Empathic Behavior in Children and Adolescents with High-Functioning ASD

Background:  

Autism has been described as an empathy disorder. A lack of empathy is emphasized in diagnostic guidelines and clinical observations of autism spectrum disorder (ASD). Yet, surprisingly few studies have directly looked at the empathic behavior of children with ASD during real life interactions. The few studies that did, suggest a reduced attention to and concern for others’ emotions in children with ASD compared to controls (e.g. Sigman, Kasari, Kwon, & Yirmiya, 1992). It is unclear whether these findings can be generalized to the behavior of school-aged children and adolescents with ASD but normal intelligence (high-functioning ASD; HFASD).

Objectives:  

This is the first large-scale study to systematically and directly examine the empathic behavior of children and adolescents with HFASD and typically developing (TD) peers.

Methods:  

We compared the empathic behavior of 203 school-aged children and adolescents with HFASD and 65 TD peers (6-18 years) using: (1) direct observations of participants’ responses to affective states (happiness, sadness, and pain) displayed by an adult interviewer, and (2) parent reports of their child’s anticipated responses in comparable situations. All responses were coded by independent coders into five response categories (kappa’s ranging from .68 to 1.00). An empathic response was defined as a relevant verbal response including an empathic reference to the other’s emotional state (e.g., ‘Are you all right?’).

Results:  

A k-means cluster analysis on the five response categories was performed to create different clusters of participants with homogeneous response patterns, including a cluster of participants that responded mostly empathically. Independent of group status, significantly more participants of 13 years or older (36%) were assigned to the empathic cluster based on their directly observed behavior compared to participants younger than 13 years (16%) (χ²(3) = 11.40, p =.01). Surprisingly, an equal proportion of participants from the HFASD and TD group (29% in HFASD vs. 26% in TD) were assigned to the empathic cluster based on their responses to the interviewer’s affective states (χ²(3) = 1.92, p > .10). When based on parent reports, significantly fewer children with HFASD were classified in the empathic response cluster (35%) than children with a typical development (67%) (χ²(3) = 18.56, p <.001). 

Conclusions:  

The findings suggest empathic behavior continues to develop throughout adolescence, both in typically developing children and children with HFASD. Importantly, children and adolescents with HFASD appear equally (in)capable as typically developing peers to respond empathically during a standardized observation, but are less likely to give empathic responses according to parents. Hence, a child’s (lack of) empathic behavior during a social interaction with an unknown adult does not indicate or refute an ASD diagnosis. However, a marked absence of empathic responses as observed by parents appears to point to an atypical, and possibly autistic, development.