Fast Recognition of Anticipation-Related Potentials

Abstract

Abstract-Anticipation increases the efficiency of daily tasks by partial advance activation of neural substrates involved in it. Here we develop a method for the recognition of electroencephalogram (EEG) correlates of this activation as early as possible on single trials which is essential for Brain-Computer Interaction (BCI). We explore various features from the EEG recorded in a Contingent Negative Variation (CNV) paradigm. We also develop a novel technique called Time Aggregation of Classification (TAC) for fast and reliable decisions that combines the posterior probabilities of several classifiers trained with features computed from temporal blocks of EEG until a certainty threshold is reached. Experiments with 9 naive subjects performing the CNV experiment with GO and NOGO conditions with an inter-stimulus interval of 4 s show that the performance of the TAC method is above 70% for four subjects, around 60% for two other subjects, and random for the remaining subjects. On average over all subjects, more than 50% of the correct decisions are made at 2 s, without needing to wait until 4 s.

Learning From EEG Error-Related Potentials in Noninvasive Brain-Computer Interfaces

“…Moreover, it takes advantage of a unique characteristic of the brain channel that carries information about cognitive states of the user. Future developments along this line may take further advantage of this characteristic to improve human-machine interaction by exploiting other cognitive states or processes such as image processing [35], anticipation of future events [36], fatigue [34], or mental workload [37].…”

Discriminative channel selection method for the recognition of anticipation related potentials from CCD estimated cortical activity

“…Number of channels selected (K) for EEG (blue line with squares) and CCD (red line with triangles) for different thresholds of P inverse methods for the classification mental imagery [1,2] and cognitive states [3]. In the current paper we studied the use of CCD inverse method for the recognition of anticipation related potentials which are shown to be exploited for BCI [9,10,11]. One of the challenges in working with estimated cortical potentials is that the number of available channels get multiplied drastically and a method for choosing a subset of them for calculating classifiers is essential.…”

“…For simulating anticipatory and nonanticipatory behaviors we used a CNV paradigm with relevant (GO) and irrelevant (NOGO) conditions ( See Fig. 1 for the CNV paradigm used in the current study and refer [10] for more details).…”

“…In EEG this state appears mainly as an increasing negativity under the site of vertex electrode (Cz) which can be interpreted as pre-activation of neural substrates involved in executing actions on the occurrence of events in future [6,7,8]. In our previous studies, we argued that single trial recognition of anticipation related potentials can be exploited for BrainComputer Interaction (BCI) [9,10], and in a recent pilot experiment we demonstated a prototype of anticipation based BCI (aBCI) [11]. These studies were based on features computed using activity of the Cz electrode alone.…”

EEG error-related potentials detection with a Bayesian filter

“…The final classification is provided by weighted combination of the class posteriors. More recently, probabilistic combination of classifiers was proposed for the recognition of anticipationrelated EEG potentials in a CNV paradigm [6].…”