Emotion recognition and cognitive empathy deficits in adolescent offenders revealed by context-sensitive tasks

Abstract

Emotion recognition and empathy abilities require the integration of contextual information in real-life scenarios. Previous reports have explored these domains in adolescent offenders (AOs) but have not used tasks that replicate everyday situations. In this study we included ecological measures with different levels of contextual dependence to evaluate emotion recognition and empathy in AOs relative to non-offenders, controlling for the effect of demographic variables. We also explored the influence of fluid intelligence (FI) and executive functions (EFs) in the prediction of relevant deficits in these domains. Our results showed that AOs exhibit deficits in context-sensitive measures of emotion recognition and cognitive empathy. Difficulties in these tasks were neither explained by demographic variables nor predicted by FI or EFs. However, performance on measures that included simpler stimuli or could be solved by explicit knowledge was either only partially affected by demographic variables or preserved in AOs. These findings indicate that AOs show contextual social-cognition impairments which are relatively independent of basic cognitive functioning and demographic variables.

Keywords: adolescence; contextual processing; delinquency; ecological tasks; offenders; social cognition.

FIGURE 1

Significant group differences in emotion recognition and empathy tasks. (A) Emotional Morphing Task (EMT): mean accuracy and SD of emotion recognition. Note that group differences disappeared after adjusting for age (see Low Context-Sensitive Measure: Facial Emotion Recognition). (B) Awareness of Social Inference Test (TASIT): mean accuracy and SD for global score. (C) Dual Valence Association Task (DVAT): mean accuracy and SD for incongruent blocks. (D) Empathy for Pain Task (EPT): cognitive components (percentage and SD for intentionality question).

FIGURE 2

Multiple regression analyses. (A) Regression analysis using global score from TASIT as the dependent variable. (B) Regression analysis using the mean accuracy in incongruent blocks from the DVAT as the dependent variable. (C) Regression analysis using as dependent variable the mean percentage of correct responses in intentional, accidental, and neutral scenarios of cognitive aspect (intentionality) of the EPT.