Two-Class c-VEP EEG Dataset for Covert Attention-Based, Gaze-Independent Brain-Computer Interfacing

Recent advances in brain-computer interfacing (BCI) have led to the development of a sophisticated stimulus protocol known as noise-tagging, in which stimuli are encoded using pseudo-random noise sequences. These sequences are ideal for BCI applications because they exhibit near-zero correlation with time-shifted versions of themselves and with other stimuli, minimizing interference and enabling robust signal detection. This type of stimulation elicits a neural response known as the code-modulated visual evoked potential (c-VEP).

Traditionally, c-VEP BCI spellers require users to move their eyes, referred to as gaze-dependent BCIs, to fixate on the target letter and thereby on the corresponding stimulus sequence. This process, known as overt attention, becomes increasingly more difficult for individuals with motor impairments, such as those living with amyotrophic lateral sclerosis (ALS).

In this study, we introduce a two-class c-VEP BCI that eliminates the need for eye movements. Our gaze-independent approach uses a covert attention paradigm, in which users maintain fixation at the center of the screen while directing their attention to a stimulus presented in the visual periphery, either to the left or right of a central fixation cross.

近年来，脑机接口（brain-computer interfacing, BCI）领域的研究进展推动了一种名为噪声标记（noise-tagging）的复杂刺激范式的发展，该范式采用伪随机噪声序列对刺激进行编码。这类序列非常适用于脑机接口应用，因为它们与自身的时移版本及其他刺激之间几乎不存在相关性，能够最大限度降低干扰，实现可靠的信号检测。此类刺激可诱发出一种被称为码调制视觉诱发电位（code-modulated visual evoked potential, c-VEP）的神经响应。

传统的码调制视觉诱发电位脑机接口拼写器要求用户通过眼球运动——即依赖注视的脑机接口（gaze-dependent BCIs）——将视线固定在目标字母对应的刺激序列上。这一被称为外显注意（overt attention）的过程，对于肌萎缩侧索硬化症（amyotrophic lateral sclerosis, ALS）等运动障碍患者而言会愈发困难。

本研究提出了一种无需眼球运动的二类码调制视觉诱发电位脑机接口。我们的非注视依赖方案采用内隐注意范式（covert attention paradigm），即让用户始终将视线固定在屏幕中央，同时将注意力转向视觉外周区域（中央注视十字左侧或右侧）呈现的刺激。