Fig. 2: Stimuli and group-level data for the Curveball experiment (n = 10).

a Gabor patterns in the illusory and control conditions travelled the same physical path. However, in the illusory condition, the internal gratings drifted in the direction orthogonal to the external path (towards the periphery); causing the perception of a ‘curveball’ when viewed in the periphery. b Sliding window sampling was used at the peripheral (blue) and central (yellow) location, corresponding to the start and the end of the motion path, respectively. Window height accounted for cortical magnification. c V1 responses reconstructed in visual field. Red and blue denote positive and negative responses relative to baseline, respectively. White dashed lines denote the physical motion paths. d Individual fMRI effects (defined as illusory μ– control μ in the central location) plotted against perceptual effects. The shaded region denotes the 95% CI estimated through bootstrapping. e Group-level neural signatures collapsed across hemifields and fit with a Gaussian function. Both illusory and control signatures across both sliding window locations were centred on the physical motion path (vertical dotted lines). f Predicted neural signatures for simulated physical motion trajectories were correlated with the measured signatures. g Heat map showing correlation coefficient across various simulated physical motion paths. The best correlated (red line) signatures corresponded to the physical motion path.