Size Distortions: Space-Time Interaction

Abstract

We used a modified method of constant stimuli to measure spatial interval discrimination thresholds. Horizontal intervals were indicated by a pair of dark vertical lines on a bright background. In each experimental session, thresholds were measured for seven reference stimuli, presented in random order. Reference stimulus separations varied from 9.52 to 16.66 min−1 in increments of 1.95 min−1. The interstimulus interval (ISI) was varied (50, 200, 500, and 1000 ms) between experimental sessions. Stimulus duration was constant at 500 ms. For all ISI durations, the point of subjective equality (PSE) for small spatial separation references was less than physical equality, the PSE for larger separations was greater, and the PSE was close to physical equality for reference stimuli in the centre of the range. This result is consistent with the modular model [V D Glezer, 1995 Vision and Mind (Mahwah, NJ: Lawrence Erlbaum)]. However, the magnitude of the PSE shifts was affected by the ISI duration: at 50 and 1000 ms, the small spatial intervals were more underestimated and the large ones were more overestimated than at 200 or 500 ms. The discriminability thresholds based on the slopes of the psychometric functions varied inversely with the ISI duration, but at the ISI of 1000 ms increased again. These findings demonstrate that in the sequential mode of presentation the temporal separation can be as important as the spatial separation distribution in determining the PSE. This suggests that these size distortions result more from memory processing than from spatial processing.