What is pattern separation?
Pattern separation can be defined in many way and coming up with good operational definitions is part of the mission statement. However, a reasonable first pass is something like “the process of reducing interference among similar inputs by using non-overlapping representations“. In the brain, this is represented by using distinct neural codes.
Although this definition seems simple and straightforward, it is far from it, and there are lots of nuances and caveats. For example, what do we mean by “distinct neural codes”? What is “interference”? How is quantified? How do we determine if a region or cell ensemble performed pattern separation? What kind of data can allow us to conclude that such a computation occurred? In other words, what is its neural basis? What kind of flexible behavior does it support? Etc…
Patternseparation.com was created with two things in mind:
1. To provide a universal resource on pattern separation as a neurocomputational process including articles, readings, post-publication peer-review and discussions of various findings.
2. To provide an interface for interaction among colleagues who do work in this area of investigation and discuss pertinent issues related to pattern separation. This includes discussions of operational definitions of pattern separation at several levels (computational, neural, behavioral) and the applications of findings in this realm to translational investigations of clinical disorders.