| Autor: |
Bayliss, Jessica D., Inverso, Samuel A., Tentler, Aleksey |
| Předmět: |
|
| Zdroj: |
CyberPsychology & Behavior; Dec2004, Vol. 7 Issue 6, p694-704, 11p, 1 Black and White Photograph, 1 Illustration, 3 Charts, 2 Graphs |
| Abstrakt: |
Brain–computer interfaces (BCIs) are now feasible for use as an alternative control option forthose with severe motor impairments. The P300 component of the evoked potential has provenuseful as a control signal. Individuals do not need to be trained to produce the signal, and itis fairly stable and has a large evoked potential. Even with recent signal classification advances,on-line experiments with P300-based BCIs remain far from perfect. We present twopotential methods for improving control accuracy. Experimental results in an evoked potentialBCI, used to control items in a virtual apartment, show a reduced response exists whenitems are accidentally controlled. The presence of a P300-like signal in response to goal itemsmeans that it can be used for automatic error correction. Preliminary results from an interfaceexperiment using three different button configurations for a yes/no BCI task show that theconfiguration of buttons may affect on-line signal classification. These results will be discussedin light of the special considerations needed when working with an amyotrophic lateralsclerosis (ALS) patient. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
