1. RNP
  2. Motor Learning
  3. Feedback, guidance and autonomy
  4. The guidance hypothesis: why your help works today and betrays you tomorrow

Motor LearningSeries 10/13

SeriesFeedback, guidance and autonomy10 of 13

The guidance hypothesis: why your help works today and betrays you tomorrow

The guidance hypothesis explains why overly frequent feedback guides immediate performance but undermines lasting learning.

LabO RNP·4 min read·reviewed on

In short. The guidance hypothesis holds that overly frequent guidance or feedback improves performance during practice but degrades long-term learning, because it reduces error instead of teaching the system how to deal with it.

You're spotting the weightlifter with the bar overhead. Your hand brushes their elbow, your voice corrects the angle in real time. The rep is clean. You can feel it's clean under your fingers. Then you take your hand away, stop cueing, and the third rep falls apart. The movement that looked learned only existed because you were there. That is exactly what the guidance hypothesis describes.

What the guidance hypothesis actually says

The guidance hypothesis highlights a central phenomenon of motor learning. When you guide a lot, performance goes up. The movement gets clean, the error disappears, the learner succeeds right in front of you. The problem comes later, when you're no longer there.

The mechanism is trickier than it looks. Guidance reduces error, and that is precisely the issue. An error that gets suppressed is an error the system learns nothing from. You hand over a clean trajectory, but you take away the information the learner would have extracted by getting it wrong on their own. You're not creating learning, you're creating dependency.

What improves immediate performance is not necessarily what promotes learning. The guidance hypothesis is the starkest illustration of that rule, closely linked to the performance illusion that any overly dense assistance produces.

The professional mistake everyone makes

Mistaking success under guidance for autonomous skill. It's the most widespread mistake in the field, and the most costly.

A guided learner can produce a correct action without ever understanding why it works. They execute the right form, but the right form doesn't belong to them. It belongs to you. The same scenario plays out everywhere. You guide, the learner succeeds, you remove the guidance, the learner falls apart.

Look at the athlete who trains under constant coaching. Every rep corrected, every set commented on, every movement adjusted out loud. In practice, they're flawless. Then competition arrives. Without you in their ear, the system no longer knows what to do, because it never learned to guide itself. The guidance prevented perceptual autonomy from developing.

Same mechanics with the child who is physically guided and succeeds at the task as long as the hand is there, then fails the moment it's gone. Same mechanics with the person who progresses under manual assistance but carries nothing over into daily life. In all three cases, the movement happened, but the sensory information was never picked up by the one who was moving. Transfer of learning never took place.

The RNP training space, free for life

Get access to RNP training, no strings attached.

Welcome! Introduce yourself

SecureNo cardNo password

By continuing, you agree to our Terms of Use and our Privacy Policy.

Why feedback frequency changes everything

If the guidance hypothesis sounds abstract, it becomes concrete as soon as you look at how feedback is dosed. Constant feedback favors immediate performance. Intermittent feedback favors retention. The motor learning evidence all points the same way.

Overly frequent feedback reduces exploration, smothers informative error and builds artificial anticipation. The system waits for your input before it has even interpreted its own sensations. It outsources its judgment.

When feedback is given matters as much as how much. Immediate feedback reassures but prevents self-assessment. Delayed feedback forces the system to interpret its own reafference before receiving information from outside. Good timing makes self-assessment come before correction. That's the whole point of feedback frequency and timing, and the reason why the type of feedback, KR or KP, and its source, internal or external, determine what the system actually retains.

The RNP reading: guidance has an exit date

For some people, guidance is necessary at the start. Perception that's too noisy, postural insecurity, neurosensory defensiveness. A system whose sensorimotor loop isn't yet reliable copes poorly with silence and reads the absence of feedback as a threat. Removing all help too early doesn't produce autonomy, it produces confusion.

So the question is never about removing guidance on principle. It's about knowing when to remove it.

That's where RNP becomes an exit criterion for guidance. It tells you when the system is ready to explore on its own, when internal calibration has become readable enough for the learner to do without you. Extending guidance beyond that window doesn't consolidate anything, it locks in dependency. You read the state of the system before deciding, instead of applying a dogma. The line with feedback dependency is drawn exactly here, in the expiry date you set for your help.

How to avoid building dependency

Getting out of the guidance trap can't be improvised. It's planned from the very first session, keeping in mind that every piece of feedback must have a scheduled end.

Fade feedback gradually rather than cutting it off all at once. Set up intrinsic success criteria, the ones the learner can feel without you. Accept functional error as a source of information, not a failure to be erased. And encourage self-assessment by letting the learner describe what they felt before you confirm.

Autonomy is never given, it's built, and it's built slowly. Not in one session, not in a month. It takes patience and tolerance for a temporary regression. Removing guidance will lower performance before it raises learning. That's the price of real autonomy, and it's a good sign.

Guidance is scaffolding. If you forget to take it down, the building never stands on its own.

In one sentence: the guidance hypothesis reminds you that feedback that's always there buys today's performance by mortgaging tomorrow's learning.

Field test. Remove all guidance for three reps and watch. If quality collapses immediately, you hadn't built learning, you were maintaining dependency. If it holds, or degrades slowly with self-correction, the system is starting to guide itself.

Scope: a functional, non-medical approach. It complements healthcare follow-up; it does not replace it.

€0Your free RNP PathsStart