Abstract
Introduction: In natural social interaction, we are able to monitor the facing direction of others even when viewing them with the far periphery. This experiment assessed the accuracy of judgments of facing direction of a viewed person as a function of the retinal eccentricity of the viewed person and the presence/absence of motion cues signaling rotations of the person's head. Method: The subject (S) viewed a trained “looker” (L) from a distance of 2 m. On each trial, a computer informed L which target to look at and L then aligned his head with the target using a laser pointer mounted on the head. The facing directions covered a range of 160 degrees. S viewed L through a computer-controlled electronic shutter. On some trials, S could see L continuously, including the head rotations between trials when moving from one target to the next. On other trials, the shutter limited viewing to the stationary head positions of L. S aligned a hand-controlled pointer with the judged facing direction of L. The subject fixated a video camcorder, which was positioned to give the desired retinal eccentricity of the looker. The camcorder was used to confirm that S held proper fixation during all judgments. Results: The pointing directions of S were used to compute correlations with L's facing directions. For the 10 subjects of the first experiment, the facing/pointing correlations averaged .96 for an eccentricity of 45 degree in both conditions (with and without visible head rotations). For 90 degree eccentricity, 9 out of 10 subjects performed at above chance when head rotations were visible and 6 out of 10 performed at above chance when only static head position was visible.