Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Clinical Trial
. 2006 Feb;59(2):127-40.
doi: 10.1016/j.ijpsycho.2005.02.007. Epub 2005 Jul 27.

Effects of SOA and flash pattern manipulations on ERPs, performance, and preference: implications for a BCI system

Affiliations
Clinical Trial

Effects of SOA and flash pattern manipulations on ERPs, performance, and preference: implications for a BCI system

Brendan Z Allison et al. Int J Psychophysiol. 2006 Feb.

Abstract

P3 brain-computer interfaces (BCIs) are synchronous communication systems that allow users to communicate interest in a target event by choosing to attend to it while ignoring other events. In such a system, a cogneme refers to the user's response to: "/attend to the event/" or "/ignore the event/". The present study examined subjects' ability to generate more cognemes per minute (by varying stimulus onset asynchrony or SOA), or requiring fewer cognemes to convey a message (by varying the pattern of stimulus presentation). Both of these have implications for improved information throughput in a P3 BCI. SOAs of 125, 250, and 500 ms were used. Additionally, the conventional "single flash" approach was compared to a new "multiple flash" condition in which half of the stimuli in an 8 x 8 grid were flashed simultaneously. In both conditions, P3-like component amplitudes decreased with faster SOAs at low target probabilities, but the trend did not hold for higher probabilities. The multiple flash condition produced more robust ERPs at the faster speeds. The results also indicate that attend/ignore differences were more apparent following multiple flashes for low target probabilities, but less apparent for high target probabilities. Although information throughput alone does not support the superiority of one approach over the other, only six cognemes are needed in the multiple flash conditions to identify a character, compared to sixteen cognemes in the single flash condition. This suggests that the former approach could operate more rapidly. Thus, the present results suggest that the multiple flash approach may be a more efficient and faster basis for a P3 BCI system.

PubMed Disclaimer

MeSH terms

LinkOut - more resources