E ects of Variations in Neural Network Topology and

Electroencephalogram, or EEG, signals are an important source of information for the study of underlying brain processes. Such studies now provide a framework for the development of a new modality of human-computer interaction based on EEG. Current research in this area only detects a small number of mental states. In this article, EEG from one subject who performed three mental tasks are classiied by neural networks. Using a sixth-order autoregressive (AR) model of half-second windows of six-channel EEG, a classiication accuracy of 89% on test data is achieved. A cross-validation study of a variety of neural network topologies showed that a network with one hidden layer of 20 units produced the best performance. It was also found that averaging the output of the network over consecutive inputs improved performance. K-means clustering of the resulting neural networks' weights identiied key components of the AR representation.