Impaired emotional processing and clinical psychopathology: a case control study using a hierarchical bifactor model

Introduction. Emotional processing (EP) represents a change from an unresolved emotional state to a resolved one. This transition successfully occurs for most people. However, failures can happen at the facet level, including 1) the tendency to avoid emotions, 2) the suppression of emotions, 3) the inability to control the expression of emotions, and 4) an impaired emotional experience. Thus, people who persist in an unresolved state, show 5) signs of unprocessed emotions, impeding other experiences and behaviors to proceed without disruption. With the widespread use of the Emotional Processing Scale (EPS), there is evidence that mental disorders are associated with EP failures, as shown by differences between healthy and clinical samples. Because these failures can be represented by a hierarchical arrangement of facets with one broad factor on the top, we propose that this general factor, representing a global impairment in EP, might be a transdiagnostic vulnerability factor. Methods: We administered the EPS to a mixed psychiatric sample (n=183) and a control group (n=231). The patients were classified as psychotics or bipolar disorders, depressed, affective disorders, and personality disorders. A bifactor model was used to obtain hierarchical factor scores subsequently used to compare patients to healthy controls and specific patient groups. Results: Model’s fit was acceptable (CFI=.92, TLI=.90, RMSEA=.05, SRMR=.06) and the general factor accounted for was 65% of the variance. Using the summated ratings, the psychiatric sample obtained systematically higher scores than healthy controls on the five facets and the total score. Conversely, using factor scores there were significant differences in the general factor only. Comparing patient groups, no differences were found using summated ratings, while we found marginally significant differences at the facet level (especially 2 and 3). Conclusions. Our findings are consistent with the transdiagnostic hypothesis.