Social relatedness disturbances and facial expression recognition deficits in children with reactive attachment disorder

The purpose of this study was to explore and examine the relationship between social relatedness disturbances and facial expression recognition among an outpatient clinic sample (N = 34) of children (aged 6--19 years) with Reactive Attachment Disorder of Childhood and Infancy (RAD). Specifically, this study investigated whether, and to what extent, facial expression recognition was associated with the two dimensions of social relatedness disturbances in RAD (i.e., indiscriminate/disinhibited behaviors and withdrawn/inhibited behaviors). Numeric scores for social relatedness dimensions, indiscriminate/disinhibited (DIS) and withdrawn/inhibited (INH) were explanatory variables and facial expression recognition error and attribution bias scores were response variables in multiple regressions analyses. Analyses results revealed partial support for two hypotheses and no support for a third hypothesis.;Based on attachment theory, attachment studies, and facial expression recognition studies, the following hypotheses were generated: (1) DIS dimension scores would be positively associated with recognition errors for sad, angry, and fearful facial expressions, (2) INH dimension scores would be associated with attribution bias scores for the recognition of sad and fearful in other facial expressions, (3) Associations found for Hypotheses 1 and 2 would be stronger for facial expressions displayed with low intensity than for those displayed with high intensity. Partial support for hypotheses was found in statistically significant linear regression models for a positive association between DIS dimension scores and recognition errors for sad in child facial expressions, and for a positive association between age and INH dimensions scores and attribution bias scores for the recognition of sad in other facial expressions. No support was found for Hypothesis 3. An unexpected finding was a positive association between INH dimension scores and attribution bias scores for the recognition of happy in other facial expressions. The current study also found evidence that adds to the previous research findings that RAD subtypes do not appear to be mutually exclusive or independently distinct subtypes. Results of this study add to the small body of empirical findings that will enable researchers and clinicians to understand more about individual differences within RAD.