Procedural Memory

Procedural memory is the long-term memory system that stores how to do things, from riding a bike to executing a practiced peek, as opposed to declarative memory, which stores facts and events that can be consciously recalled and stated. Cohen and Squire's (1980) study of amnesic patients provided the clearest dissociation between the two: patients with severe damage to declarative memory, unable to recall having practiced a mirror-tracing task from one day to the next, still showed steady, measurable improvement on that same task session after session. The skill itself was being retained and refined by a separate memory system, even though the patients had no conscious memory of ever practicing it. This double dissociation is the foundation for treating skill learning and fact learning as distinct brain systems rather than one general memory faculty.

Reference: Cohen & Squire, 1980. The NeuroRank implementation holds the canonical form and scales interference via task-irrelevant stimulus density.

1. Retaining an improved mouse-control skill across play sessions even when a player cannot articulate what specifically got better.
2. Executing a well-drilled rotation or macro pattern correctly under pressure despite being unable to recite its steps out loud.
3. Improving at a mechanic through blind repetition, without ever consciously studying a guide or tutorial for it.

A single NeuroRank combine cannot observe procedural learning directly, since that requires comparing performance across sessions separated by practice, not one sitting. The platform's retest flow, where a returning player's scores can be compared against an earlier profile, is the feature most suited to surfacing procedural-memory gains, though NeuroRank does not currently isolate skill consolidation from other session-to-session variance.