Improving the performance of a gaze independent P300-BCI by using the expectancy wave

Abstract

Objective. A P300-BCI conveys a subject's intention through recognition of their ERPs. However, in the case of visual stimuli, its performance depends strongly on eye gaze. When eye movement is impaired, it becomes difficult to focus attention on a target stimulus, and the quality of the ERP declines greatly, thereby affecting recognition efficiency. Approach. In this paper, the expectancy wave (E-wave) is proposed to improve signal quality and thereby improve identification of visual targets under the covert attention. The stimuli of the P300-BCI described here are presented in a fixed sequence, so the subjects can predict the next target stimulus and establish a stable expectancy effect of the target stimulus through training. Features from the E-wave that occurred 0~300ms before a stimulus were added to the post-stimulus ERP components for intention recognition. Main Results. Comparisons of 10 healthy subjects before and after training demonstrated that the expectancy wave generated before target stimulus could be used with the P300 component to improve character recognition accuracy (CRA) from 85% to 92.4%. In addition, CRA using only the expectancy component can reach 68.2%, which is significantly greater than random probability (16.7%). The results of this study indicate that the expectancy wave can be used to improve recognition efficiency for a gaze-independent P300-BCI, and that training contributes to induction and recognition of the potential. Significance. This study proposes an effective approach to an efficient gaze-independent P300-BCI system.

Evaluation of Different Types of Stimuli in an Event-Related Potential-Based Brain–Computer Interface Speller under Rapid Serial Visual Presentation

“…However, other paradigms and modalities may also be potentially suitable for patients without oculomotor control. Some of these visual paradigms include those previously demonstrated by, for example, [ 36 ] or [ 37 ], which are based on the use of covert attention. While these systems perform adequately, they may face challenges with increased stimuli, given that they used only two and six selectable elements, respectively.…”

“…Second, it would be interesting to evaluate the specific factors that can affect user comfort (e.g., workload [ 54 , 55 ] or usability [ 56 ]). Third, it might be relevant to combine the findings of this work with the results from prior research (e.g., alternative gaze-independent paradigms [ 37 ]). In conclusion, incorporating the insights from this study with those of prior research is crucial for advancing our understanding and application of ERP-BCI systems, ultimately benefiting users, and addressing their unique needs.…”

Towards an Eye-Brain-Computer Interface: Combining Gaze with the Stimulus-Preceding Negativity for Target Selections in XR

“…Kato et al [49] also used the CNV to develop a switch for activating or deactivating a BCI system. In other works [123], the CNV was used in conjunction with a P300-based BCI to improve its accuracy from 85% to 92.4% in identifying visual targets. While the CNV appears promising as a passive mental 'click', its reliance on imagined motor activity can distract users from their primary task.…”

“…In contrast, passive BCIs have emerged as a compelling avenue, where users do not deliberately control or manipulate their thoughts nor does the stimuli need to be altered to evoke a specific brainwave response, facilitating smoother interactions and reducing user cognitive load [126]. Passive BCI paradigms tapping into slow negative waves, specifically the Stimulus-Preceding Negativity (SPN) and the Contingent Negative Variation (CNV), have gained attention only in recent literature [41, 94, 104, 123, 129], though they were proposed back in 1966 for ‘the direct cerebral control of machines’ [118].…”