A Framework to Design Engaging Interactions in Socially Assistive Robots to Mitigate Dementia-Related Symptoms

This is the authors' abstract. We don't add key points for this paper.

People with dementia (PwD) often exhibit behavioral, physiological, and social symptoms that affect their relationships and reduce their quality of life. Socially assistive robots (SARs) have proven effective in reducing some of these symptoms. However, the design of SARs for dementia care remains a challenging task. In this article, we present a human-centered design framework to assist researchers, designers, and roboticist from the conceptualization of engaging prototypes of SARs until the assessment of their efficacy at producing dementia-relevant health outcomes. The framework incorporates several key principles of a Human-Centered Artificial Intelligence (HCAI) approach. It incorporates the needs and perspectives of the different stakeholders into the design and evaluation of the interactions. It fosters accessibility and inclusion through personalization and encourages human–robot collaboration by facilitating natural and engaging interactions that build trust and rapport. We illustrate the use of the framework with the design of an SAR to support a Cognitive Stimulation Therapy. From this case study, we derive a set of design insights for the development of SARs behaviors and interactions for dementia care.