EEG PREDICTABILITY - ADEQUACY OF NONLINEAR FORECASTING METHODS

The predictive properties of EEG segments were analysed. The sample in cluded alpha, delta as well as spike and wave EEG activity recordings. Most of these segments are better described with non-linear autoregre ssive models, and a non-linear forecasting algorithm is routinely requ ired. In terms of their predictive properties, segments can be divided into unpredictable, predictable and very predictable, these three gro ups being similarly represented among the alpha activity EEG segments. In EEG segments with alpha activity, poor predictability is associate d with poor organization of the rhythmic pattern. Concerning dynamic p roperties, it was found that cyclic skeletons were highly represented among the very predictable segments, which reflect a contribution of t he deterministic component of the autoregressive model to the predicta bility of the segments. Notable contributions of the noise component m ay explain the properties of unpredictable segments. These results poi nt to a great diversity of predictive patterns among EEG recordings. O ther factors besides the existence of chaotic dynamics must be regarde d.