Alpha Band Resting-State EEG Connectivity Is Associated With Non-verbal Intelligence.

Ilya Zakharov,Anna Tabueva,Yulia Kovas,Sergey Malykh,Timofey Adamovich

doi:10.3389/fnhum.2020.00010

Abstract

The aim of the present study was to investigate whether EEG resting state connectivity correlates with intelligence. One-hundred and sixty five participants took part in the study. Six minutes of eyes closed EEG resting state was recorded for each participant. Graph theoretical connectivity metrics were calculated separately for two well-established synchronization measures [weighted Phase Lag Index (wPLI) and Imaginary Coherence (iMCOH)] and for sensor- and source EEG space. Non-verbal intelligence was measured with Raven’s Progressive Matrices. In line with the Neural Efficiency Hypothesis, path lengths characteristics of the brain networks (Average and Characteristic Path lengths, Diameter and Closeness Centrality) within alpha band range were significantly correlated with non-verbal intelligence for sensor space but no for source space. According to our results, variance in non-verbal intelligence measure can be mainly explained by the graph metrics built from the networks that include both weak and strong connections between the nodes.

Highlights

Information processing in the brain is reflected in brain oscillations (Ward, 2003; Buzsáki and Draguhn, 2004; Clayton et al, 2015; Sadaghiani and Kleinschmidt, 2016)
The scatterplots for the relationship between wPLIbased metrics and intelligence are presented in Figure 2 (The descriptive statistics and correlations for other frequency bands, as well as scatterplots for Imaginary Coherence (iMCOH)-based metrics can be seen in Supplementary Tables S2–S5 and Supplementary Figures S1, S2)
There were a number of significant correlations between non-verbal intelligence and connectivity metrics for other frequency bands and EEG source space, none of them were consistent for the different samples and synchronization measures

Summary

Introduction

Information processing in the brain is reflected in brain oscillations (Ward, 2003; Buzsáki and Draguhn, 2004; Clayton et al, 2015; Sadaghiani and Kleinschmidt, 2016) It is still not clear how neurobiological factors contribute to more effective cognitive performance. In a seminal studies (Haier et al, 1988, 1992), using the Positron Emission Tomography (PET) method, participants with higher scores on Raven’s progressive matrices were found to consume less glucose comparing to participants with lower scores. Later these results were extended to more types of brain activity measures (EEG, fMRI and so on) and different types of tasks (see Neubauer and Fink, 2009 for review)

Objectives

Methods

Results

Conclusion