A visual and neural evaluation of the affective impression on humanoid robot appearances in free viewing
This study aims to characterize the fixation pattern and neutral dynamics underlying affective impressions on humanoid robot appearances, which have a moderate anthropomorphism, to provide an objective evaluation of the affective impressions. Thirty appearance pictures of existing humanoid robots le...
Saved in:
| Published in: | International journal of industrial ergonomics 2022-03, Vol.88, p.103159, Article 103159 |
|---|---|
| Main Authors: | , , , |
| Format: | Article |
| Language: | eng |
| Subjects: | |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This study aims to characterize the fixation pattern and neutral dynamics underlying affective impressions on humanoid robot appearances, which have a moderate anthropomorphism, to provide an objective evaluation of the affective impressions. Thirty appearance pictures of existing humanoid robots leaving negative, neutral, or positive impressions were selected online. Users’ eye-tracking and electroencephalography signals for the appearances were co-registered in a free viewing paradigm, and eye-tracking metrics and eye fixation-related potentials were extracted. The results showed that the head attracted the most eye-tracking metrics, including glance count, fixation count, fixation time, and fixation duration. The torso attracted the second most, the legs and the hands attracted the third most, and the arms and feet drew the least. Humanoid robot appearances giving positive impressions caught less glance counts and shorter fixation duration than the appearances giving negative or neutral impressions on both partial and whole appearances. The appearances giving positive impressions evoked the smallest P1 amplitude, and the appearances giving neutral impressions evoked the largest P1 amplitude, compared with that giving negative impressions. The findings suggest that glance count and fixation duration could distinguish the positive impression on humanoid robot appearances, and the P1 amplitude could distinguish the three affective impressions.
These findings could promote the understanding of fixation pattern on humanoid robot appearances, which could be referred to directly when designing appearances for humanoid robots. Moreover, this study could provide designers with a visual and neural method of evaluating affective impressions on humanoid robot appearances in free viewing.
•The head of humanoid robot appearances caught the most visual attention.•The appearances giving positive impressions caught less glance counts and shorter fixation duration.•The fixation-related P1 amplitude varied across humanoid robot appearances giving positive, neutral, or negative impressions.•The co-registration of eye-tracking and EEG signals could distinguish affective impressions on humanoid robot appearances. |
|---|---|
| ISSN: | 0169-8141 1872-8219 |