Context Matters: The Measure of Emotion Regulation in Autism

Background:  Emotion regulation (ER) development in autism spectrum disorder (ASD) is not clearly documented, despite frequent reference across literature and assessment measures.  ER is a reported difficulty in ASD, though is currently not a core target of intervention objectives (Mazefsky Pelphrey, & Dahl, 2012), which may be reflective of limited assessment tools that adequately capture ER skills across multiple contexts.  The modal modeltheory presents the following steps in ER:  1) situation selection; 2) situation modification; 3) attention deployment; 4) cognitive change; 5) and response modulation (Gross & Thompson, 2007).  The modal model provides a logical framework within which ER may be assessed in young children.

Objectives:  Three research goals are examined:  1) the development of an observational measure based on the modal model of Gross & Thompson, the Emotion Regulation Assessment for Autism (ERAA); 2) the comparison of ERAA ratings across diagnostic groups; and 3) the examination of ER difficulties as they relate to ASD across different play contexts and partners.

Methods:  Thirty-one children diagnosed with ASD and 31 children with typical development were recruited.  Parents completed an Infant-Toddler Social and Emotional Assessment (ITSEA; Briggs-Gowan & Carter, 2000) and Behavior Rating Inventory of Executive Function – Preschool Version (BRIEF-P; Gioia, Espy, & Isquith, 2003).  All children were administered a Mullen Scales of Early Learning (Mullen, 1995), and children with ASD were administered the Autism Diagnostic Observation Schedule, Second Edition (Lord et al., 2013) and the Childhood Autism Rating Scale, Second Edition (CARS2; Schopler, Van Bourgondien, Wellman, & Love, 2010).  Two 15-minute play samples were conducted with a parent and unfamiliar adult.  Play samples were blind-coded using the ERAA and SAP-O (SCERTS model; Prizant, Wetherby, Rubin, & Laurent, 2003).  ER symptoms were analyzed across play context (joint interactive play, independent play, disruptive play) and play partner (parents and strangers).

Results:  The ERAA produced near-perfect reliability, with Cohen’s kappa values resting at .81. The ERAA significantly distinguished children with typical development from children with ASD, t(60) = 7.20, p<.001.  A significant correlation between the ERAA and ASD symptoms was also found, r(31) = .51, p<.01, indicating that more children with more autism symptoms demonstrated more difficulty regulating emotion.  A significant three-way interaction was found among play context, partner, and diagnosis, with independent play with a stranger resulting in more ER difficulties.  However, joint-interactive and disruptive play resulted in more discrepant scores between diagnostic groups, despite the overall lower ERAA score. A significant relationship was not found between the ERAA and the ITSEA, suggesting the measurement of different constructs.  Executive function skills did not contribute significantly to ER difficulties.  

Conclusions:  Implications for the current study suggest that emotion regulation symptoms are separate and unique from ASD symptoms. Data suggest that independent play should be further explored in ASD, while disruptive interactions may not be as troublesome as previously reported. Given the results of the current study, the ERAA is a promising tool for assessing emotion regulation in young children and may aid in intervention planning.