A P300-based brain-computer interface for people with amyotrophic lateral sclerosis

Abstract

Objective: The current study evaluates the efficacy of a P300-based brain-computer interface (BCI) communication device for individuals with advanced ALS.

Methods: Participants attended to one cell of a N x N matrix while the N rows and N columns flashed randomly. Each cell of the matrix contained one character. Every flash of an attended character served as a rare event in an oddball sequence and elicited a P300 response. Classification coefficients derived using a stepwise linear discriminant function were applied to the data after each set of flashes. The character receiving the highest discriminant score was presented as feedback.

Results: In Phase I, six participants used a 6 x 6 matrix on 12 separate days with a mean rate of 1.2 selections/min and mean online and offline accuracies of 62% and 82%, respectively. In Phase II, four participants used either a 6 x 6 or a 7 x 7 matrix to produce novel and spontaneous statements with a mean online rate of 2.1 selections/min and online accuracy of 79%. The amplitude and latency of the P300 remained stable over 40 weeks.

Conclusions: Participants could communicate with the P300-based BCI and performance was stable over many months.

Significance: BCIs could provide an alternative communication and control technology in the daily lives of people severely disabled by ALS.

Figures

Figure 1

A) The picture shows the P300 speller matrix displayed on a screen in a participant’s home. The participant is wearing a 16-channel EEG cap that is connected to an amplifier. His task was to attend to the letter shown in parentheses (top line letter n) and count how many times the letter n flashed in the matrix. The rows and columns were flashed in random order. The second line provided the participants with feedback of results. B) An example of the 7 × 7 matrix used for free spelling. Completed text is shown on the top line and the word currently being spelled is shown on the bottom line (in German).

Figure 2

Online (open bars) and offline (shaded bars) accuracy for each of the six participants averaged across all 10 copy-spelling feedback sessions. The offline analysis, which used optimized classification coefficients, provided higher accuracy.

Figure 3

Online (solid line) and offline (dotted line) accuracy for each of the ten feedback sessions averaged across all participants. Offline accuracy is higher for all sessions.

Figure 4

Average P300 amplitude (A) and latency (B) for all participants for each copy-spelling session. Amplitude values are shown as proportion of the average amplitude for the first session.