Skip to main content
. 2018 Apr 1;169:80–93. doi: 10.1016/j.neuroimage.2017.12.013

Fig. 5.

Fig. 5

(A) Results for the subject classification analyses in percentages. Monte Carlo permutation tests were used to determine the significance threshold for each classification accuracy (see Materials and Methods), significance is indicated by a *. The left panel shows the classification of participant groups based on their neural response patterns for birds. The classifier was able to make a distinction between ornithologists and control participants in all three aROIs. The right panel shows the classification for the category of minerals, with a significant distinction between mineralogists and control participants in the high-level visual aROI. (B) Results for the generalization analysis. Generalization accuracies are depicted by the dark grey bars, the light grey bars show the maximal generalization that could be expected given the classification performances for the two expertise effects in each specific aROI (see Materials and Methods). Significance is indicated by an asterisk.