Attention Deficit Hyperactivity Disorder (ADHD) is one of the most prevalent neuropsychiatric disorders in childhood and adolescence and its diagnosis is based on clinical interviews, symptom questionnaires, and neuropsychological testing. Much research effort has been undertaken to evaluate the usefulness of neurophysiological (EEG) data to aid this diagnostic process. In the current study, we applied deep learning methods on event-related EEG data to examine whether it is possible to distinguish ADHD patients from healthy controls using purely neurophysiological measures. The same was done to distinguish between ADHD subtypes. The results show that the applied deep learning model (“EEGNet”) was able to distinguish between both ADHD subtypes and healthy controls with an accuracy of up to 83%. However, a significant fraction of individuals could not be classified correctly. It is shown that neurophysiological processes indicating attentional selection associated with superior parietal cortical areas were the most important for that. Using the applied deep learning method, it was not possible to distinguish ADHD subtypes from each other. This is the first study showing that deep learning methods applied to EEG data are able to dissociate between ADHD patients and healthy controls. The results show that the applied method reflects a promising means to support clinical diagnosis in ADHD. However, more work needs to be done to increase the reliability of the taken approach.

Efficient action control is indispensable for goal-directed behaviour. Different theories have stressed the importance of either attention or response selection sub-processes for action control. Yet, it is unclear to what extent these processes can be identified in the dynamics of neurophysiological (EEG) processes at the single-trial level and be used to predict the presence of conflicts in a given moment. Applying deep learning, which was blind to cognitive theory, on single-trial EEG data allowed to predict the presence of conflict in~95% of subjects 33% above chance level. Neurophysiological features related to attentional and motor response selection processes in the occipital cortex and the superior frontal gyrus contributed most to prediction accuracy. Importantly, deep learning was able to identify predictive neurophysiological processes in single-trial neural dynamics. Hence, mathematical (artificial intelligence) approaches may be used to foster the validation and development of links between cognitive theory and neurophysiology of human behavior.

Neurophysiological features like event-related potentials (ERPs) have long been used to identify different cognitive sub-processes that may contribute to task performance. It has however remained unclear whether “classical” ERPs are truly the best reflection or even causal to observable variations in behavior. Here, we used a data-driven strategy to extract features from neurophysiological data of n = 240 healthy young individuals who performed a Go/Nogo task and used machine learning methods in combination with source localization to identify the best predictors of inter-individual performance variations. Both Nogo-N2 and Nogo-P3 yielded predictions close to chance level, but a feature in between those two processes, associated with motor cortex activity (BA4), predicted group membership with up to ~68%. We further found two Nogo-associated features in the theta and alpha bands, that predicted behavioral performance with up to ~78%. Notably, the theta band feature contributed most to the prediction and occurred at the same time as the predictive ERP feature. Our approach provides a rigorous test for established neurophysiological correlates of response inhibition and suggests that other processes, which occur in between the Nogo-N2 and P3, might be of equal, if not even greater, importance.

Cognitive control processes are advantageous when routines would not lead to the desired outcome, but this can be ill-advised when automated behavior is advantageous. The aim of this study was to identify neural dynamics related to the ability to adapt to different cognitive control demands – a process that has been referred to as ‘metacontrol.’ A sample of N = 227 healthy subjects that was split in a ‘high’ and ‘low adaptability’ group based on the behavioral performance in a task with varying control demands. To examine the neurophysiological mechanisms, we combined event-related potential (ERP) recordings with source localization and machine learning approaches. The results show that individuals who are better at strategically adapting to different cognitive control demands benefit from automatizing their response processes in situations where little cognitive control is needed. On a neurophysiological level, neither perceptual/attentional selection processes nor conflict monitoring processes paralleled the behavioral data, although the latter showed a descriptive trend. Behavioral differences in metacontrol abilities were only significantly mirrored by the modulation of response-locked P3 amplitudes, which were accompanied by activation differences in insula (BA13) and middle frontal gyrus (BA9). The machine learning result corroborated this by identifying a predictive/classification feature near the peak of the response-locked P3, which arose from the anterior cingulate cortex (BA24; BA33). In short, we found that metacontrol is associated to the ability to manage response selection processes, especially the ability to effectively downregulate cognitive control under low cognitive control requirements, rather than the ability to upregulate cognitive control.

Goal-directed actions frequently require a balance between antagonistic processes (e.g., executing and inhibiting a response), often showing an interdependency concerning what constitutes goal-directed behavior. While an inter-dependency of antagonistic actions is well described at a behavioral level, a possible inter-dependency of underlying processes at a neuronal level is still enigmatic. However, if there is an interdependency, it should be possible to predict the neurophysiological processes underlying inhibitory control based on the neural processes underlying speeded automatic responses. Based on that rationale, we applied artificial intelligence and source localization methods to human EEG recordings from N = 255 participants undergoing a response inhibition experiment (Go/Nogo task). We show that the amplitude and timing of scalp potentials and their functional neuroanatomical sources during inhibitory control can be inferred by conditional generative adversarial networks (cGANs) using neurophysiological data recorded during response execution. We provide insights into possible limitations in the use of cGANs to delineate the interdependency of antagonistic actions on a neurophysiological level. Nevertheless, artificial intelligence methods can provide information about interdependencies between opposing cognitive processes on a neurophysiological level with relevance for cognitive theory.