Please note: This seminar will take place in DC 1304.
Xiaoye Michael Wang, Assistant Professor
School of Optometry and Vision Science, University of Waterloo
When users perform well in an immersive environment, what does that performance reveal, and what might it conceal? In this talk, I will draw on research spanning vision science, motor control, computer graphics, and human-computer interaction to examine how interactive systems change the information available for perception and action, how users adjust to those changes, and whether our evaluations capture the consequences that matter.
I will present three connected examples from research on virtual and mixed reality (VR/MR). First, I will introduce a geometrical account of the vergence-accommodation conflict and a software-based approach to reducing its effects on targeted movement, illustrating how a perceptual model can inform a software solution to an otherwise hardware-level problem. Second, I will show how added interaction latency can impair spatial accuracy before reliable changes emerge in movement time or throughput, highlighting the importance of task-relevant performance measures. Third, I will compare VR and video-passthrough MR on the same headset, showing that users can eventually achieve similar accuracy even though the two environments initially bias movement in opposite directions, require different amounts of practice to stabilize, and leave different residual effects after the headset is removed.
Together, these findings suggest that evaluating an immersive system requires more than asking whether users can successfully perform a task: we also need to understand how the system changes perception and action, whether our measures reveal those changes, and whether their effects carry over beyond the immersive environment. These questions are particularly important for applications such as surgical training, where performance within the simulator is ultimately intended to support performance outside it.
Bio: Dr. Xiaoye Michael Wang is an Assistant Professor in the School of Optometry and Vision Science at the University of Waterloo. He trained in psychology, philosophy, vision science, and human movement science, completing his PhD at Indiana University Bloomington followed by postdoctoral fellowships at York University’s Centre for Vision Research and the University of Toronto’s Faculty of Kinesiology and Physical Education.
His research examines how people perceive three-dimensional space and use vision to guide action. Combining behavioural experiments with computational and geometric modelling, he studies how binocular vision and motion information support depth perception and movement control. His current work focuses on how virtual and mixed reality alter perception and action, how people adapt to these changes, and how learning transfers between immersive and physical environments. He also develops computational methods for analysing human movement, including learned movement representations for quantifying basketball movement patterns in collaboration with the Sports Performance Lab at MLSE. His work connects fundamental questions about human perception and motor control with applications in immersive technology, training, rehabilitation, and sport.