Using the Social Responsiveness Scale to Characterize Social Deficits in Children Referred for Aggressive Behavior

Background: The Social Responsiveness Scale (SRS; Constantino & Gruber, 2003) is a measure of social deficits that has been mostly used in children with autism spectrum disorder (ASD). While the measure reflects social deficits (e.g., awareness, motivation, communication), commonly seen in ASD, there is evidence that it may also reflect dimensions that relate to social behavior in children with other psychiatric diagnoses, such as anxiety (Cholemkery et al., 2014, Settipani et al., 2012). Aggressive behavior and irritability is another phenotypic presentation that may be associated with social deficits (Sukhodolsky and Scahill 2012). However, the associations between social deficits as reflected by the SRS and aggressive or disruptive behavior in children without ASD has yet to be examined.

Objectives: The goal of the current study was to examine the relationships between scores on measures related to aggression and SRS scores in typically developing children referred for aggressive behavior relative to matched typically controls without any psychiatric or developmental psychopathology.

Methods: A total of 87 parents completed a battery of measures of disruptive behavior and social function as part of ongoing research study of anger and aggression in children. Fifty-seven children were referred for disruptive behavior (19 female; age M=11.54, SD=2.32) and the remaining 30 were recruited as healthy controls (6 female; M=12.95, SD=2.10). Thus, the sample reflected a wide range of aggression and social behavior profiles. Parents reported on their child’s aggression using the Child Behavior Checklist (CBCL), Inventory of Callous Unemotional Traits (ICU), The Home Situations Questionnaire (HSQ), and the Disruptive Behavior Rating Scale (DBRS). Social abilities were reported using the SRS.

Results: As expected, the TD clinic sample demonstrated higher scores on all aggression-related measures. SRS Total scores (M=66.47/”Mild” range for corresponding T-score), SD=28.40) were also significantly higher within the TD clinic sample than the healthy control subsample (M=22.76/”Normal” range for corresponding T-score, SD=15.43; t[84]=7.72, p<.00).

When examining the full sample, SRS Total scores were positively correlated with several indicators of aggression and disruptive behavior (p<.01): CBCL Aggressive Behavior subscale, r(84)=.62, HSQ Total r(84)=.56, DBRS Total, r(85)=.55, and ICU Total, r(85)=.48. Additionally, via hierarchical regression, age (β=.01, t=-2.54, p=.01) and Total SRS scores (β=.60, t=7.03, p<.01) were shown to significantly predict CBCL Aggressive Behavior; this was not true for gender or SRS subscales.

Conclusions: The results of the current study suggest that typically developing children with aggressive behavior surpass clinical cutoffs on the SRS and that, more generally, SRS scores are correlated with various types of aggression, irritability and noncompliance. This finding indicates that Total SRS scores can be added to clinical evaluation of children who are not on the ASD spectrum in order to characterize social deficits associated with other behavioral problems; as examined in this report, aggression and irritability. SRS subscales may be less useful for such characterization. Past studies have shown some similar relationships in samples of siblings of youth with ASD (Hus et al., 2013) but not ASD samples (Kanne & Mazurek, 2011), further suggesting the utility of differential use of this instrument across populations.