Validation of the Robot Social Attributes Scale (RoSAS) for Human-Robot Interaction through a Human-to-Robot Handover Use Case


This work aims to validate the Robotic Social Attributes Scale (RoSAS) for human-robot interaction.

September 24, 2017
IROS 2017 Workshop on Human-Robot Interaction in Collaborative Manufacturing Environments



Matt Pan (Disney Research)

Elizabeth Croft (University of British Columbia, Canada)

Gunter Niemeyer (Disney Research)


The RoSAS evaluates the social perception of robots. It measures an inventory of 18 items that are collected into three robot attributes: competence, warmth, and discomfort. We apply RoSAS to a human-to-robot handover and confirm the internal consistency and unidimensionality of each attribute. This validates the scale in this use case and suggests a more general applicability to other physical human-robot interactions including collaborative manipulation or turn-taking tasks.

Copyright Notice