Figure S1: Stimulus presentation scheme

A. Grayscale images were presented for 200 ms, with a 600 ms gray screen in between images. Subjects were instructed to fixate and performed a one-back task indicating whether the same object was repeated (regardless of changes in scale or viewpoint). Image order was pseudo-randomized. B. Objects belonged to one of 5 possible categories. Here and throughout the manuscript, each category is indicated by a separate color (red=animals, green=chairs, blue=faces, black=fruits, yellow=vehicles). There were 5 exemplar objects per category. These images show the “default” viewpoint for each exemplar. C. Each object was presented in one of 5 possible transformations, illustrated here for only one object. There were 3 depth rotations (0, ~45 and ~90 degrees) at the same scale and 3 scales (visual angles of 1.5 degrees, 3 degrees and 6 degrees) with the same rotation for each object. The duration of each session depended on clinical constraints and subject fatigue (min duration = 7 min., max duration = 40 min, mean=19.6±7.9 mins). In many cases we were able to record several sessions per subject (min=1, max=7, mean=2.6±1.6). Results were very consistent across sessions; throughout the manuscript we merged the responses across all sessions for each subject.