Evaluation of Eeg Power Spectrum Measures Using Fourier and Wavelet Based Transformation Techniques as a Function of Task Complexity

The purpose of this study is to provide a comparison between the widely applied Fourier Power Spectrum (FPS) measure and the Wavelet counterpart (WT) in applications related to continuous EEG/ MEG signal analysis tasks. Both power spectrum measures are used as EEG signal features capable of characterizing and differentiating four discrete tasks in terms of cognitive complexity. This study is not oriented towards the investigation of the brain areas involved in such cognition operations. The latter test is providing the dataset to perform the comparison between the different analysis methods. The different analysis methods were evaluated by testing the tasks for statistical significance using ANOVA on both the dependent Fourier Power Spectrum and Wavelet Power Spectrum (WPS) variable. The results of this study indicated that even if FPS already provides satisfactory results, it seems that Wavelet Averaged Global Power Spectrum is proved to be of particular interest since it provides greater resolution and specificity. WT is able to discriminate in greater detail the different tasks.