Neurofeedback training of executive function in autism spectrum disorder: distinct effects on brain activity levels and compensatory connectivity changes

Abstract

Background: Deficits in executive function (EF) are consistently reported in autism spectrum disorders (ASD). Tailored cognitive training tools, such as neurofeedback, focused on executive function enhancement might have a significant impact on the daily life functioning of individuals with ASD. We report the first real-time fMRI neurofeedback (rt-fMRI NF) study targeting the left dorsolateral prefrontal cortex (DLPFC) in ASD.

Methods: Thirteen individuals with autism without intellectual disability and seventeen neurotypical individuals completed a rt-fMRI working memory NF paradigm, consisting of subvocal backward recitation of self-generated numeric sequences. We performed a region-of-interest analysis of the DLPFC, whole-brain comparisons between groups and, DLPFC-based functional connectivity.

Results: The ASD and control groups were able to modulate DLPFC activity in 84% and 98% of the runs. Activity in the target region was persistently lower in the ASD group, particularly in runs without neurofeedback. Moreover, the ASD group showed lower activity in premotor/motor areas during pre-neurofeedback run than controls, but not in transfer runs, where it was seemingly balanced by higher connectivity between the DLPFC and the motor cortex. Group comparison in the transfer run also showed significant differences in DLPFC-based connectivity between groups, including higher connectivity with areas integrated into the multidemand network (MDN) and the visual cortex.

Conclusions: Neurofeedback seems to induce a higher between-group similarity of the whole-brain activity levels (including the target ROI) which might be promoted by changes in connectivity between the DLPFC and both high and low-level areas, including motor, visual and MDN regions.

Keywords: Autism spectrum disorders; Dorsolateral prefrontal cortex; Neurofeedback; Rt-fMRI.

Fig. 1

Schematic representation of the neurofeedback protocol

Fig. 2

Probability map of the target ROI selected online in the DLPFC for all participants (maximum probability = 44.82%)

Fig. 3

Mean T value across subjects in the target ROI (DLPFC) in neurofeedback runs for ASD and neurotypical individuals (NT). Points are laid over a 1.96 standard error of the mean (SEM) (95% confidence interval) in lighter color and a 1 standard deviation (SD) in darker color

Fig. 4

Scatterplot showing the relation between ADOS-2 (total score) and DLPFC in ASD group

Fig. 5

Control group RFX GLM map for all neurofeedback runs during ‘Imagery’ condition. Sagittal (x=-1, talairach coordinate) and axial (z=17 for left image and z=42 for right image, talairach coordinates) views, showing the typical pattern of activation during executive function tasks, including DLPFC, intraparietal sulcus (IPS), supplementary motor area (SMA), anterior insula, premotor cortex, inferior frontal gyrus (IFG) and basal ganglia, with deactivation of DMN (precuneus, medial frontal area and lateral parietal/angular gyrus)

Fig. 6

Whole-brain RFX GLM map comparing groups (ASD>NT) in pre-neurofeedback (left) and first neurofeedback conditions (right) during the ‘imagery’ condition. Coronal (left top, y=13, talairach coordinate) and axial (left bottom, z=55, talairach coordinate) views

Fig. 7

Seed-based (DLPFC) connectivity analysis, within-group comparisons transfer> pre-neurofeedback for ASD (A) and control groups (B). A–Cluster of hyperconnectivity with DLPFC in the transfer run compared to the pre-neurofeedback run located in the right pre and postcentral gyri. B-C luster of hypoconnectivity with the DLPFC in the transfer run compared to the pre-neurofeedback run located bilaterally in the parahippocampal gyri (here represented by the medial view of the left hemisphere) in the control group

Fig. 8

Seed-based (DLPFC) connectivity analysis, within-group comparisons first neurofeedback> pre-neurofeedback in control group. Clusters of hyperconnectivity with the DLPFC in the first neurofeedback run comparing to pre-neurofeedback run located in the left superior frontal gyrus, left ventral mPFC and left postcentral gyrus in the right superior view (left) and in the right SMA in the right medial view (right)

Fig. 9

Seed-based (DLPFC) connectivity analysis between group comparison (ASD>NT) during the transfer run (‘Imagery’>’Baseline’). Clusters of hyperconnectivity with the DLPFC in the left middle frontal gyrus, left middle temporal gyrus, left ventral mPFC and right occipital pole