Emotion recognition, face processing and crystallized intelligence in personality pathology: A multimethod study

Personality disorders (PDs) are commonly associated with impairments in social cognition, particularly emotion recognition. However, evidence also suggests deficits in more basic cognitive functions, including face processing (e.g., detection of non-emotional facial features) and indicators of general intelligence (e.g., crystallized intelligence). We addressed two exploratory research questions: first, how emotion recognition, face processing, and crystallized intelligence relate to personality pathology across categorical and dimensional PD models, and second, whether these associations primarily reflect general or specific personality pathology. We further hypothesized that the association between personality functioning and emotion recognition would be attenuated when controlling for face processing or crystallized intelligence. In a clinical sample (n = 219), cognitive abilities were assessed using performance-based measures, and personality pathology was assessed using interviews and self-reports. Structural equation modeling, including correlated factor and bifactor-(S-1) models, was applied to separate general from specific personality pathology variance. General personality pathology showed strong negative associations with emotion recognition and small-to-moderate negative associations with face processing and crystallized intelligence. Associations with specific personality pathology were substantially attenuated when controlling for general personality pathology. The link between personality functioning and emotion recognition was only minimally accounted for by face processing or crystallized intelligence. These findings suggest that cognitive deficits in PD may primarily reflect severity rather than disorder-specific variance, while the association between personality functioning and emotion recognition appears not to be reducible to face processing or crystallized intelligence. Our study suggests that dimensional models provide a more nuanced framework for understanding cognitive impairment in personality pathology.