Pupillary Responses to Robotic and Human Emotions: The Uncanny Valley and Media Equation Confirmed.

Anne Reuten,Maureen van Dam,Marnix Naber

doi:10.3389/fpsyg.2018.00774

Abstract

Physiological responses during human–robots interaction are useful alternatives to subjective measures of uncanny feelings for nearly humanlike robots (uncanny valley) and comparable emotional responses between humans and robots (media equation). However, no studies have employed the easily accessible measure of pupillometry to confirm the uncanny valley and media equation hypotheses, evidence in favor of the existence of these hypotheses in interaction with emotional robots is scarce, and previous studies have not controlled for low level image statistics across robot appearances. We therefore recorded pupil size of 40 participants that viewed and rated pictures of robotic and human faces that expressed a variety of basic emotions. The robotic faces varied along the dimension of human likeness from cartoonish to humanlike. We strictly controlled for confounding factors by removing backgrounds, hair, and color, and by equalizing low level image statistics. After the presentation phase, participants indicated to what extent the robots appeared uncanny and humanlike, and whether they could imagine social interaction with the robots in real life situations. The results show that robots rated as nearly humanlike scored higher on uncanniness, scored lower on imagined social interaction, evoked weaker pupil dilations, and their emotional expressions were more difficult to recognize. Pupils dilated most strongly to negative expressions and the pattern of pupil responses across emotions was highly similar between robot and human stimuli. These results highlight the usefulness of pupillometry in emotion studies and robot design by confirming the uncanny valley and media equation hypotheses.

Highlights

Developments in material, electronic, and computer sciences have advanced substantially over the last few decades
Additional general linear models that predicted either recognition scores (Canniness: B = 0.001, p = 0.012; Pupil size: B = 0.09, p < 0.001) or pupil size (Canniness: B < 0.001, p = 0.723; Recognition: B = 0.611, p < 0.001) suggested that emotion recognition scores is partially responsible for both uncanny ratings [r(318) = 0.14, p = 0.013] and weaker pupil dilations [r(318) = 0.24, p < 0.001] of the eerie robots that fall within the uncanny valley but that uncanny ratings do not relate to weaker pupil dilations [r(318) = 0.01, p = 0.804]
We investigated the recognition of emotional facial expressions displayed by robots and humans and measured pupillary responses as objective markers of visuo-emotional processing by the nervous system

Summary

Introduction

Developments in material, electronic, and computer sciences have advanced substantially over the last few decades. One central assumption in robotic design is that the more humanlike socalled “social” robots appear, the more the user will expect the robot to behave like a human being (Duffy and Joue, 2004), and to engage into social interactions with them. Robots that look a lot but not quite like humans appear odd and eerie (Tinwell et al, 2011). This phenomenon is often explained in the context of the uncanny valley hypothesis (Mori, 1970), a theoretical assumption that the level of eeriness is explained by an observer’s unfamiliarity with humanlike robotic faces. It is interesting to discuss whether familiarity or another construct underlies the uncanny valley (e.g., Brenton et al, 2005; Kätsyri et al, 2015), we here only focus on how the uncanny valley can be measured

Objectives

Methods

Results

Conclusion