Abstract
Each saccadic eye movement changes the retinal signal. We perceive the world as stable. The classical explanation for visual stability is that information about the eye movement is combined with feedback to discount self-induced retinal signals. We applied classification image (CI) analysis to investigate saccadic suppression of displacement (SSD). We compared CIs with the motor behaviur of the eye. We tested 5 observers who reported the location (left or right) of a pre-saccadic stimulus (a small square embedded in noise). During saccade execution, the noise matrix refreshed and the target stepped horizontally backward or forward. Observers reported the direction of the trans-saccadic target step. We compared CIs from peripheral (Figures 1A-B) and foveal tasks (Figures 1D-E), landing errors of primary and secondary saccades (Figures 1C and 1F). Peripheral perceptual CIs were indistinguishable from saccadic CIs. In the foveal task, perceptual responses required less target signal than saccades (Figure 2A) and a strong saccadic bias was observed towards backward target steps (Figure 2B). Furthermore, we found a spatially similar distortion in perceptual CIs and landing positions of the corrective saccades. The distortion is stronger for backward than forward eye movements (Figure 2C). We discuss that visual stability arises from an interplay between the efference copy and a prior believe of stability (e.g. step was not caused by an internal error). When the post-saccadic target position falls within an elliptic region equivalent to saccadic variability (Figure 1C), perceptual CIs and landing positions of secondary saccades display a similar distortion (center of mass shifts away from target), stability is assumed. Outside this region, as with forward steps, displacements are noticed and positions are remapped.
Meeting abstract presented at VSS 2017