Title
Getting acquainted: First steps for child-robot relationship formation
Author
Ligthart, M.E.U.
Neerincx, M.A.
Hindriks, K.V.
Publication year
2022
Abstract
In this article we discuss two studies of children getting acquainted with an autonomous socially assistive robot. The success of the first encounter is key for a sustainable long-term supportive relationship. We provide four validated behavior design elements that enable the robot to robustly get acquainted with the child. The first are five conversational patterns that allow children to comfortably self-disclose to the robot. The second is a reciprocation strategy that enables the robot to adequately respond to the children’s self-disclosures. The third is a ‘how to talk to me’ tutorial. The fourth is a personality profile for the robot that creates more rapport and comfort between the child and the robot. The designs were validated with two user studies (N1 = 30, N2 = 75, 8–11 years. o. children). The results furthermore showed similarities between how children form relationships with people and how children form relationships with robots. Most importantly, self-disclosure, and specifically how intimate the self-disclosures are, is an important predictor for the success of child-robot relationship formation. Speech recognition errors reduces the intimacy and feeling similar to the robot increases the intimacy of self-disclosures. Copyright © 2022 Ligthart, Neerincx and Hindriks.
Subject
Child-robot interaction
Getting acquainted
Human-robot interaction
Relationship formation
Self-disclosure
Social robots
User study
To reference this document use:
http://resolver.tudelft.nl/uuid:3497c60e-e406-4502-ac28-c5b37097ccc8
DOI
https://doi.org/10.3389/frobt.2022.853665
TNO identifier
977312
ISSN
2296-9144
Source
Frontiers in Robotics and AI, 9 (9)
Document type
article