Virtual avatar communication task eliciting pseudo-social isolation and detecting social isolation using non-verbal signal monitoring in older adults

Social isolation and loneliness are two of the main causes of mental health problems or suicide, not only in younger adults but also in older adults. Thus, identifying an effective method to detect social isolation is important in the field of human-machine interaction. However, to the best of our knowledge, no effective method has been developed to elicit pseudosocial isolation tasks to evaluate social isolation detection systems for older adults. This study has two research aims: 1. To develop a virtual avatar conversation cyberball task to evoke pseudosocial isolation in older adults and, 2. to identify non-verbal indicators that replace social isolation in older adults. To achieve these objectives, 22 older men were recruited as participants. They were asked to communicate with two virtual avatars on a monitor and then to rate the follow-up questions provided to evaluate the level of social isolation and emotions; meanwhile, facial expressions and gaze patterns were recorded by a camera and an eye tracker. In the results, the developed virtual avatar conversation cyberball task successfully induced pseudosocial isolation in older adults, and this social isolation was detected by the intensity of inner/outer eyebrow and eyelid movements and the blink frequency.