Practice against what you represent: compare each attempt to the target

How you practice depends on what you represent — evaluate each attempt against your internal model of the target, not against how it felt.

Why it works

In Anders Ericsson’s account of deliberate practice, mental representations are the pivot: practice builds them, and they in turn make better practice possible. A mental representation is a domain-specific structure — a chess player’s sense of a position, a musician’s sense of how a phrase should sound — that lets an expert perceive a situation as a meaningful whole rather than a set of parts, and Ericsson argued this is what most distinguishes experts from committed amateurs. The practical consequence is that the feedback loop needs two things, not one: an accurate representation of the target, and an accurate read of what you just did. Most learners have only the second, and even that only as a feeling — "that didn’t sound right" — without being able to name what was wrong. That is the point at which practice stops improving anything, because a rep you cannot diagnose is a rep you cannot correct. The loop is also why the two develop together: sharpening the representation sharpens the gap analysis, and each diagnosed gap sharpens the representation further. It is the same reason a novice benefits so much from a coach — the coach is temporarily supplying the representation the learner has not yet built.

How to do it

1. Before each practice session, vividly rehearse the target representation — what it should look, sound, or feel like when done correctly.
2. After each attempt, articulate specifically what matched and what did not, in words rather than as a verdict.
3. If you cannot name the gap, that is a signal your representation is too coarse — go study the target (a recording, a model, an expert) before doing more reps.
4. Focus corrections on the most significant specific gap, not on general improvement.

Evidence

Specific, knowledge-based feedback is consistently more effective than general or affective feedback in skill acquisition research. The representation-to-attempt comparison model is Ericsson’s theoretical account of why deliberate practice works — empirically consistent but theoretical in form. (mechanistic)

Measuring "mental representation quality" directly is not possible; this is inferred from performance characteristics. The mechanism is principled and consistent with what is directly observable.

Sources

• Ericsson, Krampe & Tesch-Römer (1993), "The role of deliberate practice in the acquisition of expert performance", Psychological Review

Common mistake

Ending each attempt with "that was okay" or "that felt off" without being able to specify what the gap is — leaving the practice loop without the targeted correction that drives the next rep.