106 - The Prepotent Response EEG Task: Evaluating Inhibitory Control in Preschoolers with Elevated ADHD Symptoms

Publication Number: 1098.106

Matthew Zimon, Boston Children's Hospital, Brighton, MA, United States; Anne Arnett, Boston Children's Hospital, Boston, MA, United States

Background: The Go/No-Go (GNG) task is a laboratory measure of inhibitory control, in which consecutive Go trials build prepotency for motor response, while No-Go trials elicit suppression of the habituated prepotent response. Previous literature has linked prepotent response inhibition and concurrent neurophysiology to attention deficit hyperactivity disorder (ADHD) in school-aged children. While prepotency level moderates GNG performance and is sensitive to inhibitory control problems in school-aged children, it remains unknown whether the same associations are found in preschoolers.

Objective: This study investigated task performance and neurophysiological response during No-Go trials using a GNG task with concurrent electroencephalography (EEG) in school-aged and preschool-aged children. We expected reduced accuracy and attenuated late P3 amplitudes among children with ADHD, especially at higher prepotency levels.

Design/Methods: Participants were 83 school-aged children with ADHD (60%) or typically developing (TD) and 79 preschool-aged children with elevated (63%) or low ADHD symptoms. GNG was completed during EEG acquisition with a high-density cap. The GNG task had a 40% No-Go frequency with age-appropriate versions for each cohort. Prepotent response level was defined as the number of correct Go trials preceding a No-Go trial, categorized as low (0 Go trials), medium (1 Go trial), and high (2+ Go trials) levels. The late P3 event-related potential (ERP) component was extracted at the centro-parietal region during correct No-Go trials at 375-525ms and 425-600ms for school-aged and preschool-aged children respectively.

Results: Across age cohorts, accuracy was lower at high levels relative to low levels of prepotency (B = -21.4, SE = 1.9, p < .001; B =-16.3, SE = 2.2, p<.001). In the school-aged cohort only, there was a group x prepotency interaction, such that accuracy was lower in ADHD relative to TD children only at high levels of prepotency (Figure 1). Late P3 amplitude was greater for high compared to low levels of prepotency in both cohorts (B =0.64, SE = 0.27, p=.02; B =1.51, SE = 0.44, p<.001) (Figure 2). No significant differences in the late P3 component were found between ADHD and TD groups.

Conclusion(s): Prepotency level moderates sensitivity of the GNG task to ADHD at school age. Differences in both accuracy and late P3 amplitude were associated with prepotency level across age cohorts. The GNG task may need to be further adapted to find associations between laboratory and clinical measures of response inhibition in young children.

No-Go Accuracy and ADHD Dx Group by Prepotency Level, School-aged Cohort
Figure 1. No-Go Accuracy and ADHD Dx group at low, mid (medium), and high prepotency levels for the school-aged cohort. Correct No-Go trial rate was significantly lower in the ADHD group only at high levels of prepotency. Both groups exhibited reduced No-Go accuracy at high compared to low prepotency levels.

ERP Waveform by Prepotency Level and Group
Figure 2. Stimulus-locked waveform at correct No-Go trials for school-aged (Left) and preschool-aged (Right) groups. Dashed lines indicate the timeframe of the late P3 event-related potential (ERP) component at 375-525ms and 425-600ms for the school-aged and preschool-aged cohorts respectively. No-go trials with high levels of prepotency evoked a greater late P3 amplitude than low levels of prepotency. The ADHD group showed reduced late P3 amplitude on average compared to the TD group, but this difference was not significant.