Computational Processes that Appear to Model Human Memory.

This paper presents two computable functions, omega and epsilon, that map networks into networks. If all cognition occurs as an active neural network, then it is thought that omega models long-term memory consolidation and epsilon models memory recall. A derived, intermediate network form, consisting of chordless cycles, could be the structural substrate of long-term memory; just as the double helix is the necessary substrate for genomic memory.