Differences in Temperament between Children with Autism Spectrum Disorder and Typical Development in the Study to Explore Early Development

Background: The Behavioral Style Questionnaire (BSQ) is a widely used temperament measure in children with autism spectrum disorder (ASD)^1,2. Recent evidence suggests that the BSQ’s original factor structure does not best describe diverse samples of children with ASD or typical development.³ Instead, children with ASD and typical development may share an alternative 9-factor temperament structure measuring (1) Maladaptivity (e.g., “bothered by plan changes”), (2) Environmental Sensitivity (e.g., “sensitive to noises”), (3) Quiet Persistence (“practices [to mastery]”), (4) Social Inattention (e.g., “does not acknowledge [when called]”), (5) Social Approach (e.g., “approaches [unknown children]”), (6) Activity (e.g., “[frequently] runs”), (7) Crying, (8) Rhythmicity (e.g., “hungry at dinner”), and (9) Food Openness (e.g., “tries new foods”). To date, there are no data published on these alternative temperament factors comparing children with ASD to children in a population-comparison group.

Objectives:

This study compared children with ASD and children in a population-comparison group (POP) on the alternative 9-factor BSQ temperament structure.

Methods: Study data were collected in the Study to Explore Early Development-Phase I (SEED1). SEED1 is a case-control study funded by the Centers for Disease Control and Prevention and conducted at multiple sites throughout the U.S. (California, Colorado, Georgia, Maryland, North Carolina, and Pennsylvania).⁴ SEED participants were recruited through developmental disability service organizations (ASD) or a sample of state vital records (POP). Children were aged 2-5 years at the time of study enrollment. Clinicians confirmed ASD status with gold-standard diagnostic instruments administered during an in-person evaluation of the child. BSQ data were obtained via a self-administered questionnaire completed by a caregiver.

Revised BSQ (BSQ–R) scales were created by summing items shared among ASD and POP children in exploratory factor analysis (i.e., both groups shared items with factor loadings ≥ .40): Maladaptivity (k [items]=4), Environmental Sensitivity (k=7), Quiet Persistence (k=5), Social Inattention (k=5), Social Approach (k=5), Activity (k=4), Crying (k=5), Rhythmicity (k=2), and Food Openness (k=2). Higher scores on Maladaptivity, Social Inattention, Crying, and Activity indicate potential behavioral concerns, while lower scores on Environmental Sensitivity, Quiet Persistence, Social Approach, Rhythmicity, and Food Openness indicate potential concerns. Consideration of scale residuals suggested that the following BSQ-R scales were non-parametrically distributed: Maladaptivity, Quiet Persistence, Social Approach, Crying, Rhythmicity and Food Openness. T-tests were used to compare ASD and POP children for parametric data; Mann-Whitney tests were used to compare ASD and POP children for non-parametric data.

Results:

Children with ASD scored higher than POP children on Maladaptivity, U=19.18, p<.001; Social Inattention, U=13.67, p<.001; and Crying, U=5.77, p<.001. Children with ASD scored lower than POP children on Environmental Sensitivity, t(1221)=10.45, p<.001; Quiet Persistence, U=15.16, p<.001; Social Approach, t(1164)=23.71, p<.001; Rhythmicity, U=6.53, p<.001; and Food Openness, U=13.67, p<.001. There were no differences on Activity.

Conclusions:

These results show that the BSQ-R scales may distinguish ASD from POP children on eight of the nine subscales. Specifically, children with ASD had more problems than POP children on all temperament domains except Activity. The BSQ-R scales may be useful for developing metrics of previously unstudied phenotypic factors in children with ASD.