Abstract
Subjects often mislocalize the position of visual targets that are presented near the time of saccade onset (Ross et al 1997). This mislocalization is manifested as a perceived compression of visual space toward the saccade target and a shift in the direction of the saccade. One explanation for the pre-saccadic spatial shift is that afferent processing delays cause uncertainty about the timing of localization targets (LTs) with respect to saccade onset (e.g. Schlag & Schlag-Rey 2002). However, in recent studies, varying the afferent processing delays by modulating luminance or contrast of LTs did not strongly affect the spatial shift (Boucher et al. 2001, Michels & Lappe 2004). To further investigate the role of afferent delays, we measured localization errors as well as the perceived temporal relationship between LT presentation and saccade onset, for different saccade amplitudes and LT luminances.
We found that the temporal relationship between saccade onset and LT presentation was systematically misperceived by all subjects. This temporal error is consistent with timing differences between incoming visual information and the eye-position signals, and it is greater for LTs at low luminance. Furthermore, spatial localization errors for LTs flashed in a time window between 50 ms before and 20 ms after saccade onset were dependent on both stimulus luminance and saccade amplitude. High LT-luminance in particular resulted in a reduced spatial shift.
These changes in spatial perception at different luminance levels are consistent with the idea that afferent delays play an important role in generation of pre-saccadic localization errors. The effect of saccade amplitude on perisaccaic spatial localization may be taken as further evidence for the interaction of the eye position signal and visual signals in modulating the resulting perceptual errors.
Supported by CIHR and NSF cognitive neuroscience grant