EEG cross-frequency coupling associated with attentional performance: An RDoC approach to attention

摘要

19th biennial IPEG Meeting: Nijmegen, The Netherlands. 26-30 October 2016. The quality of attentional performance plays a crucial role in goaldirected behavior in daily life activities, cognitive task performance, and in multiple psychiatric illnesses. The Research Domain Criteria (RDoC) approach put forward by the National Institute of Mental Health aims to investigate cognitive constructs while abandoning the conventional diagnostic system of psychiatric illnesses. The current study used an RDoC approach to investigate functions underlying attentional performance. One of the previously postulated physiologic mechanisms that could explain variance in attentional performance is the quality of interplay between neuronal networks. Various attempts to visualize this interplay have been made using different approaches. In our current study, we aimed to validate the approach of functional Independent Component Analysis (fICA) based on electroencephalograms (EEG’s) for this purpose. This method yields components that reflect EEG cross-frequency coupling patterns between networks (details about the method can be found elsewhere). We first applied fICA to combined Eyes Open resting state EEG and EEG during an n-back task data in a large sample of healthy adults (n = 1397), yielding 32 components. Secondly, we obtained individual component loadings for every subject for the two conditions as well as a difference loading score (Loadingtask-LoadingEO) per network. Thirdly, we operationalized attentional performance by differentiating between attenders (n = 704) versus non-attenders, (n = 320) on the n-back task and found a significant difference between groups for the difference loading score for component 10. We proposed that component 10 reflects the anticorrelated interaction of an attention network and a resting state network. This finding was cross-validated in an adolescent Attention-Deficit/Hyperactivity Disorder (ADHD) population (n = 80), clinically suffering from attentional problems. As expected, the difference loading scores in this group was similar to the pattern observed in non-attenders. Furthermore, it was accompanied by a lower overall loading on component 10 in both conditions. The current findings seem to validate fICA as a method to visualize neuronal networks and their interactions. Combining this method with objective behavioral measures may contribute to the understanding of brain mechanisms involved in attention and attentional problems such as observed in multiple psychiatric illnesses.