To investigate neural mechanisms of human psychology with electroencephalography (EEG), we typically instruct participants to perform certain tasks with simultaneous recording of their brain activities. The identification of task-related EEG responses requires data analysis techniques that are normally different from methods for analyzing resting-state EEG. This review aims to demystify commonly used signal processing methods for identifying task-related EEG activities for psychologists. To achieve this goal, we first highlight the different preprocessing pipelines between task-related EEG and resting-state EEG. We then discuss the methods to extract and visualize event-related potentials in the time domain and event-related oscillatory responses in the time-frequency domain. Potential applications of advanced techniques such as source analysis and single-trial analysis are briefly discussed. We conclude this review with a short summary of task-related EEG data analysis, recommendations for further study, and caveats we should take heed of.

Electroencephalography (EEG) is a powerful tool for investigating the brain bases of human psychological processes non-invasively. Some important mental functions could be encoded by resting-state EEG activity; that is, the intrinsic neural activity not elicited by a specific task or stimulus. The extraction of informative features from resting-state EEG requires complex signal processing techniques. This review aims to demystify the widely used resting-state EEG signal processing techniques. To this end, we first offer a preprocessing pipeline and discuss how to apply it to resting-state EEG preprocessing. We then examine in detail spectral, connectivity, and microstate analysis, covering the oft-used EEG measures, practical issues involved, and data visualization. Finally, we briefly touch upon advanced techniques like nonlinear neural dynamics, complex networks, and machine learning.