Figure 3.

Identification of Videos via decoded Emotion Scores

(A) Mean video identification accuracies from region-wise decoders. Dots indicate accuracies from individual brain regions (five subjects averaged; dashed lines, chance level, 50%; see Figure S1C for individual subjects). Color and shape of each dot indicate locations of individual brain regions as Figure 2C.

(B) Mean video identification accuracies from ensemble decoders. Dots indicate accuracies of individual subjects (error bars, 95% C.I. across subjects; dashed line, chance level, 50%).

(C) Histogram of video identification accuracies for individual videos (ensemble decoder; five subjects averaged; dashed line, chance level, 50%).

(D) Two-dimensional maps of emotional experiences constructed from reported and decoded scores. Each dot corresponds to each video and is colored according to its original reported scores using a weighted interpolation of unique colors assigned to 27 distinct categories (see Cowen and Keltner, 2017 for the color schema).