Transfer entropy (TE) is introduced as an alternative measure of effective connectivity in neuroscience, based on information theory. Unlike model-based approaches such as Granger causality or dynamic causal modeling, TE does not require a specific model of the interaction and is inherently non-linear. The authors investigate the applicability of TE to electrophysiological data, specifically magnetoencephalography (MEG) recordings from a simple motor task. They demonstrate that TE improves the detection of non-linear interactions and is robust against linear cross-talk, which is common in MEG data due to volume conduction. The study shows that TE can effectively detect causal relationships between brain regions, even in the presence of signal cross-talk and non-linear dynamics. The method is validated through simulations and applied to real MEG data, providing a robust tool for analyzing effective connectivity in neuroscience.Transfer entropy (TE) is introduced as an alternative measure of effective connectivity in neuroscience, based on information theory. Unlike model-based approaches such as Granger causality or dynamic causal modeling, TE does not require a specific model of the interaction and is inherently non-linear. The authors investigate the applicability of TE to electrophysiological data, specifically magnetoencephalography (MEG) recordings from a simple motor task. They demonstrate that TE improves the detection of non-linear interactions and is robust against linear cross-talk, which is common in MEG data due to volume conduction. The study shows that TE can effectively detect causal relationships between brain regions, even in the presence of signal cross-talk and non-linear dynamics. The method is validated through simulations and applied to real MEG data, providing a robust tool for analyzing effective connectivity in neuroscience.