The Autism Preschool Peer Interaction Observation Scale (APIOS): Naturalistic Observation to Assess Social-Communicative Skills during Ongoing Interaction with Peers

Background: Peer relations in typical development are cardinal for children's development of ample cognitive, linguistic, and social skills (e.g., Hay, Caplan, & Nask, 2009). Longitudinal evidence shows that individual variations in behavior and in responding to peers’ behavior at early ages predict later social competence, even when accounting for factors such as temperament and attachment relations (e.g., Hay et al., 2009). Peer relations do not develop typically in children with autism spectrum disorder (ASD), thus limiting peer relationship experiences in young children with ASD and forming the basis for reduced peer engagement across development (Manning & Wainwright, 2010). Yet, naturalistic observation procedures to reliably evaluate spontaneous peer engagement are scarce, and those that are developmentally oriented at the preschool years are even more limited.

Objectives: The major aim of the current study was to examine the validity and reliability of a newly developed peer observation scale, the Autism Peer Interaction Observation Scale (APIOS), in (a) differentiating between preschoolers with ASD and typical age-mates regarding their social functioning during real-life peer interactions in and out of natural classroom environments; and (b) linking the APIOS with frequently used standard reports to assess social-adaptive functioning (Vineland), social dysfunction (SRS-2), autism severity (ADOS), and IQ (Mullen).

Methods: Eighty-five preschoolers (age 3-6 years) participated in the study: n=50 individually included high-functioning children with ASD (HFASD; IQ>75) and n=35 typically developing age-mates matched on chronological age, IQ, and mothers’ education. The theory-based APIOS (Shefer & Bauminger-Zviely, 2016) comprises a 3-hour observational procedure assessing preschoolers’ social-communication abilities and difficulties during spontaneous peer interactions during various activities in and out of preschool. The APIOS includes 9 categories and 17 subcategories for adaptive social functioning (nonverbal communication, functional and complex social behaviors, prosocial behaviors, social and imaginary play, conversation, range and quality of facial expressions) and another 3 categories for non-adaptive behaviors (aggressive, stereotyped-repetitive, and irregular sensory interest), rated by the observer along a 4-point scale from 1 (typical) to 4 (very atypical behavior). APIOS inter-rater reliability, established for approximately 20% of preschool observations, ranged from 85.7% to 96.4%. To evaluate adaptive skills and social functioning, the Vineland and SRS-2 were completed by children's teachers.

Results: Main results verified the APIOS’s distinctive value in differentiating HFASD from typical groups, with the latter showing more adaptive socio-communication and fewer non-adaptive behaviors. Based on the APIOS, we also clustered behaviors according to severity of socio-communicative deficit. In HFASD, the most deficient behaviors were imaginary play and conversation, and the more intact behaviors were nonverbal behavior and functional social communication. The APIOS correlated well with the Vineland, SRS-2, and IQ: Children observed to demonstrate more intact peer relations on the APIOS showed higher IQ, better adaptive skills, and fewer social atypicalities. In contrast, only a few significant correlations emerged between the APIOS and ADOS.

Conclusions: We discuss the importance of developing appropriate naturalistic peer-engagement observation procedures to obtain comprehensive assessment of social-communication in HFASD and also to provide intervention guidelines.