Visual Attention’s Past Shapes The Future of BCI
A Brain-Computer Interface (BCI) offers a communication option to those who suffer from amyotrophic lateral sclerosis (ALS) and other motor disabilities. The majority of BCIs use a visual interface. Visual paradigms offer the highest communication rates when compared to other modalities such as auditory or tactile. One of the greatest challenges in BCI is signal-to-noise ratio (SNR). There is substantial research in signal processing to improve classification techniques and SNR. However, past research in visual attention has been largely overlooked and could improve SNR and usability. This work will review the present BCI paradigms and examine them under visual attention principles. The original visual paradigm, Row/Column (RC), was developed as a fast and simple method to present all of the items of a matrix rapidly. The RC paradigm design did not consider visual attention research topics such as; Gestalt grouping, attentional blink, flanker effects, and space based attention. Attentional principles can explain errors in BCI paradigms (adjacency error, attentional blink, non-target distraction) and how other paradigms have improved performance over RC. In addition, researchers use event-related potential components associated with different types of perception such as motion onset, covert attention, and color processing in an attempt to improve classification. By combining visual attention and signal processing research, a more effective BCI that is easier to use can be brought to those to need it most.
Ryan, David B. and Sellers, Eric W., "Visual Attention’s Past Shapes The Future of BCI" (2012). ETSU Faculty Works. 914.