Aug 2026· Applied Sciences· 0 citations· 85 references
TL;DR
HEART, a healthcare-specific evaluative architecture comprising Human-Centred Communication, Ethical and Trustworthy Deployment, Ethical and Trustworthy Deployment, Adaptive and Embodied Intelligence, Relationship Continuity, and Translational Healthcare Value is proposed.
Abstract
Large language models (LLMs) are expanding the capabilities of socially assistive robots (SARs) through natural dialogue, personalisation, multimodal reasoning, retained interaction context, and adaptive behaviour in healthcare. Integrating generative language models into robots, however, complicates evaluation because fluent output may exaggerate perceived competence and increase the risks of hallucination, overtrust, privacy exposure, relationship dependency, and unsafe reliance on advice or actions. This PRISMA-informed review synthesises healthcare robotics, human–robot interaction, LLM-enabled systems, ethics, implementation, and care delivery. Database searches returned 128 records, of which 110 were unique after deduplication. Supplementary retrieval and assessment yielded 85 substantive sources spanning background mapping, primary analysis, and governance. Studies focused mainly on feasibility, usability, acceptability, dialogue quality, and short-term engagement, whereas longitudinal safety, governance of retained interaction context, comparative effectiveness, workflow integration, and sustained healthcare value received limited attention. These gaps indicate that evaluation of LLM-enabled SARs must account for physical presence, social role, interaction memory, and potential actions rather than focus on conversational performance alone. The review therefore proposes HEART, a healthcare-specific evaluative architecture comprising Human-Centred Communication, Ethical and Trustworthy Deployment, Adaptive and Embodied Intelligence, Relationship Continuity, and Translational Healthcare Value. HEART uses boundary rules, operational indicators, qualitative labels, and non-additive deployment gates to separate evaluative domains, define assessable outcomes, summarise reported support, and prevent strengths in one area from masking critical safety or governance failures. Future research should validate HEART through longitudinal and comparative assessment of hallucination severity, language-to-action safety, long-term effects, equity, and post-deployment monitoring.
A LINE-deployed conversational AI system grounded in the WHO iSupport for Dementia framework, built around an Assessment-Before-Intervention dialogue mechanism, which offers practical implications for the design of AI-assisted care systems in dementia and other emotionally sensitive healthcare contexts.
Abstract Population aging, chronic illness, workforce shortages, and rapid adoption of artificial intelligence (AI), robotics, virtual nursing, remote monitoring, and digital platforms are reshaping fundamental care. Yet existing literature focuses mainly on technical performance, implementation, and ethical risk, with limited attention to how technology alters the relational conditions of person-centered care. This theory-informed narrative review synthesized multidisciplinary literature identified through PubMed, CINAHL, Scopus, and Web of Science, supplemented by reference-list screening and purposive inclusion of seminal works. Seventy-eight empirical, review, conceptual, ethical, policy, and methodological sources informed the synthesis. The review reconceptualizes person-centered fundamental care through posthumanization in nursing—a socio-technical condition in which care practices, judgment, and relationships are increasingly co-produced by human and technological actors—and proposes relational augmentation as an evaluative framework. The synthesis indicates that task substitution is inadequate because it privileges efficiency, standardization, and replacement over interpretation, dignity, and relationship. Relational augmentation instead asks whether technologies strengthen relationship support, integrated recognition of need, and a supportive care context. Across monitoring and prediction, documentation and workflow, communication and education, social robotics, and virtual nursing, the framework identifies both functional benefits and relational risks. It provides a practical basis for evaluating whether technological innovation preserves trust, dignity, equity, contextual understanding, and professional judgment in person-centered fundamental care.
Fanghong Nie, Cheng Xu, Yijun Huang et al.· Journal of Multidisciplinary...· 0 citations
The rise of LLMs has enabled CUIs to increasingly mimic human social and conversational cues, e.g., tone of voice and emotional expressions. However, this mimicry usually lacks strategic communicative intent, placing the cognitive burden of mutual understanding on the user. At the same time, CUIs based on general-purpose, task-agnostic LLMs are being deployed across varied domains with distinct, context-specific conversational needs, including, e.g., healthcare, education, and journalism. Therefore, there is a growing need to transition from arbitrary, domain-agnostic generation of pragmatic cues to strategic adaptation of both visual and linguistic interface features. This workshop proposes a paradigm shift toward designing context-specific CUIs that actively support communicative success through pragmatic cues. Bringing together perspectives from HCI and social sciences, we will explore how users appropriate conversational AI across domains. Through cross-disciplinary dialogue, the workshop aims to establish a shared vocabulary, identify domain-specific challenges, and lay the groundwork for future collaboration.
Laura Spillner, Johanna Rockstroh, Paul Goerke et al.· International Conference on...· 0 citations
Background The global rise in dementia, closely linked to aging populations, necessitates innovative interventions to address care challenges. Artificial Intelligence-Enabled Social Robots (AI-ESRs) present a promising approach by leveraging multi-modal communication to enhance social engagement and alleviate isolation. However, evidence regarding their efficacy remains inconsistent and fragmented. This systematic review evaluates the effectiveness of AI-ESRs in enhancing social interactions among people with dementia, examines factors contributing to outcome variability, and provides recommendations for optimizing their integration into care models. Methodology A comprehensive search was conducted across five electronic databases (PubMed, Web of Science Core Collection, EBSCO, Scopus, and Cochrane Library) for peer-reviewed articles published up to May 2026. Inclusion criteria encompassed studies addressing AI-ESRs interventions in dementia care, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Data extraction was standardized to ensure consistency across selected studies. Results 16 studies met the inclusion criteria, comprising seven randomized controlled trials (RCTs) and 9 pre-post studies. 15 studies demonstrated the potential of AI-ESRs to significantly enhance multiple dimensions of social interactions. These included verbal and non-verbal communication, the frequency and quality of interactions with caregivers, peers, and robotic systems, as well as behavioral indicators (e.g., smiling and laughter) and participation in activities. One study reported no statistically significant differences between intervention and control groups (p = 0.18) but noted a marginal improvement in communication scores. Conclusions AI-ESRs show promise in facilitating social interactions among patients with dementia, with their effectiveness influenced by participant characteristics, intervention design, and contextual factors. Further research is warranted to explore long-term outcomes, tailor interventions to individual needs, and address ethical and implementation challenges, thereby advancing the integration of AI-ESRs into dementia care.
Junjie Wang, Jing Shen, Jia-Yi Wang et al.· Frontiers in Psychology· 0 citations
This study investigates digital mental health as a high-stakes problem of conversational information access and interaction. Using Design Science Research Methodology, it develops
Cognitive Theatre
, a risk-aware conversational agent informed by cognitive behavioural therapy and implemented through a controller-mediated role-decomposed architecture. The system separates support into specialised roles for risk assessment, support selection, structured intervention, and user-facing delivery, and coordinates them through risk-aware routing. The artifact was evaluated through safety-routing, latency, AI-based comparative evaluation, and human comparative evaluation. Results show that, under controlled model-generated conditions, the architecture escalated all 50 extreme-risk inputs and received higher response-level ratings than two baselines in AI-based assessment and a two-scenario human evaluation. The study contributes design knowledge for building structured, inspectable, and context-sensitive conversational systems in sensitive domains, and extends research on human-centred conversational information access.
Yiming Zhou, M. Honary, Amjad Fayoumi· Information Systems Frontier...· 0 citations
The advent of Large Language Models has accelerated interest in empathetic conversational agents. Despite a surge in empirical research, artificial empathy remains deeply fragmented, often serving as a catch-all term for diverse interactional phenomena. Addressing this conceptual gap, we systematically review 89 empirical studies to map how human-machine empathy is operationalized. Our synthesis reveals that empathy is highly situated and driven by functional goals, like health and well-being, transactional service, social interaction, and learning support. Within these contexts, we classify affective responsiveness by its directional flow, detailing how agents project, elicit, or mediate empathy. We structure the literature into a cohesive framework spanning linguistic, paralinguistic, identity, and architectural strategies. Furthermore, our methodological evaluation reveals a reliance on adapted clinical metrics, a scarcity of longitudinal studies, and a disproportionate focus on text-based over voice-based interfaces. Ultimately, this review equips researchers and practitioners with an actionable foundation for designing, measuring, and implementing contextually appropriate and empathetic agents.
Supriya Khadka, Smit Desai· International Conference on...· 0 citations