“Decoding the variable velocity of lower-limb stepping movements from EEG” by Korik et al. introduces a new deep learning technique for processing EEG to analyze the velocity of different parts of the lower leg, as well as examining the topographical activity and functional connectivity of key brain regions. The network developed in this study combines convolutional neural networks (CNNs) and long short-term memory (LSTM) units [in hybrid architecture. Subjects were split into two groups: one that received verbal cues for forward movement only and one that received verbal cues for forward and backward movement. The fibular head, the mid-fibular point, and the lateral malleolus were used as lower-leg markers. The technique developed by Korik et al. was compared to typical linear regression analysis, and this technique had a higher decoding accuracy overall, with an especially high accuracy on the forward-backward direction and for the fibular head. The topographical mapping and functional connectivity identified brain regions and patterns of activation that were significant between groups, which likely relates to the difference in auditory cues.