Motion-induced position shift in early Alzheimer's disease

Abstract

The localization of object position in space is one of the most important visual abilities in humans. Motion-induced position shift is a perceptual illusion in which the position of a moving object is perceived to be shifted in the direction of motion. In this study, we wanted to explore whether and how Alzheimer's disease (AD) affects this illusion. We recruited a group of patients with early AD and a group of age-matched healthy controls. In our experiments, two drifting Gabor patches moving in opposite directions were presented and participants were asked to report whether the upper Gabor appeared rightwards or leftwards of the lower one. We measured the psychometric functions, of which the point of subjective alignment was taken as the magnitude of motion-induced position shift. We compared the position shift across the two groups at three different retinal eccentricities. We found that position shifts were systematically smaller in the AD group as comparing to the elderly control group. Our data demonstrated that AD patients were less prone to motion-induced position shift. The results add to the existing knowledge of perceptual deficits in AD patients. We suggest that motion induced position shift may be effective as a new behavioral indicator for AD identification.

Figure 1

Task designs. (A) Observers viewed a pair of drifting Gabor patches moving in opposite directions around fixation (one above and the other one below). They were required to judge whether the upper Gabor is leftwards of the lower one by keyboard presses. (B) The schematic illustration of motion-induced position shift. When the two Gabor patches are physically aligned, they appear to be horizontally displaced in the direction of motion. (C) Our experiments manipulated the vertical eccentricity of the Gabor patches (relative to the fixation spot) and horizontal offsets between Gabors.

Figure 2

The psychometric measurement of position shifts in two representative subjects. The y-axis plotted the probability of perceiving the upper Gabor to be leftwards of the lower one as a function of horizontal offset. Data were shown for one AD patient and one elderly subject at 3° eccentricity. We fitted a cumulative Gaussian function to the data and the point of 50% performance was taken as the magnitude of position shift.

Figure 3

The dependence of position shift on the retinal eccentricity in two groups of participants. Results are shown for mean ± SEM averaged within each group.