Based on human neurophysiology, it has been shown that the human brain and spinal cord can partly be repaired by movement-based learning. It seems that even to a very limited extent, new nerve cells can be built anew in the human central nervous system. Neural network learning starts with the knowledge of basic human neural network functions and their communication with the outside world. Neural network functions can only be explored thoroughly if it is partly known what impulse patterns run into and out of the networks. Even though the gained knowledge is rudimentary, it has immediate consequences for learning and the repair of the human central nervous system. This book discusses the theory of neural network learning. It provides research on neural network learning rates in healthy patients and patients with central nervous system injuries; neural network learning for coma patients; improving health in geriatric and cancer patients; and improving mental functions in patients with depression and anxiety.