Fig. 1: Experimental design and fixation decoding.
From: Dynamic face-related eye movement representations in the human ventral pathway
a An example trial sequence in the behavioral experiment. For anonymization, the face image shown here was complemented, binarized, and denoised using the functions of imcomplement, imbinarize, and bwareopen in MATLAB 2021b. b An example trial sequence in the Gaze Session of the MEG experiment. c An example block of the stimuli sequence in the Image Session of the MEG experiment. d An example face and house image overlaid with the fixations from one example participant (upper left panel), and the fixation patterns (lower left panel) collected from the same participant during the view of faces (blue star) and houses (green cross). These fixation patterns were used to train a classifier in discriminating face- and house-related fixation patterns. The trained classifier was then used to classify the fixation patterns collected during the following of gaze tracks from the current observer (SF vs. SH, upper right panel) and the fixation patterns during the following of gaze tracks from another participant (OF vs. OH, lower right panel, i.e., cross-experiment classification). e The prediction accuracies of the cross-experiment classification are shown as a function of the comparison (SF vs. SH and OF vs. OH) and the number of fixations included in the classifications. Chance-level accuracies were derived by permutating the labels of the categories, rendering a distribution of chance accuracies. The shaded areas indicate accuracies below the 95% percentile of the distribution of chance accuracies obtained from the permutation-based classification. **p < 0.01, *p < 0.05 for between self and other (Bonferroni-corrected).
