The Effect of Robot’s Facial Features on Users’ Perception: Evidence from Subjective, Pupillometry, and Electroencephalography Measures

AbstractThe present study aimed to investigate how facial features, including facial ratio, eye shape, and mouth presence, influence users’ perceptions of social robots in terms of anthropomorphism, trustworthiness, and overall impressions. Using electroencephalogram (EEG) and eye-tracking technologies, we conducted a 2 (face ratio) × 2 (eye shape) × 2 (with or without mouth) full factorial experiment designed within-subject. EEG signals and pupil diameters were recorded while participants viewed images of robots with different facial design features. The results showed that the shape of robot’s eyes significantly influenced users’ perceptions, with round eyes being associated with higher ratings of anthropomorphism, trustworthiness, and overall positivity. Robots with lower facial width-to-height ratios (fWHR) induced smaller average pupil diameters in users than those with high fWHR. Additionally, robots with lower fWHR were perceived as more anthropomorphic, trustworthy and users produced higher theta rhythm when the eye shape was round but did not when it was rectangular. The presence of a mouth led to higher ratings of anthropomorphism rather than trustworthiness and other measures. Overall, the findings highlight the importance of facial features in shaping users’ perceptions of social robots and provide practical implications for designing social robots with anthropomorphic facial features.Keywords: Facial featurestrustworthinessanthropomorphismtheta rhythmpupillometry AcknowledgmentWe would like to acknowledge the participants of the study for their time and involvement in the experiment.Disclosure statementNo potential conflict of interest was reported by the author(s).Additional informationFundingThis work was supported by the National Natural Science Foundation of China (Grant No. 72071035 and Grant No. 71771045) and the Natural Science Foundation of Anhui Higher Education Institutions (Grant No. 2023AH051077).Notes on contributorsZenggen RenZenggen Ren is a Ph.D. candidate at the Department of Industrial Engineering, School of Business Administration, Northeastern University, China. He obtained his Master degree in Human Factors from Northeastern University in 2019. His research interests include human factors, human-robot interaction, and intelligent interaction.Fu GuoFu Guo is a professor of Industrial Engineering at School of Business Administration, Northeastern University, China. She is the peer-reviewer for the National Natural Science Foundation of China. Her research interests include human factors, kansei engineering, user experience design, human-robot interaction, occupational safety and health, consumer behavior.Mingming LiMingming Li is a lecturer of Industrial Engineering at College of Management Science and Engineering, Anhui University of Technology, China. He obtained his Doctor degree in Human Factors from Northeastern University in 2023. His research interests include human factors, human-robot interaction, user experience design, and human-computer interaction.Mingcai HuMingcai Hu is a lecturer of Industrial Engineering at School of Management, Jiangsu University, China. He obtained his Doctor degree in Human Factors from Northeastern University in 2021. His research interests include human factors, kansei engineering, user experience design.Vincent G. DuffyVincent G. Duffy is a full professor of Industrial Engineering and Agricultural & Biological Engineering at Purdue University. He has been the chair or co-chair of Applied Human Factors and Ergonomics International Conference and Human-Computer Interaction International Conference. His research interests include digital human modeling, safety engineering, and ergonomics.