Memory consolidation

In the brain

A memory isn’t fixed the moment it forms. Over hours to years it’s reorganized: the hippocampus captures an episode fast, then replays it — especially during slow-wave sleep, carried on sharp-wave ripples — to the neocortex, which slowly integrates it into long-term knowledge^{Diekelmann 2010}.

Whether the hippocampus ever fully hands off, or is always needed to replay a detailed memory, is still debated⚖^{Nadel 1997}.

The model

The Complementary Learning Systems account: a fast hippocampal learner and a slow neocortical one. Interleaving replay of old and new lets the slow system extract structure without catastrophically overwriting what it already knew^{McClelland 1995} — the reason sleep matters for memory at all.

In AI

In practice: an offline “sleep” pass that merges redundant memories and extracts patterns. engram (now Counterparts) runs a consolidation cycle between sessions (merge duplicates, distill old episodics to gist); Microsoft filings cluster a vector store with DBSCAN/HDBSCAN and select by age and effectiveness. engram’s two-pass design — a cheap model finds merge candidates, a stronger one rewrites them — mirrored the fast/slow split directly; Counterparts keeps the split: arithmetic merges exact duplicates, and a nightly dream merges near-copies in better words, keeping the originals, so a whole dream can be undone.