Perception and anticipation are important determinants in karate sports. Using temporal and spatial occlusion technique in video presentations is a common method to determine anticipatory cues but the lack of information about depth video presentations seems to affect the results. The aim of this study is compare the responses of karate athletes to occluded attacks shown on a video screen and in a virtual environment. Five expert karate athletes were filmed two synchronized high-speed cameras while responding to nine temporally and spatially occluded sequences of a competition relevant attack at first in a CAVE (Cave Automatic Virtual Environment) and then on a life-size video screen. Their responses were rated as 'correct' or 'incorrect'. The results of the Wilcoxon test show significant differences (Z = -2.325, p