Efficient methods of presenting visual stimuli for the P300-based cognitive BMI

• TAKAI Hideaki

  Advanced Industrial Science and Technology

• MINAMI Tetsuto

  Toyohashi University of Technology

• HASEGAWA Ryohei P.

  Advanced Industrial Science and Technology

Bibliographic Information

Other Title

• Ｐ３００に基づく認知型ＢＭＩにおける効率の良い刺激提示方法の検討

Abstract

In Japan, where population aging and the trend toward nuclear families are rapidly advancing, various problems attributed to changes in disease structures and life styles have become serious. Especially when someone develops difficulty in speaking and writing, there is a tendency for the person to become socially isolated because he/she does not have the ability to convey his/her intention and in some cases this might become a matter of life and death. The brain-machine/computer interface (BMI/BCI) technology is expected to allow treatment of patients with disabilities in brain or physical functions and improve the quality of life for handicapped people. One of the good examples of BMIs is the P300 speller (Farwell and Donchin, 1988). The previous P300 speller systems give the timing to the subjects with the change of brightness or colors. In this study, we suggested a new method of short-time presentation of face or word stimuli. We used those stimuli to decrease the prediction errors between adjacent choices based upon the knowledge from previous studies showing that the face and word elicited N170 component mainly in the right and left hemispheres respectively. We assumed that the difference would contribute to the decrease of adjacent errors. We compared the success rates of predictions among four conditions of visual stimuli (“brightness”, “face only”, “word only” and “face/word combination”). The better performance was obtained by “word only” and “face/word combination” than by “brightness”. These results suggest that P300-based communication will be more efficient by our methods.

Journal

• Transactions of Japan Society of Kansei Engineering

  Transactions of Japan Society of Kansei Engineering 10 (2), 89-94, 2011

  Japan Society of Kansei Engineering

Keywords

• Brain Machine Interface (BMI)
• EEG
• P300

Details

• CRID: 1390282680302060416
• NII Article ID: 130001361990
• DOI: 10.5057/jjske.10.89
• ISSN: 18845258
• ISSN: 18840833
• Text Lang: ja
• Data Source:
  • JaLC
  • Crossref
  • CiNii Articles
  • KAKEN
• Abstract License Flag: Disallowed