Education using humanoid robots can have a positive impact in many fields, including in medical or physical training. This study investigated the effects of robot interactions with respect to facial expressions, gestures, voices and their combinations on the education of the elderly regarding information and communications technology (ICT) from functional and emotional perspectives. In this study, the robot's interaction methods were divided into four categories: (1) voice, (2) voice and expression, (3) voice and gesture, and (4) voice and expression and gesture. An experiment involving an educational application with a humanoid robot was conducted with a total of 15 elderly people over the age of 60. The effect of the humanoid robot's interaction method on education was identified by means of subjective survey evaluation and practice performance data analysis, including error rate, task success rate, and number of retrainings. Through the experiment, functional and emotional aspects of effects were measured. The results showed that performance and perceived effectiveness were not significantly affected by the type of robot interaction, but the degree to which the robot felt like it had emotions, the degree to which the robot felt like a human, and the degree to which the robot was friendly were significantly different according to the interaction type employed by the humanoid robot. The best effect was achieved when voice and gesture were used together during tutoring. Recognizing that ICT education using humanoid robots increases interest and participation in education, such robots are concluded to be a suitable method for performing ICT education. In addition, when designing robotic interactions, the use of the robot's voice and gestures together is expected to lead to greater anthropomorphism, resulting in a stronger relationship with humanoid robots.
Keywords: education; elderly; human–robot interaction; interaction type; robot.