This research shows that there is a heavy and slow write path and a secondary read path that is quick. This affects response time for memory recall.
Reminds me of database design.
I think this is useful and groundbreaking research, but when articles use words like "neural circuit" it frustrates me. Circuit is a word with a specific meaning, and to talk about neural circuits seems to be projecting an implicit model that we don't really have justification for. Also the imprecision of the term "memory", in the sense of "this is where a memory is stored" seems also to overstate what is being measured, which is of course, behavior in response to stimuli. That is related to memory in the broad sense, but is it a memory? Is it even reasonable to think of memories as distinct and discrete things?