Title
Towards A Fully Spatially Coded Brain-Computer Interface: Simultaneous Decoding Of Visual Eccentricity And Direction
Abstract
By encoding visual targets with different locations relative to a stimulus, spatially coded brain-computer interface (BCI) has regained interest nowadays. Recent spatially coded BCI studies have demonstrated the feasibility of single-stimulus, multi-target BCIs, suggesting their potentials for simple and efficient applications. However, these studies have only decoded the visual direction information from the neural responses. To fully utilize the visual spatial information, it is necessary to include the visual eccentricity information as well. In the present study, the decodability of visual eccentricity information for BCI application was investigated for the first time. Sixteen targets were encoded simultaneously with eight directions and two eccentricities relative to a visual motion stimulus. Distinct neural spatial patterns and response strengths of motion-onset visual evoked potentials were elicited in the 16 attention conditions. The offline analysis reached an average classification accuracy of 63.1 +/- 11.5%, and the best-performing participant achieved an accuracy of 81.9%, well above the chance level (i.e., 6.25%) for 16-target classification. The results suggested the feasibility of simultaneous decoding of visual eccentricity and direction information towards a fully spatially coded BCI.
Year
DOI
Venue
2019
10.1109/EMBC.2019.8856586
2019 41ST ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC)
Field
DocType
Volume
Spatial analysis,Computer vision,Eccentricity (behavior),Computer science,Brain–computer interface,Offline analysis,Artificial intelligence,Visual motion,Decoding methods,Stimulus (physiology),Encoding (memory)
Conference
2019
ISSN
Citations 
PageRank 
1557-170X
0
0.34
References 
Authors
0
5
Name
Order
Citations
PageRank
Jingjing Chen131.76
Bo Hong225333.74
Yijun Wang330846.68
Xiaorong Gao459881.99
Dan Zhang55410.62