Does bimodal stimulus presentation increase ERP components usable in BCIs?
Does bimodal stimulus presentation increase ERP components usable in BCIs?
复制标题
DOI：
10.1088/1741-2560/9/4/045005
复制
发表时间：
2012-08
影响因子：
4
通讯作者：
M. Thurlings;A. Brouwer;Jan B. F. van Erp;B. Blankertz;P. Werkhoven
中科院分区：
文献类型：
--
作者：
M. Thurlings;A. Brouwer;Jan B. F. van Erp;B. Blankertz;P. Werkhoven
基于事件相关电位（ERP）的脑机接口（bci）利用大脑对参与和忽略刺激的反应差异。通常使用视觉刺激。最近，触觉刺激被认为是一种与视线无关的替代方法。由于多感觉整合，双峰刺激可引起额外的大脑活动，这可能用于脑机接口。我们研究了视觉-触觉刺激呈现对ERP组件链、脑机接口性能（分类准确率和比特率）和被试任务性能（目标计数）的影响。10名参与者被要求通过在视觉、触觉或视觉-触觉刺激序列中（空间上）关注目标来导航视觉显示。我们观察到，参与视觉触觉刺激（与视觉或触觉刺激相比）会导致早期ERP成分的增强（N1）。这种双峰N1可能会提高BCI性能，正如离线分类准确性的不显著正趋势所表明的那样。与视觉刺激相比，视觉-触觉刺激降低了后期ERP分量（P300），这与被试任务表现的不显著负向趋势是一致的。我们讨论这些发现的影响空间注意在高水平与低水平的刺激加工比较。此外，我们从实际的角度和未来的应用来评估双峰脑机接口。
Event-related potential (ERP)-based brain–computer interfaces (BCIs) employ differences in brain responses to attended and ignored stimuli. Typically, visual stimuli are used. Tactile stimuli have recently been suggested as a gaze-independent alternative. Bimodal stimuli could evoke additional brain activity due to multisensory integration which may be of use in BCIs. We investigated the effect of visual–tactile stimulus presentation on the chain of ERP components, BCI performance (classification accuracies and bitrates) and participants’ task performance (counting of targets). Ten participants were instructed to navigate a visual display by attending (spatially) to targets in sequences of either visual, tactile or visual–tactile stimuli. We observe that attending to visual–tactile (compared to either visual or tactile) stimuli results in an enhanced early ERP component (N1). This bimodal N1 may enhance BCI performance, as suggested by a nonsignificant positive trend in offline classification accuracies. A late ERP component (P300) is reduced when attending to visual–tactile compared to visual stimuli, which is consistent with the nonsignificant negative trend of participants’ task performance. We discuss these findings in the light of affected spatial attention at high-level compared to low-level stimulus processing. Furthermore, we evaluate bimodal BCIs from a practical perspective and for future applications.