Contribution of Severity of Autism and Age on Dog Attraction to Children with Autism Spectrum Disorder

Background: Positive effects of pets at home on children with autism spectrum disorders (ASD) are well known (Grandgeorge et al., 2012; Viau et al., 2010). These benefits are linked to the strength of the bond established between the child and the animal (Carliste, 2014; 2015). However, we do not know what are the antecedents or variables that would facilitate or limit the effects of the dog's assignment at home.

Objectives: The present research aims to identify indicators of attraction for dogs observed in ASD children before introducing the animal into the family, and to determine if and how this attraction is influenced by factors such as severity of ASD, sex and age.

Methods: This project was realized in collaboration with Mira Foundation, a non-profit organization involved in the training and donation of assistance dogs since 1987. All participating families have an ASD child and applied to receive a dog from the Foundation. As a first step in the evaluation of families’ application by the Foundation, all families must participate to a videorecorded assessment session, in which child and parents interact freely with a trained dog for approximately 20 minutes. Those videorecording are then coded using two evaluation instruments, (1) the Childhood Autism Rating Scale (CARS; Schopler et al., 1988) and (2) an observation grid assessing 12 behaviors related to the attraction for the dog (grid created by MIRA’s team and research collaboration). A total of 748 ASD children (17.5% of girls) and their families were involved in the study. ASD children were aged between two and sixteen years old (Mean age: 7.2 ±3.4) and were diagnosed with ASD by independent teams of physicians and professionals based on DSM-IV criteria.

Results: Preliminary results highlighted the reliability of the two instruments (Cronbach alpha: .90 for each). Factor analyzes with varimax rotation identified 2 main factors (53,4% of the total variance) for the CARS associated with: (1) the deficits in social communication, (2) the child emotional reactivity (anxiety, emotion, adaptation). This result partially replicated the factor structure reported in previous researches (DiLalla & Rogers, 1994; Park & Kim, 2015). Furthermore, analysis on the attraction grid revealed two factors (60,5% of the total variance) associated with: (1) the child’s behaviors toward the dog (look, proximity, contact, manipulation), (2) the type of interactions expressed in presence of the dog (facial, verbal, playful). When crossing the two evaluation instruments, our main findings revealed a strong correlation between social communication (CARS) and physical engagement of the child (attraction grid) (r= .43 **). Sex did not have any influence on these covariations. In addition, child’s age was significantly related to the first factor of attraction for the dog (r= - .33 **). ASD severity and child’s age had an independent influence on the attraction toward dogs (all p<0.05).

Conclusions: Our study may provide relevant information for predicting which ASD children could be attracted by dogs and would benefit the most from the dog assignment at home.