With the rapid acceleration of global population aging, nursing robots are increasingly regarded as a crucial technological solution for addressing caregiver shortages and improving the quality of life of older adults. However, existing studies have p...
With the rapid acceleration of global population aging, nursing robots are increasingly regarded as a crucial technological solution for addressing caregiver shortages and improving the quality of life of older adults. However, existing studies have predominantly focused on functional and usability aspects, while insufficient attention has been paid to the emotional experience and trust-building mechanisms of older adults during actual interactions with robots. To address this gap, the present study integrates four theoretical frameworks—Inclusive Design, Emotional Interaction, Human–Robot Interaction (HRI), and User Experience (UX)—and proposes a comprehensive concept termed Inclusive Emotional Interaction Features, which aims to explain how emotional cues shape user experience and trust formation in elderly-robot interactions.
This study adopts a mixed-method approach combining interviews and experiments. First, in-depth interviews were conducted with 8 HRI experts and 10 older adult users to extract the key interaction features most relevant to elderly users. Based on the four theoretical frameworks, a four-dimensional structure of Inclusive Emotional Interaction Features was developed: emotional accessibility (appearance friendliness), emotional expressiveness (vocal qualities), empathetic responsiveness (motion synchrony), and interaction transparency (interface clarity). Subsequently, two controlled experiments were conducted. Experiment 1 (n = 40) employed a 2×4 factorial design to examine how high vs. low levels of the four features influence emotional experience (pleasure, psychological safety, emotional resonance) and trust (ability, benevolence, integrity). Experiment 2 (n = 40) focused on micro-motion frequency, comparing low, moderate, and high frequencies to test its nonlinear effects on trust. All data were collected using Likert-scale questionnaires (1–5) and analyzed using t-tests, ANOVA, MANOVA, and quadratic regression.
Results show that: (1) all four inclusive emotional interaction features significantly enhanced older adults’ emotional experience, with motion synchrony and vocal emotional expressiveness showing the strongest effects; (2) high-inclusive interaction environments significantly improved trust across all three dimensions, with benevolence trust exhibiting the largest effect size; (3) emotional cues (e.g., voice, facial expression) had a stronger influence on trust formation than informational cues (e.g., interface clarity, interaction distance), aligning with established HRI and UX findings that early-stage judgments are predominantly emotion-driven; (4) micro-motion frequency demonstrated a clear inverted U-shaped effect, with moderate frequency (once every 6–8 seconds) producing the highest trust levels, while overly frequent movements generated tension and reduced trust.
The contributions of this study are threefold. First, it integrates Inclusive Design, Emotional Interaction, HRI, and UX to construct a novel framework of Inclusive Emotional Interaction Features, extending the theoretical boundaries of emotional experience design in nursing robots. Second, it empirically establishes a sequential mechanism linking emotional experience to trust formation, addressing gaps in prior research that overlooked affective mediation. Third, it provides experimental evidence for the nonlinear effects of micro-motion rhythms, offering new insights into motion-based affective cues in HRI.
This study provides theoretical and practical guidance for designing emotionally attuned and trust-enhancing nursing robots for older adults. Future research may incorporate real-world deployment, longitudinal tracking, and individual differences to build more ecologically valid trust models in elderly–robot interaction.