Exploring the interface of neuroscience and language acquisition : a systematic review of transferability in neurofeedback training

Electroencephalogram (EEG)-neurofeedback training (NFT) is currently receiving an increasing amount of attention following a series of empirical evidence supporting the correlation between EEG-NFT and cognitive enhancement. The present study has continued that focus by exploring the transferability of EEG-NFT effects in the domain of language acquisition, in particular, the working memory construct associated with second language acquisition (WM-SLA).

While positive effects of EEG-NFT have been found on cognitive performance in healthy subjects (Gruzelier, 2014b; Yeh et al., 2020; Da Silva & De Souza, 2021; Viviani & Vallesi, 2021), little research has been conducted investigating the correlation between EEG-NFT and WM-SLA. The current study aimed to examine the empirical evidence for the validation of EEG-NFT effects on WM-SLA in healthy subjects. It also aimed to compare the EEG-NFT parameters to determine an optimal paradigm for the efficacy of training and future studies. Measurement of both cognitive performance and neuroelectric changes were examined for further research.

To discuss the protocols and outcomes, a systematic review was carried out by synthesizing the empirical results of EEG-NFT studies on cognitive performance in healthy subjects. A total of 17 reports, published between 2007 and 2021, were selected based on the method of PICOS, out of 572 records available on the databases of PubMed, PsycoINFO, ERIC, and Scopus, of which 16 studies were included in the meta-analysis in order to describe between-subject effect size. Additionally, both behavioral and neuroelectric measurements were examined for better design of the neurofeedback trials.

Findings from the current review revealed that EEG-NFT had a significant effect on WM-SLA. Further moderator analysis showed more significant performance gains in the training frequency of alpha, the fronto-parietal area, visual feedback, the longer training dose. No significant gains were found in the use of strategy instruction. Regarding the measurements of behavioural and neuroelectric outcomes, the degree of difficulty in WM tests required better design, and the phasic EEG was found to be a better predictor than tonic EEG for cognitive changes.

This study explored the interface of neuroscience and language acquisition by proposing a refined EEG-NFT paradigm of EEG-NFT to optimize cognitive performance in the process of second language acquisition. As a nascent domain, more studies are required to further validate the effect of EEG-NFT on WM-SLA.