Figures & data
Table 1. Participants’ characteristics.
Table 2. Recognition accuracy of neutral expressions: Model fit (AIC) and effect statistics (Z score, P-values, odds ratios) for models with alternate grouping variables (GVs).
Table 3. Confusions of neutral with other emotional expressions: Effect statistics (Z score, P-values, odds ratios) for the standard model and models with better model fit.
Figure 1. Confusions in diagnostic groups and for groups based on CTQ-subscales sexual abuse, physical neglect, and emotional abuse. Error bars indicate standard errors (calculated on link level and transformed back to response level). Neutral expressions were mistaken as pride or disgust in so few cases that models could not be fitted.
![Figure 1. Confusions in diagnostic groups and for groups based on CTQ-subscales sexual abuse, physical neglect, and emotional abuse. Error bars indicate standard errors (calculated on link level and transformed back to response level). Neutral expressions were mistaken as pride or disgust in so few cases that models could not be fitted.](/cms/asset/2041be8f-9a45-4de1-80a5-a3937ae32743/zept_a_1682929_f0001_b.gif)
Table A1. Group sizes for model fit comparisons.
Table A2. Confusions of other emotional expressions with neutral: Effect Statistics (Z score, P-values, odds ratios) for the Standard Model and Models with better Model Fit.