Fig 1.
Eighteen participants volunteered and were screened, thirteen participants were included in the study and randomly assigned to a group.
Table 1.
Characteristics of study participants.
Fig 2.
After enrollment, participants attended six weekly sessions. A week prior to the training, baseline ADHD symptoms and cognitive functioning were assessed (pre-test). Then the four weekly training sessions commenced (60-min training time/session). A week after, the last training session, the behavioral post-assessment, was done (post-test). ADHD = Attention Deficit Hyperactivity Disorder, fMRI = functional magnetic resonance imaging, NP-test = Neuropsychological tests, MSIT = Multi Source Interference task, SA-DOTS = Sustained Attention DOTS task, SART = Sustained Attention to Response Task, 2-back WM = 2-back Working memory task, WAIS = Wechsler Adult Intelligence Scale, DS = Digit Span, LNS = Letter Number Sequencing, VC = Vocabulary, BD = Block Design, QCM = Questionnaire of Current Motivation, MCT = Mental Calculation Task.
Fig 3.
Localization of dACC target regions.
Activation during the localization tasks is depicted for the interference task (MSIT, panel A, peak activation x = 8, y = 10, z = 45, Talairach space) and the mental calculation task (MCT, panel B, peak activation x = -8, y = 10, z = 48, Talairach space). To verify if the targeted dACC region was localized/defined within the online procedure, a whole-brain random effects GLM analysis (p<0.05, corrected at cluster level) was performed for the contrast interference vs. no interference (MSIT, panel A), and the contrast mental calculation vs. mental singing (MCT, panel B). Panel C depicts the location of the online defined individual dACC target regions around their average location (x = 2, y = 15, z = 39, Talairach space, neurofeedback group = blue tints, control group = green tints). The slightly more anterior location of the dACC target regions respective to the peak of activation may have resulted from a preference for the most anterior clusters during the localization procedure, which have been shown to be hypo-activated in ADHD patients [26].
Fig 4.
Activation levels within dACC target regions during localization tasks.
Participants across groups showed significantly increased activation levels in the dACC target region during the interference condition of the MSIT (panel A, marked with an asterisk), and the mental calculation condition of the MCT (panel B). There were no changes in performance over time, confirming that both groups consistently performed the task as instructed.
Fig 5.
Activation levels within dACC target regions during MRI training and transfer runs.
The results demonstrate that all participants achieved the expected up-regulation of activation levels within the dACC target regions with training. Both groups showed a marked and significant increase in dACC activation level from the third session onwards, which was evident during training runs (panel A, marked with an asterisk), and marginally significant during transfer runs (panel B). Importantly, participants were able to sustain increased activation levels until the end of training. There were no significant group differences and neither group demonstrated ability to differently up-regulate the dACC activation levels during the medium and high difficulty condition.
Table 2.
Pre-post behavioral assessment (within group).