Bolt Series · Human Performance Calibration · Article 22
Experts do not necessarily see more. They often see what matters sooner.
Watch an expert in a fast sport and something can look almost supernatural.
The ball has barely left the opponent.
The expert is already moving.
The beginner reacts later.
It is tempting to say the expert has faster reflexes.
Sometimes that may be part of the story.
But decades of research on perceptual-cognitive expertise show something more interesting.
Experts are often better at identifying the cues that actually predict what is about to happen.
Expert anticipation is a real phenomenon
A major meta-analysis of perceptual-cognitive expertise in sport found that experts outperformed non-experts in response accuracy and response speed when extracting information from sport-specific situations.
Experts also showed systematic differences in visual search.
They did not simply look everywhere harder.
They often used fewer, longer fixations and extracted more useful information from the available scene.
A 2022 systematic review and meta-analysis in combat sports found large expert advantages in anticipation accuracy and faster response times.
A 2025 review of goalkeeping similarly found that higher-level goalkeepers showed superior anticipation and more effective cue use.
Expert prediction is therefore not just confidence.
It can reflect better information pickup.
The cue matters
Consider a bowler in cricket.
A novice may watch the whole movement and still fail to know where the ball will go.
A highly skilled player can use particular information from the hand, arm and movement pattern to anticipate earlier.
Research on expert anticipation has shown that this advantage can disappear when the normal sequence of useful cues is disrupted.
That tells us something important.
The expert advantage is not magic. It depends on sensitivity to information that predicts the outcome.
Sprinters can become better observers of sprinting too
A 2023 study compared sprinters with non-sprinters on a visual discrimination task involving videos of walking and sprinting.
The sprinters were more accurate at detecting movement differences.
Participants who identified a particularly informative cue performed better, and the trained sprinters benefited more from that cue than non-experts.
This fits the larger pattern.
Experience can teach a person not merely to perform a system, but to notice what is diagnostically important inside it.
Now move this into education
Two students look at a Mathematics question.
The novice notices:
“There are lots of words.”
The more expert learner notices:
“The quantity is changing at a constant rate, so this is probably a linear relationship. The distracting information is not part of the model.”
Both students are looking at the same question.
They are not extracting the same information.
Prediction quality changes because cue quality changes.
Experts also have better histories to compare against
An experienced performer has seen more cases.
More successful attempts.
More failures.
More variations.
More feedback.
That history can improve prediction because the current event is compared with a richer library of previous events.
A beginner may think:
“This feels difficult.”
An expert may think:
“This feels difficult in the same way that happens when my timing is late, not when my technique has broken down.”
The second judgement contains more categories.
More categories can produce better predictions.
But experience alone does not guarantee expertise
This is an important safeguard.
A person can repeat the same activity for years and still become a poor judge.
A 2025 paper on athlete selection makes the point sharply: experience does not automatically create expertise when the environment provides weak cues or poor feedback.
For judgement to improve through experience, the person needs opportunities to learn from cues that genuinely predict outcomes and feedback that is timely and informative enough to correct mistaken assumptions.
This matters enormously in education.
A student can complete hundreds of questions and still learn the wrong lesson if the feedback never reveals why an answer succeeded or failed.
Experience becomes expertise only when the environment allows prediction errors to teach.
Expert predictions are often domain-specific
Another safeguard is equally important.
Being beautifully calibrated in one domain does not make someone a universal predictor.
Recent metacognition research finds some shared structure across tasks, but also substantial domain specificity.
A 2026 study across multiple tasks found only weak evidence for a fully domain-general metacognitive ability, with stronger clustering within particular kinds of tasks.
That makes intuitive sense.
An expert sprinter may be exquisitely calibrated about sprint readiness and ordinary at judging an unfamiliar financial decision.
A brilliant mathematician may know exactly when a proof is secure and still misjudge their ability in a social situation.
Calibration has to be earned against the relevant world.
Why experts can still be wrong
Experts can overfit to familiar patterns.
Conditions can change.
The signal can be noisy.
A rare event can violate the usual pattern.
And expertise itself can create confidence that discourages reconsideration.
So the lesson is not:
Experts know.
It is:
Experts can become better predictors when expertise teaches them which cues matter, what patterns repeat and how to update from feedback.
What this looks like in a strong learner
A strong learner does not merely say:
“I think I’m ready.”
They can increasingly say:
“My retrieval is secure because I can reproduce it after a delay. My transfer is less secure because I still miss the method when the question is disguised. I expect to do well on routine items but not yet on the hardest application problems.”
That prediction has better cues.
Better cues produce better decisions.
Why this matters for education
We often teach students answers.
We should also teach them which signals deserve trust.
- Recognition is weaker evidence than retrieval.
- One correct answer is weaker evidence than repeated transfer.
- Effort is not the same as accuracy.
- Feeling fluent immediately after studying is not proof of durable learning.
- A repeated pattern deserves more weight than one surprising event.
Expertise is partly the acquisition of better actions.
It is also the acquisition of better cues for judging those actions.
Experts often predict better because experience has taught them where the real information is.
Education should aim to give learners that same advantage.
Evidence and further reading
- Meta-analysis — Perceptual-Cognitive Expertise in Sport
- Systematic Review and Meta-Analysis — Expert Anticipation in Combat Sports
- Expertise and Training of Anticipation in Goalkeeping
- Action Perception in Athletes: Expertise Facilitates Perceptual Discrimination
- Why Experience Fails to Foster Expertise in Athlete Selection
- Metacognition in Decision-Making Across Domains and Modalities
