Abstract
Abstract: :
Purpose: Previous studies (Matin, 1972; Pola, 1976; Honda, 1990; Dassonville, et al., 1992; Miller, 1999) have shown that subjects tend to mislocalize a flashed target when it is presented just before, during or shortly after the occurrence of a saccade. The usual account for this mislocalization is that it comes from a "slow" extravisual (exV) signal that starts to change before the saccade and continues to change during and after the saccade. However, all of these studies overlook the important fact that the flash creates a visual (V) signal that persists in vision for as long as 300 ms. Interaction of this V signal persistence with the exV signal could have a significant influence on the psychophysical exV signal (the exV signal as derived from experiment). In this study, we explore the effects of V signal persistence on the psychophysical exV signal, using a slow-signal model and a fast-signal model. Methods: The slow-signal model, in line with the previous studies (above), has an exV signal that begins to change before a saccade and continues to change during and after the saccade. In contrast, the fast-signal model has an exV signal that begins to change just after a saccade and follows approximately the same rapid time-course as the saccade. In both models, a V signal of appropriate persistence (approx. 250 ms) is created by passing a target flash (5 ms) through a cascade of 1st-order lags (time-constant = 15 ms). This V signal, when convolved with the exV signal, gives the psychophysical exV signal at the time of the flash. Thus, flashes were presented at 5 ms intervals from 300 ms before until 200 ms after the saccade in order to determine the overall psychophysical exV signal. Results: For both models, the interaction of the V signal persistence and the exV signal resulted in a slowly changing psychophysical exV signal. However, in the slow-signal model, the psychophysical exV signal began to change 300 ms before the saccade and stopped changing at the onset of the saccade, whereas, in the fast-signal model, the psychophysical exV signal began to change 100 ms before and continued until 100 ms after the saccade. Conclusion: The psychophysical exV signal generated by the slow-signal model is unlike the previous experimental results (above). On the other hand, the psychophysical exV signal produced by the fast-signal model bears a strong resemblance to the previous findings. The fast-signal model suggests that: (1) the slow psychophysical exV signal observed in previous studies may be an artifact of using a flashed target, and (2) the actual exV signal may have a time course similar to that of a saccade.
Keywords: 406 eye movements • 585 space perception • 619 vision and action