Neural Correlates of Robot Personality Perception: An fNIRS Study

Robot personality design has garnered research interest for its crucial role in enhancing the robot’s social capabilities and promoting user experience. This study aims to use machine learning classification techniques to predict the personality of a robot by analyzing the neural activities in the prefrontal cortex (PFC) when individuals interact with the robot that features a specific personality design (i.e., extroverted or introverted). We recruited 64 participants and divided them into two groups to interact with an extroverted or introverted robot. We collected data using a functional near-infrared spectroscopy (fNIRS) device and, after data preprocessing, selected signal means as features for analysis. After applying six machine learning methods for data classification, we found significant differences in the performance of different classifiers. In addition, we observed that personality classification based on left and right brain data showed different performance. According to the results, we can determine the type of robot personality with which users are interacting based on the medial PFC activities during user‒robot interactions.