Abstract
PURPOSE. Human observers report that the extent of perceived motion smear is less during smooth pursuit than when comparable motion of the retinal image occurs during steady fixation. We asked if this reduction of perceived motion smear is attributable to an increase in the speed of the temporal impulse response during pursuit eye movements, as reported previously to occur during saccades (Burr & Morrone, 1996).
METHODS. Increment thresholds were determined for pairs of successively presented flashes (5.9 ms each) of a long horizontal line, presented on a 65 cd/sq m background field, 1 deg above or below a continuously visible cross. The stimulus-onset asynchrony (SOA) between the first and second flash varied randomly among blocks of trials from 5.9 to 234 ms. The observer either fixated on the stationary cross (fixation condition) or tracked the cross as a mirror galvanometer moved the entire display at 8 deg/s (pursuit condition). Monitoring of horizontal eye position indicated that small saccades occurred during target presentations on fewer than 10% of the trials in the pursuit condition. Temporal impulse response functions were modeled as impulse responses of a second order linear system and were estimated using an optimization procedure. When summed at the various SOAs, each estimated impulse response function best fit the aggregate increment threshold data of 6 normal observers.
RESULTS. The measured 2-pulse increment thresholds varied similarly with SOA in the fixation and pursuit conditions. However, the estimated temporal impulse response function indicated a natural temporal frequency ca. 10% higher and a damping ratio ca. 20% larger during the pursuit than the fixation condition.
CONCLUSIONS. Although the fitted temporal impulse response function indicates a faster temporal response during pursuit than fixation, we conclude that the difference in response speed is too small to account for previously reported differences in perceived motion smear.