Exploring the influence of a dancing virtual agent on the evocation on human emotions

Dance has been the universal language of humans for thousands of years. It has been used as a form of expression and therapy. It is only natural then that virtual humans, often known as Intelligent Virtual Agents (IVAs), should also be able to dance and help humans to express themselves through dance. This thesis presents a study investigating the impact of dancing IVAs on human emotions. The experiment was based on a ‘repeated measures’ design with one within subjects factor (dancing character) and one between-subjects factor (display order). The study analysed the responses of 55 participants comprised of dancers and non-dancers watching a dancing IVA perform three different dances that represent anger, sadness and happiness in different display orders. Analysis of the study data showed statistically significant results confirming a number of hypotheses that watching a dancing IVA depicting different emotions can influence human emotions. The participants’ anger, sadness and happiness were significantly dependent on which dancing character’s emotion they watched. The results of the study also showed that the participants can recognize the emotions depicted by the dancing IVA. Moreover, correct recognition was not a factor for the influence of the dancing IVAs on human emotions except for the happy dancing IVA. These results suggest that IVAs could be useful for dance therapy.