File S1 - Human Emotion Experiences Can Be Predicted on Theoretical Grounds: Evidence from Verbal Labeling

Supporting Information. Figure S1, Prototype matching with Manhattan distance for two-dimensional artificial appraisal data. User-observed appraisal input is depicted as data clouds in the 5 by 5 appraisal space. Theoretical emotion centroids for joy (circle), rage (diamond) and fear (square) are situated in this appraisal space. For a given user’s appraisal input (black dot), the GEA system calculates the Manhattan distance to each emotion centroid. The emotion with the shortest distance is outputted as the GEA’s prediction, in this case fear. Table S1, Standardized appraisal profiles for the 13 Geneva Emotion Analyst emotions. Displayed are the theoretically predicted profiles (THEO), the empirically found profiles for the GEA data (GEA), and the empirically found profiles for the GRID data (GRID). Pearson correlations between each pair of columns are displayed on the bottom row. Empty cells indicate that the corresponding appraisal item was not present for the respective profile. Refer to Table S2 for appraisal variable names. Table S2, List of appraisal questions in the GEA questionnaire, including appraisal check name, data abbreviation, and weight used in the adjustment of the distance functions.
(DOCX)