Ashley Perez

Major and Classification

Computer Science (Games)

Faculty Mentor

Maja Matarić


Viterbi School of Engineering

Research Gateway Project

Perceived Empathy of Socially Assistive Robots

Project Abstract

Communicating empathy is a difficult skill for socially assistive robots. Work developing computational models of empathy has been growing rapidly, demonstrating the importance for machines to learn this skill. Despite this, questions remain about how beliefs in robot agency may mediate perceptions of robot empathy. Do people really believe that robots can feel emotions? Building a relationship between robot and user is important to the field of Human-Robot Interaction (HRI), especially in the context of healthcare. A robot’s failure to be perceived as empathetic by its human user could be detrimental to the human- robot relationship. This work studies the difference in viewer’s perceptions of cognitive and affective empathetic statements made by a robot in response to a disclosure. In this within-subjects study, participants (n=200) watch videos in which a robot responds to a human who is disclosing negative emotions around COVID-19, with either affective or cognitive empathetic responses. Using an adapted version of the RoPE Scale, participants will rate the robot’s perceived empathy in both cases. We hypothesize that the cognitive statements made by the robot will be perceived as more believable and more empathetic. We will translate these findings towards the creation of an autonomous facilitator robot for support groups.