Memory replay & reactivation

Replay is the sequential re-expression, during rest and sleep, of neural activity patterns that occurred during a prior experience — for example a rodent's place cells firing in the same (or time-reversed) order as when it ran a track, but time-compressed roughly ten- to twentyfold and often bundled inside sharp-wave ripples. Reactivation, more broadly, denotes any post-experience recurrence of an experience-related pattern, whether or not strictly ordered.

Replay is a candidate mechanism for consolidation and for offline planning, and it is central to prosthetic ambitions: if a system can detect that a particular memory is being replayed, it can in principle reinforce, bias, or interfere with that specific memory. Reading replay content in humans remains hard — it requires simultaneous single-unit or high-resolution population recordings and decoders that generalize from awake activity to compressed offline sequences, and most human evidence is indirect (pattern-similarity or MEG-based decoding rather than literal spike-sequence replay).