Evaluation Of P300 Speller Performance Using Large Language Models Along With Cross-Subject Training

TL;DR

This paper enhances P300 speller BCI performance for ALS patients by integrating large language models and cross-subject training, achieving significant speed improvements in communication, especially with GPT2 for multi-word prediction.

Contribution

It introduces the use of advanced language models and multi-subject training techniques to optimize stimuli presentation and word prediction in P300 speller BCIs, significantly improving communication speed.

Findings

Approximately 10% speed improvement from character-level optimizations.

Around 40% speed gain using GPT2 for multi-word prediction.

Performance levels within 10% of theoretical limits achieved.

Abstract

Amyotrophic lateral sclerosis (ALS), a progressive neuromuscular degenerative disease, severely restricts patient communication capacity within a few years of onset, resulting in a significant deterioration of quality of life. The P300 speller brain computer interface (BCI) offers an alternative communication medium by leveraging a subject's EEG response to characters traditionally highlighted on a character grid on a graphical user interface (GUI). A recurring theme in P300-based research is enhancing performance to enable faster subject interaction. This study builds on that theme by addressing key limitations, particularly in the training of multi-subject classifiers, and by integrating advanced language models to optimize stimuli presentation and word prediction, thereby improving communication efficiency. Furthermore, various advanced large language models such as Generative Pre-Trained Transformer (GPT2), BERT, and BART, alongside Dijkstra's algorithm, are utilized to optimize stimuli and provide word completion choices based on the spelling history. In addition, a multi-layered smoothing approach is applied to allow for out-of-vocabulary (OOV) words. By conducting extensive simulations based on randomly sampled EEG data from subjects, we show substantial speed improvements in typing passages that include rare and out-of-vocabulary (OOV) words, with the extent of improvement varying depending on the language model utilized. The gains through such character-level interface optimizations are approximately 10%, and GPT2 for multi-word prediction provides gains of around 40%. In particular, some large language models achieve performance levels within 10% of the theoretical performance limits established in this study. In addition, both within and across subjects, training techniques are explored, and speed improvements are shown to hold in both cases.