Wavelet theory may be a widely used feature extraction method for raw electroencephalogram (EEG) signal processing. the character of the EEG signal is non-stationary, therefore applying wavelet transform on EEG signals may be a valuable process for extraction promising features. On the opposite hand, determining the right wavelet family may be a challenging step to urge the simplest fitted features for top classification accuracy. during this paper, therefore, we focused on a comparative study of various Discrete Wavelet Transform (DWT) methods to seek out the foremost convenient wavelet function of wavelet families for a non-stationary EEG signal analysis to be wont to classify mental tasks. For the classification process, four different mental tasks were selected to and that we grouped each with another one to line dual tasked sets including all possible combinations. Feature extraction steps are performed using wavelet functions haar, coiflets (order 1), biorthogonal (order 6.8), reverse biorthogonal (order 6.8), daubechies (order 2) and, daubechies (order 4). Later, a selected feature reduction formula is applied