Hippocampus NeoCortex interaction: Long term vs Short term learning

How does the brain learn and memorize? This post explores the contribution of the Hippocampus and Neocortex regions of the brain to understand the neural basis of learning and memory. The inspiration for this post came from an anecdote regarding playing darts and getting better at it. According to my cognitive science professor, when learning a new technique or trying to improve old ones, taking periodic breaks from the learning activity helps a lot. Thus, if you're practicing hitting bulls eyes, practice for 15 minutes and then go do something else. Then come back and practice for 15 more minutes and so on. What actually is going on in the brain during this process?

          A memory and learning system would have two goals: 1) Remember specific information and 2) Extract generalities across many experiences. From a computational standpoint, the architectures for these goals would be different. While remembering specifics would require separate representations that are learned automatically and quickly, representations for extracting generalities will have to be overlapping between multiple representations of specific instances thereby requiring slow learning. Thus, the brain being the wonderful machine it is, has two separate neural architectures that are specifically optimized for the above two goals. 

          While the hippocampus essentially does fast learning where it encodes everything automatically to remember specifics (for instance how did I make the last bulls eye at darts), the neocortex does slower learning to extract generalities (for example, what is the strategy I should follow to keep making bulls eyes). Thus, in the darts playing example, as you keep practicing, the specific shots you make keep overwriting your previous attempts because the hippocampus is now engaged in extracting specifics about the situation. On the other hand, when you go and sit down, and do some other task, the neocortex takes over and extracts generalities from the experience of the past 15 minutes. Thus, since your goal is to get better at hitting bulls eyes, the neocortex goes over all the times you succesfully hit a bulls eye and does task driven learning. It tries to accumulate the experience out of all the different representations (in this case the conditions where you hit a bulls eye) and integrate them into a strategy. 

           In conclusion, the next time that you're trying to learn/get better at a motor skill (I am not sure whether this approach works for mental tasks), practice for a while so that the hippocampus creates all these representations of how to perform the task, then go sit down and think about something totally different so that the neocortex can take over behind the scenes and generate a strategy from all these representations.