Learning
Feedback Loop
An outcome teaches something when you understand its meaning. A feedback loop connects trying, noticing and adjusting without flooding every attempt with advice.
- Attempt
- Information
- Assessment
- Adjustment
An attempt produces information that changes the next attempt. The returning connection shows information actually influencing action.
What changes after the correction?
Your coach gives a useful cue. You nod, then do exactly the same thing again. The feedback arrived, but the loop is not closed yet. A feedback loop connects an attempt, information about its consequence and a changed next attempt. That information can come from a coach or video, but also from what you feel yourself, such as losing balance. Hearing good or bad is less useful than knowing what you are trying to change now.
Feedback loops are a broad concept in fields including control engineering and learning research. There is no reason to name one sporting coach as their inventor. Augmented feedback research examines information added to what a learner perceives themselves. Sigrist and colleagues' review discusses different forms and design questions. Suitable content and timing depend on the task and learner.
Making a recurring balance problem visible
A boxer repeatedly loses balance after a combination. They believe the final punch is too slow, while the coach notices an earlier step traveling too far forward. A short video can compare these observations. The coach selects one point of attention and asks for several gentle attempts.
Before another correction, ask what the boxer noticed. As independent perception improves, feedback may become less frequent. The loop strengthens because the learner begins reading relevant consequences themselves. Continual commentary can instead direct all attention towards the coach.
Immediate is not invariably better
A safety problem may require an immediate stop and explanation. In an ordinary learning task, a short delay can allow independent assessment of the experience. Feedback after several attempts may sometimes help more than correcting every movement. There is no general rule that the fastest feedback always produces the best learning outcome.
Success may also contain chance. One winning exchange does not establish a sound choice, and one failure does not disprove an idea. Examine several attempts under appropriate conditions. Resulting Fallacy helps distinguish outcomes from decision quality.
Use additional information deliberately. A video identifying five errors without priorities provides more data but not necessarily a better next attempt. First establish the question the information should answer. A learner should be able to explain what they intend to change.
Close the loop with another attempt
Choose a recurring problem and agree what information you need. Make several attempts, describe your own perception first and compare it with your partner or coach. Change one element and observe again.
If feedback does not influence another attempt, it remains an isolated comment. Deliberate Practice uses the loop within targeted improvement. Scientific Method makes predictions and tests more explicit. Kaizen retains a useful adjustment in everyday practice.
Sources and further reading
- Sigrist et al.: Augmented feedback in motor learning
Review of feedback forms and design considerations; not a universal immediate-feedback rule.
See an idea alongside another
Related models
Deliberate Practice
Choose a specific task where a correction is followed by another attempt.
Scientific Method
Know which prediction the feedback should help you check.
Resulting Fallacy
Do not judge the choice only by one exchange won or lost.
Kaizen
Keep the adjustment that works in your normal practice.