Back

Intellectual disability and cerebellar hypoplasia in autism reported by associative learning

Welsh, J. P.; Munson, J.; St. John, T.; Meehan, C. N.; Tran, E.; Begay, K. K.; Dager, S. R.; Estes, A. M.

2021-03-28 neurology
10.1101/2021.03.26.21254368 medRxiv
Show abstract

ObjectiveTo determine how impairments in associative learning in autism spectrum disorder (ASD) relate to intellectual disability (ID) and early-childhood cerebellar hypoplasia. MethodsTrace and long-delay eye blink conditioning (EBC) were performed in 62 children age 11.2 years having: 1) ASD with ID (ASD+ID); 2) ASD without ID (ASD-noID); or 3) typical development (TD). The sub-second timing of conditioned eye-blink responses (CRs) acquired to a tone paired with a corneal air puff was related to brain structure at age 2 years and clinical measures across ages 2-12 years. Because CR timing is influenced strongly by cerebellar function, EBC was used to test hypotheses relating cerebellar hypoplasia to ASD. ResultsChildren with ASD+ID showed early-onset CRs during trace EBC that were related to early-childhood hypoplasia of the cerebellum but not of the cerebral cortex, hippocampus, or amygdala. Children with ASD-noID showed early-onset CRs only during long-delay EBC without cerebellar hypoplasia. Using EBC measures, logistic regression detected ASD with 81% sensitivity and 79% specificity while linear discriminant analysis separated ASD subgroups based on ID but not ASD severity. MRI of additional 2-year-olds with ASD indicated that early-onset CRs during trace EBC revealed ASD+ID more readily than cerebellar hypoplasia, per se. ConclusionsEarly-childhood cerebellar hypoplasia occurs in children with ASD+ID that demonstrate early-onset CRs during trace EBC. Trace EBC reveals the relationship between cerebellar hypoplasia and ASD+ID likely by engaging cerebro-cerebellar circuits involved in intellect. We emphasize that the cerebellum optimizes sensory-motor processing at sub-second intervals, impairments of which may contribute to ID.

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.