Fig. 1: Experimental design, analysis pipeline, and hypotheses.
a CSST paradigm involved Certain Go, Uncertain Go, and Stop trials. b Dual control model decomposed reactive control and proactive control processes in the CSST. c We hypothesized that children with ADHD will show higher temporal variability and lower spatial stability of trial-evoked brain responses than TD children. d The Gaussian model was applied to assess the temporal variability of trial-evoked neural responses. e Representational similarity analysis (RSA) was used to examine the spatial stability of trial-evoked activation patterns. f ICP index was computed to measure the extent to which proactive control was implemented in Certain and Uncertain Go trials, which modulates trial-wise behavioral fluctuation. g Inter-subject pattern similarity analysis was used to investigate the group-specificity of neural coding during proactive and reactive control. h Core regions of interests (ROIs) in the salience (red color) and frontoparietal (blue color) network, defined from an independent meta-analysis study30. CSST stop-signal task, SSD stop-signal delay, TD typically developing, ADHD attention-deficit/hyperactivity disorder, SSRT stop-signal reaction time, ICP inhibitory control pattern, SuccStop Successful Stop trials, rAI right anterior insula, preSMA presupplementary motor cortex, rdlPFC right dorsal lateral prefrontal cortex, rPPC right posterior parietal cortex, RT reaction time, RSA representational similarity analysis.
