Changes in Brain Activation and Connectivity in Children with Autism before and after a Visualization Language Intervention

Background:  Deficits in language comprehension have been widely reported in children with autism spectrum disorders (ASD). Recent evidence from neuroimaging research suggests that people with ASD tend to recruit visuospatial imagery to comprehend language (Kana et al., 2006). Such increased reliance on visuospatial regions may also underlie weaker coordination between primary visual and association areas. 

Objectives:  This fMRI study investigates the brain activation and connectivity patterns in ASD during language comprehension. The main objective was to determine the impact of a 10-week visualizing language intervention on improving functional activation and connectivity during a visual imagery language task in children with ASD.

Methods: 32 children with ASD (ages 8-13 years) were imaged twice 10 weeks apart, with 17 children receiving the intervention soon after the initial fMRI scan, and 15 wait-list control children. In addition, 27 typically-developing (TD) children, age and IQ-matched, were imaged once. A language comprehension task, presented while the children underwent fMRI, included high and low imagery sentences. The data were used to investigate intervention related effects on brain activation and connectivity. Data were acquired from a Siemens 3.0T Allegra head-only scanner and analyzed using SPM8. 

Results:  (1) Between-group analysis of the ASD groups at their pre-imaging session and typically-developing (TD) controls revealed differences in high-imagery sentences, with the ASD group recruiting Wernicke's area (BA 40) and right precentral and right lingual gyri; (2) Specific differences were found in the visual word form area of the mid-fusiform region, with the ASD group failing to activate this area pre-imaging session; (3) Children with ASD who received the intervention between imaging sessions showed increased activation in the right middle occipital gyrus and lingual gyrus pre- to post-imaging session. There were no significant pre- to post-imaging session differences in the wait-list control group; (4) Between-group results revealed greater activation in the intervention group as compared to the wait-list control group in both language regions and frontal regions (p<0.001, cluster corrected); (5) A functional connectivity analysis revealed greater synchronization between the medial prefrontal gyrus and a number of language and visual association brain regions in the children with ASD who received the intervention compared to the wait-list controls.

Conclusions:  This study introduces a novel and intensive language-based remediation treatment that is designed to use nonverbal sensory input, an area relatively intact in individuals with ASD, in order to develop oral and written language comprehension, establish vocabulary, and develop higher order thinking skills. The results of this study reveal that children with an ASD who received the V/V intervention show increased activity in visual language association regions post-intervention. In addition, when compared to our wait-list control group, the children with ASD who received the intervention showed greater activity in language and frontal regions and greater connectivity between frontal and more posterior language and visual areas. Our preliminary findings reveal the plasticity of the brain in children with autism, and suggest improvement of neural activity and synchronization due to targeted intervention.