虚拟数字讲师的手势类型与呈现频率如何影响学习?OA

In digital-intelligent learning environments,the gestures of virtual digital instructors serve as important social cues that exert significant impacts on learners' cognitive processes,emotional experiences and even learning outcomes.Accordingly,this study adopted a multimodal approach integrating eye tracking,electrodermal activity(EDA)measurement,learning experience scales,and learning outcome tests to systematically examine the effects of gesture type(beat gestures vs.deictic gestures)and presentation fre-quency(average presentation vs.enhanced presentation),as well as the underlying mechanisms.The results indicate that in terms of learning experience and outcomes,deictic gestures significantly enhanced emotional arousal,social presence,and transfer test perfor-mance.Average presentation was associated with higher levels of emotional dominance,whereas enhanced presentation was more effec-tive in improving social presence and overall learning outcomes.Among gesture combinations,beat-enhanced gestures most effectively promoted learning motivation and retention test performance.Regarding the underlying mechanisms,enhanced presentation,particular-ly deictic-enhanced gestures,helped reduce cognitive anxiety.In addition,deictic gestures and enhanced presentation promoted greater allocation of visual cognitive processing resources to the learning content area.Learning motivation,social presence,retention test per-formance and transfer test performance were positively correlated with total fixation duration on the learning content area and negatively correlated with total fixation duration on the instructor area.Furthermore,social presence and retention test performance were negative-ly correlated with electrodermal activity levels.Based on these findings,this study puts forward four optimization strategies to provide practical guidance for the design,generation,and application of virtual digital instructors and embodied intelligent agents.