Abstract
Two conditions of the picture presentation method in the driving simulator (DS) were compared for the estimation of driver gap acceptance in turning across opposite traffic. One was the stereo condition in which the picture was presented with binocular disparity consistent with the simulated scene. The other was the 2-D condition in which the picture was presented with zero binocular disparity all over the picture. We did not find significant differences between the estimations in the two conditions for the acceptance threshold and for the risk score at the acceptance threshold. Requirements for the picture presented in DS was discussed.