Conditional generative adversarial networks applied to EEG data can inform about the inter-relation of antagonistic behaviors on a neural level

Amirali Vahid,Christian Beste,Ann-Kathrin Stock,Sebastian Stober,Moritz Mückschel

doi:10.1038/s42003-022-03091-8

Amirali Vahid, Christian Beste + Show 3 more

Open Access

https://doi.org/10.1038/s42003-022-03091-8

Copy DOI

Abstract

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.

Highlights

Goal-directed actions frequently require a balance between antagonistic processes, often showing an interdependency concerning what constitutes goal-directed behavior
Whenever Go trials occur with high frequency, participants tend to establish automated response tendencies that are difficult to inhibit[8,9,10,11]. When assuming that this interdependency is reflected on the neural level, it should be possible to predict the neural processes underlying inhibitory control based on the neural processes underlying speeded automatic responses
The results can be found in the supplemental material (Supplemental Fig. 1). These results show that the same conditional generative adversarial networks (cGANs) architecture as applied for the Go/Nogo task was less well able to generate a signal in the Simon Task

Summary

Introduction

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. The generated neurophysiological pattern should show high similarities with recorded data If this was the case, applying artificial intelligence methods to EEG data could inform cognitive science and provide information about the principles underlying antagonistic classes of goal-directed behavior on a neurophysiological level. This would further our understanding of interdependencies between distinct cognitive processes on a neural level, but it may in the long-range provide an opportunity to test the predictions of computational modeling, inform future theories, and—potentially—allow for the prediction of behavioral performance in various situations

Objectives

Methods

Conclusion