Mathew Salvaris

Visual modifications on the P300 speller BCI paradigm

The best known P300 speller brain-computer interface (BCI) paradigm is the Farwell and Donchin paradigm. In this paper, various changes to the visual aspects of this protocol are explored as well as their effects on classification. Changes to the dimensions of the symbols, the distance between the symbols and the colours used were tested. The purpose of the present work was not to achieve the highest possible accuracy results, but to ascertain whether these simple modifications to the visual protocol will provide classification differences between them and what these differences will be. Eight subjects were used, with each subject carrying out a total of six different experiments. In each experiment, the user spelt a total of 39 characters. Two types of classifiers were trained and tested to determine whether the results were classifier dependant. These were a support vector machine (SVM) with a radial basis function (RBF) kernel and Fishers linear discriminant (FLD). The single-trial classification results and multiple-trial classification results were recorded and compared. Although no visual protocol was the best for all subjects, the best performances, across both classifiers, were obtained with the white background (WB) visual protocol. The worst performance was obtained with the small symbol size (SSS) visual protocol.

Classification effects of real and imaginary movement selective attention tasks on a P300-based brain–computer interface

Brain-computer interfaces (BCIs) rely on various electroencephalography methodologies that allow the user to convey their desired control to the machine. Common approaches include the use of event-related potentials (ERPs) such as the P300 and modulation of the beta and mu rhythms. All of these methods have their benefits and drawbacks. In this paper, three different selective attention tasks were tested in conjunction with a P300-based protocol (i.e. the standard counting of target stimuli as well as the conduction of real and imaginary movements in sync with the target stimuli). The three tasks were performed by a total of 10 participants, with the majority (7 out of 10) of the participants having never before participated in imaginary movement BCI experiments. Channels and methods used were optimized for the P300 ERP and no sensory-motor rhythms were explicitly used. The classifier used was a simple Fishers linear discriminant. Results were encouraging, showing that on average the imaginary movement achieved a P300 versus No-P300 classification accuracy of 84.53%. In comparison, mental counting, the standard selective attention task used in previous studies, achieved 78.9% and real movement 90.3%. Furthermore, multiple trial classification results were recorded and compared, with real movement reaching 99.5% accuracy after four trials (12.8 s), imaginary movement reaching 99.5% accuracy after five trials (16 s) and counting reaching 98.2% accuracy after ten trials (32 s).

Wavelets and ensemble of FLDs for P300 classification

Over the last few years various P300 classification algorithms have been assessed using the P300 data provided by the Wadsworth center for brain-computer interface (BCI) competitions II and III. In this paper a novel method of P300 classification is presented and compared to the state of the art results obtained for BCI competition II data set IIb and BCI competition III data set II. The novel classification method includes discrete-wavelet transform (DWT) preprocessing and an ensemble of Fishers Linear Discriminants for classification. The performance of the proposed method is as good as the state of the art method for the BCI competition II data set and only slightly worse than the state of the art method for BCI competition III data sets. Furthermore the proposed method is far less computationally expensive than the current state of the art method and could be modified for adaptive behavior in an online system.

Novel Protocols for P300-Based Brain–Computer Interfaces

The oddball protocol is often used in brain-computer interfaces (BCIs) to induce P300 ERPs, although, recently, some issues have been shown to detrimentally effect its performance. In this paper, we study a new periodic protocol and explore whether it can compete with the standard oddball protocol within the context of a BCI mouse. We found that the new protocol consistently and significantly outperforms the standard oddball protocol in relation to information transfer rates (33 bits/min for the former and 22 bits/min for the latter, measured at 90% accuracy) as well as P300 amplitudes. Furthermore, we performed a comparison of two periodic protocols with two less conventional oddball-like protocols that reveals the importance of the interactions between task and sequence in determining the success of a protocol.

Comment on 'Fast attainment of computer cursor control with noninvasively acquired brain signals'

In a recent paper by Bradberry, Gentili and Contreras-Vidal published in Journal of Neural Engineering (2011, 8 036010), an interesting method for the control of a two-dimensional mouse cursor was proposed, which apparently attained excellent control and good speed with relatively simple techniques. We believe some of the results in the paper have been misinterpreted due to a failure in appreciating the self-fulfilling nature of the success criteria adopted. In this comment, we explain the nature of the problem and attempt to assess its influence on the results reported in the aforementioned paper.

Perceptual errors in the Farwell and Donchin matrix speller

Perceptual errors such as repetition blindness (RB) and attentional blink (AB) can affect the morphology of the target-P300. These perceptual errors may occur with the use of the Farwell & Donchin P300 speller protocol and in turn interfere with the desired symbol identification. In this paper a state of the art classification algorithm is employed to classify brain-computer interface (BCI) competition II and III P300 data and its performance used to assess the presence and effect of such phenomena. The general trend of the results indicated that perceptual errors were present and their effect on classification performance was detrimental. Although the degree to which these perceptual errors affected classification performance varied across subjects.

Exploring multiple protocols for a brain-computer interface mouse

In recent years, various visual protocols have been explored for P300-based BCI. In stimulus-driven BCI paradigms such as P300 BCIs it is vital to optimise the stimulation protocol as much as possible in order to achieve the best performance. Due to the inherent variability between subjects and the complex nature of the brain it is unlikely that an optimal protocol will be identified through a single iteration of random exploration. That is why in this paper we explore 8 different visual protocol configurations based on recent literature, in the hope of identifying key features that can later be used to create further improved protocols. Results indicate that luminosity changes, the standard method of stimulation used in visual P300 BCI protocols, do provide the best performance of the variations presented here.

Evolution of a brain-computer interface mouse via genetic programming

We propose the use of genetic programming as a means to evolve brain-computer interfaces for mouse control. Our objective is to synthesise complete systems, which analyse electroencephalographic signals and directly transform them into pointer movements, almost from scratch, the only input provided by us in the process being the set of visual stimuli to be used to generate recognisable brain activity. Experimental results with our GP approach are very promising and compare favourably with those produced by support vector machines.

Eigenbrains: The free vibrational modes of the brain as a new representation for EEG

We present a new transform for EEG signals whose basis functions are well suited to represent the large-scale dynamics associated with event related potentials. The method involves instantiating an approximate model of the electrical properties of the brain as a conductor medium and then studying the free vibrational modes of the model. These form a set of basis functions, which we call eigenbrains, that can be used to meaningfully re-represent the brains electrical activity. Eigenbrains are compared to principal component analysis and independent component analysis to highlight differences and similarities.

Evolutionary synthesis of a trajectory integrator for an analogue brain-computer interface mouse

Recently significant steps have been made towards effective EEG-based brain-computer interfaces for mouse control. A major obstacle in this line of research, however, is the integration of the noisy and contradictory information provided at each time step by the signal processing systems into a coherent and precise trajectory for the mouse pointer. In this paper we attack this difficult problem using genetic programming, obtaining extremely promising results.