Using the NIH Toolbox Cognition Battery to Characterize the Cognitive Profile of Children with Autism

Background: The NIH Toolbox Cognition Battery (NIH-TCB) is a clinically validated set of digital assessments for the assessment of several cognitive domains. While preliminary studies have implemented this measure in Intellectual Disability (ID), Fragile X syndrome, and Down syndrome (Hessl et al., 2016), to our knowledge, it has not been tested on children with autism spectrum disorder (ASD). Children with ASD are believed to possess an altered cognitive profile compared to typically developing children, characterized by executive dysfunction in several areas, including speed of processing, attention shifting, and cognitive flexibility, as well as increased detail orientation (Kenworthy et al., 2008).

Objectives: This study details preliminary data using the NIH-TCB to assess the cognitive domains of cognitive flexibility, processing speed, and attention shifting in children with and without ASD.

Methods: ASD children ages 3-17 and typically developing (TD) children matched on age are currently being recruited. This abstract details data from 18 children and adolescents with ASD (mean age = 9.4 years; SD = 3.9) and 14 TD controls (mean age = 7.1 years; SD = 2.8). TD children demonstrated a significantly higher IQ (mean IQ = 117.4; p=.01) than ASD children (mean IQ = 100.2), so this was controlled for in analysis. Children were administered an abbreviated version of the Stanford-Binet Intelligence Test, Fifth Edition (SB-5) and three NIH Toolbox Cognition assessments: 1) Flanker Inhibitory Control and Attention (Flanker); 2) Dimensional Change Card Sort (DCCS); and 3) Pattern Comparison Processing Speed (PCPS). These tasks were chosen in order to evaluate attention, cognitive flexibility, and speed of processing.

Results: Three ASD children were excluded from analysis, as they were unable to complete the assessment due to difficulties in attention or understanding the task. These children were all characterized by an IQ at or below the 5^th percentile, although a select few children with an IQ in this range completed the assessment. ASD children exhibited cognitive deficits compared to TD children, with lower Age-Corrected Standard scores for the Flanker (t(27)= -2.9, p= .007), DCCS (t(27)= -3.6, p= .001), and PCPS (t(27)= -3.2, p= .004) assessments. IQ was strongly correlated with Age-Corrected Standard score for ASD (r= .69), but not TD children (r= .46) on the DCCS task. No significant correlations were found between IQ and Age-Corrected Standard score for the Flanker or PCPS tasks in either group. When controlling for IQ on the DCCS task, significant between-group differences continued to be observed in Age-Corrected Standard scores (F= 4.3, p= .047), supporting previous findings using the standard DCCS task in high-functioning autism (Faja & Dawson, 2014).

Conclusions: These preliminary findings replicate previously observed patterns of cognition in children with ASD. Children with ASD exhibited decreased cognitive flexibility, difficulties in attention switching and inhibition, and increased speed of processing compared to TD children. Although the current sample size is small, we anticipate 150 children with ASD and 75 controls by Spring 2018. These data support the potential of the NIH-TCB as a valid, integrated tool for the assessment of cognition in children with ASD.