More Evidence that Behavioral Therapy for ASD Affects Brain Function

Pam Ventola, PhD
Yale University

This week we published a paper on the brain-based responses to a behavioral treatment for ASD, Pivotal Response Treatment (PRT). We set out to complete this project because in many areas in this country, not to mention around the world, there is a dearth of empirically validated treatment options for individuals with ASD. The literature on effective treatments for ASD is lagging behind work in other areas of the disorder; by design, treatment work is costly and time-consuming. It can take months or years to collect full data on a single child, depending on the length of the treatment. Furthermore, very few studies have sought to understand the mechanisms of treatment response, which is needed to understand how treatment works and to develop even more effective treatments that are tailored to individuals with distinct profiles.

Another thing that makes treatment research difficult is that no two cases of autism are exactly the same. The differences in how each case of autism shows itself as well as how they are biologically different has baffled autism scientists for years. In addition to treating ASD, we were assessing brain function in individuals in our study. Our brain-based data reinforced the notion of ASD as a heterogeneous disorder, but the magnitude of the heterogeneity was surprising even to us.

For the study, children completed a functional MRI (fMRI) before and after a four-month treatment course of pivotal response training. This is a behavioral therapy which is based in ABA and shown to improve some of the core symptoms of ASD. It is play based and targets critical, or “pivotal” areas like motivation, initiation of social interactions, and self management. The approach utilizes the motivations unique to each child to provide opportunities for skill acquisition within naturalistic play-based interactions. The fMRI allowed us to collect a measure of how the children processed social information and gave us images of which areas of the brain lit up, or were activated, during each task. This way, we were able to measure the strength of their brain-based responses to social information and the specific locations in their brains that responded. We found, before treatment even started, that two groups of children emerged. One showed increased activation, the other showed reduced activiation in the same brain are. This brain area is called the posterior superior temporal sulcus. This finding alone demonstrates biologically based subgroups of children with ASD, and suggests that these subgroups may reflect distinct biotypes (subgroups of children with a common aspect of their genotype).


Following the four-month treatment course of PRT, children again completed an fMRI. Despite their differences in brain functioning, all of the children gained meaningful skills and showed improvements in adaptive, social and communication skills. Given that they showed different brain responses before treatment, this was somewhat surprising. What was the most exciting is that the children exhibited changes in brain functioning following the treatment. The treatment seemed to exert its effect in different ways depending on the child’s baseline brain-based profile. For example, the children who exhibited hyper-activation (too much activation) in the pSTS at the start showed a decrease in activation following treatment. The children who exhibited hypo-activation (too little activation) at the start showed an increase in activation in the pSTS following the treatment. This is the first work to demonstrate how brain-based responses to treatment may differ between groups of children with ASD. In this study, individuals with autism showed differences in biological markers. Yet behavioral based intervention showed improvements in both groups, reinforcing that individuals with ASD, despite how their brain is wired, can benefit from behavioral therapy.

This project, as a whole, is in its preliminary stages with this study representing our earliest neuroimaging data. Nonetheless, work such as this is crucial for children with ASD, their families, and care providers. As previously stated, understanding how and why treatments work is critical to bettering treatments and tailoring them to specific individuals. Specifically, this understanding will guide the refinement of existing treatments, inform the development of novel interventions, determine early efficacy indicators, and lead to the development of algorithms that predict which treatment will likely benefit a given person.

%d bloggers like this: