Development of Wavelet-Based Tools for Event Related Potentials’ N400 Detection: Application to Visual and Auditory Vowelling and Semantic Priming in Arabic Language

摘要

Neurological signals are generally very weak in amplitude and strongly noisy. As a result, one of the major challenges in neuroscience is to be able to eliminate noise and thus exploit the maximum amount of information contained in neurological signals (EEG…). In this paper, we aimed at studying the N400 wave of the Event-Related Potentials (ERPs) that may reflect the effects of vowelling and semantic priming in Arabic language. To improve the quality of the recorded ERP signals, we considered a nonlinear filtering method based on 10th order Daubechies discrete wavelet transform combined to principal component analysis (PCA). Among all tested wavelets, the Daubechies one showed high values of the used signal processing metrics. Thus, it allowed a significant enhancement of the signal to noise ratio while using only 10 ERP trials. In addition, we confirm its effectiveness while comparing the filtered outputs to those obtained using the averaging technique implemented in the conventional EEGLab toolbox. In a second step, the Mexican Hat function was used to achieve continuous wavelet analysis of the filtered signals. This time-scale analysis method permitted to get an alternative representation of the ERPs and to detect the N400 wave with significantly greater accuracy.