Compensation for smooth eye and head movements by gaze saccades during head-unrestrained tracking

Pierre Daye, Gunnar Blohm, Philippe Lefèvre

Gaze saccades are generally composed of a combination of eye and head movements. In head fixed conditions, smooth eye movements that occur during the saccade latency period are taken into account by the saccade if the nervous system has enough time to integrate eye velocity (Blohm et al., 2003, 2005 and 2006). In the present study we investigate this compensatory mechanism when the head was unrestrained in order to possibly differentiate the compensation for eye movements from the compensation for head movements occurring during the saccade latency period.
Subjects sat in front of a 1-m distant tangential screen. They were instructed to pursue with their eyes and head a sinusoidal target (Frequency [0.6 to 1.2 Hz]) moving along a straight line in 2D (Orientation [0 to 360°] and amplitude [20 to 25°] randomly chosen). 1.2-1.8s after target motion onset, a second target was briefly flashed at a random position on the screen. The position of both eyes was recorded by a video-based recording device (200 Hz), head position was recorded by active infrared marker tracking cameras (200 Hz) and gaze orientation was reconstructed (Ronsse & al., 2007).
We analyzed how the orienting gaze shift towards the flash was programmed and how these saccades compensated for the smooth eye (SED) and head (SHD) displacements during the latency period. Multiple regression analysis (R=0.94, N=899, p<0.001) showed that gaze saccades were programmed using the retinal position error of the flash [PE] (slope=1.03, partial R=0.92, p<0.001) and estimations of SED (slope=0.66, partial R=0.51, p<0.001) and SHD (slope=0.54, partial R=0.64, p<0.001) at saccade onset. In 45% of trials, we observed at least one (and up to 4) additional saccade(s) contributing to the orientation process to the flash. At the end of the orientation, subjects compensated overall for 74% of total SED and 69% of total SHD, which was compatible with previous findings (Blohm et al., 2003, 2005 and 2006).
In conclusion, we propose that the gaze control system uses similar mechanisms to program head restrained and head unrestrained saccades. Smooth eye and head displacements during gaze latency were integrated to compensate for intervening eye and head movements. These results point towards a synergy between eye and head velocity signals used by the gaze control system.

Supported by: Marie Curie (EU), FNRS (Belgium), IAP (Belgium), ESA (EU), ARC (UCLouvain, Belgium), NSERC (Canada), Botterell Fund (Queen’s University, Canada)