Mitigating the Impact of Psychophysical Effects During Adaptive Stimulus Selection in the P300 Speller Brain-Computer...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Stimulus-driven brain-computer interfaces (BCIs), like the P300 speller, utilize event-related potentials (ERPs) for device control.
• Psychophysical factors such as refractory effects and adjacency distractions can impair ERP elicitation and BCI performance.
• Conventional pseudo-random stimulus presentation may not optimize ERP elicitation.

Purpose of the Study:

• To enhance an adaptive BCI stimulus selection algorithm.
• To mitigate adjacency distractions and refractory effects in BCI paradigms.
• To improve spelling speeds and communication rates for BCI users.

Main Methods:

• Developed an enhanced adaptive algorithm for BCI stimulus selection.
• Modeled temporal dependencies of ERP elicitation within the objective function.
• Imposed spatial restrictions on the stimulus search space to reduce distractions.

Main Results:

• The enhanced adaptive stimulus selection algorithm demonstrated improved spelling speeds.
• Simulations using synthetic and human data confirmed the algorithm's effectiveness.
• The algorithm outperformed conventional BCI stimulus presentation paradigms.

Conclusions:

• The enhanced adaptive stimulus selection algorithm offers a significant improvement over conventional methods.
• This advancement can increase communication rates for individuals relying on BCIs.
• The findings support the translation of BCIs as viable communication tools for those with severe neuromuscular limitations.