Evaluating proactivity levels in socially assistive robots for elderly care: a user adoption assessment

Abstract

This study evaluates different levels of proactivity in socially assistive robots (SARs) within elderly care, focussing on user acceptance and comfort. Using the mPLiCA model, we categorise proactive behaviours from basic presence to full autonomy. A series of video scenarios was created to represent varying degrees of proactive interactions with SHARA, a socially assistive robot. Participants rated these scenarios based on usefulness, appropriateness, perceived intrusiveness and naturalness. Results indicate that higher levels of proactivity are perceived as more useful but also more intrusive. Intermediate levels, particularly dialogue-based interactions, were the most acceptable. This research offers insights into designing SARs that balance autonomy and user comfort in elderly care, contributing to the development of more effective assistive technologies.