4], [1], [6], [7] . CST (Cognitive Stimulation Therapy) is an example of such therapies designed for fostering conversational participation with repeated practice for PwDs [5] . CST is one of the leading evidence-based non-pharmacological interventions for individuals living with mild to moderate dementia [5],[8] . It has demonstrated benefits for cognition, communication, and quality of life [5], [8] . However, delivering CST consistently at scale is resource-intensive. It often relies on trained facilitators and structured group settings. Emmanuel Akinrintoyo and Nicole Salomons are with Imperial College London , [e.akinrintoyo23@imperial.ac.uk](e.akinrintoyo23@imperial.ac.uk) , [n.salomons@imperial.ac.uk](n.salomons@imperial.ac.uk) Fig. 1: A participant completing an individual cognitive stimulation therapy session in their living room with CoSTAR [11] . Recent advances in socially assistive robots (SARs) offer new opportunities to augment or support therapeutic interventions for PwDs [9], [10], [11] . SARs are being explored as facilitators of cognitive activities, companions, and communication partners. Their potential lies in their ability to deliver therapeutic tasks, scaffold interaction and maintain conversational engagement. Hence, SARs are uniquely positioned to deliver individual CST (iCST) [12], [13] because they combine embodied presence, conversational interaction, and consistent facilitation. However, limited research has examined the fine-grained conversational dynamics that emerge during robot

FAQs

What does the study analyze about robot-delivered iCST sessions?

The study performs interaction analysis of audio-recorded iCST therapy sessions and examines how conversational engagement varies with prompt personalisation, interaction phase, and participant characteristics.

How do personalised prompts affect conversational engagement compared with generic prompts?

Personalised prompts significantly increase response duration, self-referential language, and overall engagement relative to generic prompts.

What factors are linked to long-term participation and engagement patterns?

First-session conversational metrics were associated with long-term participation, and living situation influenced the patterns of conversational engagement.

Jul 16, 2026

Upload by Genevieve