Functional EEG connectivity is a neuromarker for adult attention deficit hyperactivity disorder symptoms

• Published in: Clinical neurophysiology 2020-01, Vol.131 (1), p.330-342
• Main Authors: Kiiski, Hanni, Rueda-Delgado, Laura M., Bennett, Marc, Knight, Rachel, Rai, Laura, Roddy, Darren, Grogan, Katie, Bramham, Jessica, Kelly, Clare, Whelan, Robert
• Format: Article
• Language: English
• Subjects: Adult; Alpha Rhythm - physiology; Attention Deficit Disorder with Hyperactivity - diagnosis; Attention Deficit Disorder with Hyperactivity - physiopathology; Attention deficit-hyperactivity disorder; Beta Rhythm - physiology; Case-Control Studies; Connectome; Delta Rhythm - physiology; Dimensional psychiatry; Electroencephalography; Electroencephalography - methods; Female; Functional connectivity; Gamma Rhythm - physiology; Humans; Linear Models; Machine Learning; Male; Parents; Perceptual Disorders - physiopathology; Psychomotor Agitation - physiopathology; Siblings; Symptom Assessment; Theta Rhythm - physiology
• ISSN: 1388-2457; 1872-8952
• DOI: 10.1016/j.clinph.2019.08.010

•EEG functional connectivity predicted attention deficit hyperactivity disorder (ADHD) symptoms in adults.•EEG functional connectivity may be a neuromarker for adult ADHD.•EEG connectivity did not reliably distinguish between ADHD, 1st degree relative and control groups. Altered brain functional connectivity has been shown in youth with attention-deficit/hyperactivity disorder (ADHD). However, relatively little is known about functional connectivity in adult ADHD, and how it is linked with the heritability of ADHD. We measured eyes-open and eyes-closed resting electroencephalography (EEG) from 38 adults with ADHD, 45 1st degree relatives of people with ADHD and 51 healthy controls. Functional connectivity among all scalp channels was calculated using a weighted phase lag index for delta, theta, alpha, beta and gamma frequency bands. A machine learning analysis using penalized linear regression was used to identify if connectivity features (10,080 connectivity pairs) could predict ADHD symptoms. Furthermore, we examined if EEG connectivity could accurately classify participants into ADHD, 1st degree relatives and/or control groups. Hyperactive symptoms were best predicted by eyes-open EEG connectivity in delta, beta and gamma bands. Inattentive symptoms were predicted by eyes-open EEG connectivity in delta, alpha and gamma bands, and eyes-closed EEG connectivity in delta and gamma bands. EEG connectivity features did not reliably classify participants into groups. EEG connectivity may represent a neuromarker for ADHD symptoms. EEG connectivity may help elucidate the neural basis of adult ADHD symptoms.