Investigation of Visual Stimulus With Various Colors and the Layout for the Oddball Paradigm in Evoked Related Potential-Based Brain-Computer Interface.

Miaoji Guo,Xingyu Wang,Andrzej Cichockia,Jing Jin,Yong Jiao

doi:10.3389/fncom.2019.00024

Abstract

Objective: Stimulus visual patterns, such as size, content, color, luminosity, and interval, play key roles for brain–computer interface (BCI) performance. However, the three primary colors to be intercompared as a single variable or factor on the same platform are poorly studied. In this work, we configured the visual stimulus patterns with red, green, and blue operating on a newly designed layout of the flash pattern of BCI to study the waveforms and performance of the evoked related potential (ERP).Approach: Twelve subjects participated in our experiment, and each subject was required to finish three different color sub-experiments. Four blocks of the interface were presented along the edge of the screen, and the other four were assembled in the center, aiming to investigate the problem of adjacency distraction. Repeated-measures ANOVA and Bonferroni correction were applied for statistical analysis.Main results: The averaged online accuracy was 98.44% for the red paradigm, higher than 92.71% for the green paradigm, and 93.23% for the blue paradigm. Furthermore, significant differences in online accuracy (p < 0.05) and information transfer rate (p < 0.05) were found between the red and green paradigms.Significance: The red stimulus paradigm yielded the best performance. The proposed design of ERP-based BCI was practical and effective for many potential applications.

Highlights

Brain–computer interface (BCI) enables patients suffering from movement disorders to communicate with others or interact with the outside world through electroencephalogram (EEG), magnetoencephalography (MEG), functional magnetic resonance imaging, and more (Vidal, 1973, 1977; Wolpaw et al, 2000, 2002)
We focused on Evoked related potential (ERP)-based BCI, which is one of the most promising approaches
A commonly used component in ERP is the visual evoked potential (VEP) P300 or P3, which is generally elicited by the oddball paradigm

Summary

Introduction

Brain–computer interface (BCI) enables patients suffering from movement disorders to communicate with others or interact with the outside world through electroencephalogram (EEG), magnetoencephalography (MEG), functional magnetic resonance imaging (fMRI), and more (Vidal, 1973, 1977; Wolpaw et al, 2000, 2002). A commonly used component in ERP is the visual evoked potential (VEP) P300 or P3, which is generally elicited by the oddball paradigm. Paralleling with RCP, the single-character paradigm (SCP), in which each character is individually highlighted, fully capable of avoiding adjacency distraction, has been extensively studied (Fazel-Rezai et al, 2012; Jin et al, 2015). To compare these two mainstream paradigms (RCP and SCP) fairly, Guger et al invited 100 healthy subjects to perform a spelling task, and the result showed that 72.8% (N = 81) of the subjects spelled RCP with 100% accuracy and 55.3% (N = 38) of the subjects did the same in SCP. RCP and SCP are both promising methods to establish a practical BCI system

Methods

Results

Discussion

Conclusion