A principal difficulty in trying to understand brain function, and dysfunction, is that the research entails everything from the behavior of molecules up to the behavior of people. Not only must this information be gathered, it must also be linked together to provide explanation and prediction. Computational neuroscience has emerged as a set of concepts and techniques to provide these links for findings and ideas arising from disparate types of investigation and at disparate spatial and temporal scales.
The extraordinary success of artificial neural networks and computer learning seems to demonstrate that the very recent accomplishments of the human mind -- eg playing chess, go or Jeopardy -- are not the things that are really hard to do. They are also, of course, largely not the things that evolution as pressured brains to be able to do. Instead, it's the seemingly simple things that we share with many other animals -- acute visual and auditory perception under multiple conditions, control of locomotion across uneven terrain at different speeds -- that really bring into play the complex processing for which brain coding is optimized. Lack of perceptual and motor skills have been major impediments to the development of useful robots. We have been trying to develop biomimetic models that can communicate with brains and also control a robot arm.