Tracking the signal, cracking the code: speech and speech comprehension in non-invasive human electrophysiology

Magneto- and electroencephalographic (M/EEG) signals recorded from the human scalp have allowed for substantial advances for neural models of speech comprehension over the past decades. These methods are currently advancing rapidly and continue to offer unparalleled insight in the near-to-real-time neural dynamics of speech processing. We provide a historically informed overview over dependent measures in the time and frequency domain and highlight recent advances resulting from these measures. We discuss the notorious challenges (and solutions) speech and language researchers are faced with when studying auditory brain responses in M/EEG. We argue that a key to understanding the neural basis of speech comprehension will lie in studying interactions between the neural tracking of speech and the functional neural network dynamics. This article is intended for both, non-experts who want to learn how to use M/EEG to study speech comprehension and scholars aiming for an overview of state-of-the-art M/EEG analysis methods.