Social Coordination in Human–Robot Interaction Through Reciprocal Engagement

Abstract

In this paper, we provide a theoretical and experimental investigation on human–robot interaction scenarios where collaborative activities are perceived more meaningful and easier to perform by the human agent in the situation of reciprocal engagement with the robot. Inspired by the perceptual-crossing paradigm (Auvray and Rohde in Front Hum Neurosci, [1]), we defined reciprocal engagement as an interaction concept that seeks to extend perceptual crossing to functional purposes. For instance, when a robot and a person face a door and they both intend to go through it, a designed behaviour negotiates who is to go first in appropriate manner, either the human or the robot. The objective of the study is to explore social coordination mechanisms that are easy and natural for the human agent in order to engage in interaction with a robot. Two experiments were conducted which explore interaction scenarios with and without reciprocal engagement between a person and a robot to appreciate their significance for the human actor. In the first one, interaction scenarios are evaluated on the basis on video clips; in the second one, participants were asked to interact directly with the robot according to the same scenarios. The results show a preference of participants towards reciprocal engagement situations which were considered more meaningful and easier to learn. With this work, we hope to inspire design thinking to shift from discrete, procedural design mechanisms to continuous and action-driven mechanisms of social coordination when addressing the interaction between humans and systems in smart environments.