Assessment of Intelligence: A Comparison of Intellectual and Adaptive Functioning in Preschool Autistic Children

Background: DSM-5 autism diagnosis must specify whether there is an accompanying intellectual deficiency, but few intellectual tests are available for children under age 3. Available tests to assess cognitive development mostly consist of developmental scales that may not accurately represent the potential of young autistic children, whose developmental paths are highly atypical. Indeed, school-aged autistic children perform differently depending on the test used to assess their intelligence (Dawson et al., 2007) and present a discrepancy between intelligence and adaptive level (Charman et al., 2011). Typically used tests in preschool years are Vineland Adaptive Behavior Scales (VABS) and Mullen Scales of Early Learning (MSEL). The differences and resemblances in the performance of ASD preschoolers on those two tests is not well documented.

Objectives: 1) To compare performance of young autistic (ASD), developmentally delayed (DD) and typically developing (TD) children on MSEL and VABS, respectively used as indicators of IQ and adaptive level in young autistic children. 2) To examine the difference between the similar subscales in VABS and MSEL in each group (e.g. receptive language, expressive language and fine motor subscales). 3) To explore the MSEL subscale profile in each group.

Methods: Recruitment is ongoing, and the goal is to reach groups of 40 ASD, 40 DD and 40 TD. To date, the sample includes 29 ASD, 12 DD and 31 TD children aged 18 months to 5 years old. No exclusion criteria were applied in the clinical groups.

Results: MSEL global score was significantly higher than VABS global score in autistic (p<.001) and DD (p<.01) children, but not in TD children (p=.211). As expected, ASD and DD children performed significantly lower than typical children on both tests (p<.001). On expressive language subscales, performance on MSEL was significantly lower than performance on VABS in ASD (p<.001) and DD children (p<.05), but not in TD children (p=.573). On receptive language subscales, there was a main effect of Test (p<.01), VABS score being significantly higher than MSEL score. There was also a main effect of the group, ASD and DD children scoring lower in receptive language than typical children (p< .001). Finally, ASD (p<.001) and DD (p<.05) children performed significantly lower on MSEL subscale assessing fine motor abilities than on the corresponding VABS subscale. In MSEL, visual reception scores were significantly higher than the three others in the ASD group, while there was no significant difference among subscales in the TD group. 

Conclusions: These preliminary results indicate that intellectual potential can be judged very differently depending on the type of test used in preschool children with developmental delays and/or ASD. Clinical groups performed better on MSEL compared to VABS global scores, but the reverse when comparing similar subscales from the two tests (receptive and expressive language, and fine motor abilities). This suggests that the information reported by the caregiver may differ substantially from performance in a formal evaluation setting. Hence, the two assessment methods (self-reported vs. observed) are both important and should be interpreted with prudence, especially in preschool children with atypical development.