Fig. 8: Selective peri-saccadic suppression of low spatial frequencies¹⁰ is a visual phenomenon.

a Left: subjects made saccades towards display center. Right: gratings were flashed peri-saccadically over a uniform gray background (circular “virtual monitor” surrounded by a coarse texture; saccade directions and flash locations: similar to Figs.  and 6). b Left: proportion of correct grating localizations with different spatial frequencies during fixation (“Baseline”; dashed curve) and for peri-saccadically flashed gratings (solid curve). Low spatial frequencies were associated with the strongest suppression relative to baseline. Right: ratio of peri-saccadic to baseline performance (highest spatial frequency not shown because it was at chance performance even in baseline). Suppression depended on grating spatial frequency (χ² = 13.46, p = 0.0092, df = 4, Kruskal–Wallis test; **p < 0.01 for post-hoc pairwise comparisons between the lowest and highest spatial frequencies). c Left: simulated saccade-induced image displacements by translating the virtual monitor and surrounding texture from one corner towards display center. Right: gratings appeared as in a. d The same selective suppression of low spatial frequencies occurred as with real saccades (b). “Baseline” in this context means both no saccades and no virtual monitor and texture displacements. Suppression depended on spatial frequency (χ² = 25.33, p < 0.0001, df = 4, Kruskal–Wallis test; *p < 0.05, **p < 0.01, ***p < 0.001 for post-hoc pairwise comparisons between individual spatial frequencies). e, f With a fine surround texture, both real (e) and simulated (f) saccades were associated with suppression for all spatial frequencies; suppression selectivity¹⁰ was eliminated (χ² = 0.8, p = 0.938, df = 4 for e and χ² = 7.74, p = 0.102, df = 4 for f, Kruskal–Wallis test). Error bars: s.e.m. across individual subjects’ curves. Supplementary Figs. 8–10: full time courses and controls with black surrounds around the virtual monitor. Note that in d, f, we exploited the longer time course of visual suppression (Fig. 6 and Supplementary Figs. 8 and 9) to probe perception at a later time than in b, e. This also explains why suppression appeared quantitatively weaker in d, f than in b, e.