Altered neural flexibility in children with attention-deficit/hyperactivity disorder

Abstract

Attention-deficit/hyperactivity disorder (ADHD) is one of the most common neurodevelopmental disorders of childhood, and is often characterized by altered executive functioning. Executive function has been found to be supported by flexibility in dynamic brain reconfiguration. Thus, we applied multilayer community detection to resting-state fMRI data in 180 children with ADHD and 180 typically developing children (TDC) to identify alterations in dynamic brain reconfiguration in children with ADHD. We specifically evaluated MR derived neural flexibility, which is thought to underlie cognitive flexibility, or the ability to selectively switch between mental processes. Significantly decreased neural flexibility was observed in the ADHD group at both the whole brain (raw p = 0.0005) and sub-network levels (p < 0.05, FDR corrected), particularly for the default mode network, attention-related networks, executive function-related networks, and primary networks. Furthermore, the subjects with ADHD who received medication exhibited significantly increased neural flexibility (p = 0.025, FDR corrected) when compared to subjects with ADHD who were medication naïve, and their neural flexibility was not statistically different from the TDC group (p = 0.74, FDR corrected). Finally, regional neural flexibility was capable of differentiating ADHD from TDC (Accuracy: 77% for tenfold cross-validation, 74.46% for independent test) and of predicting ADHD severity using clinical measures of symptom severity (R²: 0.2794 for tenfold cross-validation, 0.156 for independent test). In conclusion, the present study found that neural flexibility is altered in children with ADHD and demonstrated the potential clinical utility of neural flexibility to identify children with ADHD, as well as to monitor treatment responses and disease severity.

Fig. 1. Illustration of a multilayer network.

Top panel: in the multilayer network representation of temporal data, each node is connected to itself in adjacent contiguous windows. Next, each node is assigned to a functional community, represented by different colors. Bottom panel: representative correlation matrices of each sliding window.

Fig. 2. Alteration of neural flexibility in ADHD.

a A boxplot shows significantly decreased whole brain neural flexibility in subjects with ADHD as compared to TDC. b Comparisons of neural flexibility of different functional networks. Black asterisks indicate significant differences after FDR correction. SH sensorimotor hand, SM sensorimotor mouth, CO cingulo-opercular, AUD auditory, DMN default mode, MEM memory retrieval, VIS visual, FP frontoparietal, SAL salience, SUB subcortical, VA ventral attention, DA dorsal attention, CB cerebellar, UC uncertain system. Statistical significance levels: *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 3. Influence of medication on neural flexibility in ADHD.

A boxplot shows significantly increased whole brain neural flexibility in subjects with ADHD who are receiving treatment with medication as compared to unmedicated ADHD subjects (corrected p = 0.025), significantly decreased whole brain neural flexibility in unmedicated ADHD subjects as compared to the TDC group (corrected $p=0.012$), no significant difference in whole brain neural flexibility between medicated ADHD subjects and subjects in the TDC group (corrected $p=0.74$). Black asterisks indicate significant differences after FDR correction. Statistical significance levels: *p < 0.05.

Fig. 4. Successful prediction of ADHD status and severity using neural flexibility.

a The accuracy, sensitivity, specificity, and AUC when using all 264 ROIs, top 24 ROIs, and independent testing using the NYU dataset with the top 24 ROIs, respectively for the ADHD classification model, and b the spatial distribution of the most predictive 24 regions using ranked importance scores. c The R² scores when using all 264 ROIs, top 28 ROIs, and independent testing using the NYU dataset with the top 28 ROIs, respectively for the ADHD severity regression model, and d scatter plots comparing the representative neural flexibility-based severity score using tenfold cross validation and clinically obtained ADHD severity score using the top 28 regions for PKU dataset (left) and independent testing using the NYU dataset (right). e The spatial distribution of the most predictive 28 regions using ranked importance scores.