松香 敏彦

The present study investigated (1) how well humans can recognize facial expressions represented by a small set of landmarks, a commonly used technique in facial recognition in machine learning and (2) differences in conscious observational behaviors to recognized different types of expressions. Our video stimuli consisted of facial expression represented by 68 landmark points. Conscious observational behaviors were measured by movements of the mouse cursor where a small area around it was only visible to participants. We constructed Bayesian models to analyze how personality traits and observational behaviors influenced how participants recognized different facial expressions. We found that humans could recognize positive expressions with high accuracy, similar to machine learning, even when faces were represented by a small set of landmarks. Although humans fared better than machine learning, recognition of negative expressions was not as high as positives. Our results also showed that personality traits and conscious observational behaviors significantly influenced recognizing facial expressions. For example, people with high agreeableness could correctly recognize faces expressing happiness by observing several areas among faces without focusing on any specific part for very long. These results suggest a mechanism whereby personality traits lead to different conscious observational behaviors and recognitions of facial expressions are based on information obtained through those observational behaviors.

A back-channel is a short utterance made by non-primary speakers. In the Japanese language back-channel is called 'Aizuchi, ' Aizuchi has long been considered as a positive reaction to the primary speaker showing agreement or appraisal and or giving the right to continue speaking. Togashi argues that, in contrast to the conventional view, the essence of aizuchi is a sign of internal information processing in the mind of the one who uses aizuchi. The present study examined the usage of aizuchi as a non-primary speaker's internal information processing using Japanese map task dialogue. The result of our analysis showed that aizuchi often occurred when the primary speakers introduced an object known by the non-primary speakers into the conversation as predicted by Togashi's view of aizuchi. It was also suggested that the presence/absence of aizuchi was not necessarily a clear sign indicating the non-primary speakers have knowledge about the target object but rough but useful information to predict the knowledge state of the non-primary speakers.

Previous studies indicated that how people look at others' faces depends on observers' characteristics. A vast majority of these studies used aggregated data, treating gaze data in a static manner and often neglected dynamics of observational behaviors. The present study, on the other hand, examined the relationship between observers' personality traits and observational behaviors (i.e., which areas of face were looked at) using time-series eye movement data. In particular, we collected gaze data using a smile judgments task, asking participants to judge whether smiley faces were genuine or not. Observational behav-iors were analyzed using a hidden Markov model. The results showed that participants with different patterns of personality traits exhibited different transition patterns. In addition, there were also significant interactions between personality traits and the number of hidden states each participant had. Furthermore, there were significant relationships between participants' person-ality traits and the result of judgments.