Drawing parallels in human–other interactions: a trans-disciplinary approach to developing human–robot interaction methodologies

Abstract

This opinion paper discusses how human–robot interaction (HRI) methodologies can be robustly developed by drawing on insights from fields outside of HRI that explore human–other interactions. The paper presents a framework that draws parallels between HRIs, and human–human, human–animal and human–object interaction literature, by considering the morphology and use of a robot to aid the development of robust HRI methodologies. The paper then briefly presents some novel empirical work as proof of concept to exemplify how the framework can help researchers define the mechanism of effect taking place within specific HRIs. The empirical work draws on known mechanisms of effect in animal-assisted therapy, and behavioural observations of touch patterns and their relation to individual differences in caring and attachment styles, and details how this trans-disciplinary approach to HRI methodology development was used to explore how an interaction with an animal-like robot was impacting a user. In doing so, this opinion piece outlines how useful objective, psychological measures of social cognition can be for deepening our understanding of HRI, and developing richer HRI methodologies, which take us away from questions that simply ask ‘Is this a good robot?’, and closer towards questions that ask ‘What mechanism of effect is occurring here, through which effective HRI is being performed?’ This paper further proposes that in using trans-disciplinary methodologies, experimental HRI can also be used to study human social cognition in and of itself. This article is part of the theme issue ‘From social brains to social robots: applying neurocognitive insights to human–robot interaction’.

Social Cognition in the Age of Human–Robot Interaction

“…A central question in human-robot interaction studies should be what the appropriate comparison categories are for different types of robots. Of course, these could range from humans to objects to animals, and the best answer will naturally depend on the specific research question being tested [29]. To establish the place robots might occupy in our social milieu, we need to measure the (dis)similarity to animate agents (e.g., a human or pet) as well as objects (e.g., a phone).…”

From social brains to social robots: applying neurocognitive insights to human–robot interaction

“…The final piece for this section is a paper by Collins [23] which provides a comparative approach for analysing HRI, by referring to human -animal interaction. Collins points out that HRI is a system which consists of three elements: the human, the robot and their interaction.…”

Mind Meets Machine: Towards a Cognitive Science of Human–Machine Interactions

“…as well as nonsocial artificial agents and non-human animals [58]. In principle, therefore, research examining object, tool, and animal interactions seems as valuable as social cognition research (cf.…”