Effect size g = 0.63 on academic performance. That's larger than homework, class size reduction, or most one-on-one tutoring. Metacognition is trainable. MindFrame is how you train it.
Free plan includes all 19 training modes · No credit card · 4–5-minute sessions
The academic performance evidence
This isn't about IQ. It's about knowing when you know something vs when you're guessing — and most students have never been taught to tell the difference.
How to read g: Hedges' g is a standardised effect size. g = 0.20 small · 0.50 medium · 0.80+ large. g = 0.63 means the trained group outperformed about 73% of the untrained group. These are school studies, not MindFrame results. Full evidence base →
What metacognition actually fixes
Fluency feels like understanding — but recognising a concept when you read it is very different from being able to retrieve and apply it. MindFrame's calibration training exposes this gap before the exam does.
Classic calibration studies found that answers people feel near-certain about are wrong far more often than their confidence implies. Calibration training directly corrects this by giving you precise feedback on the distance between your confidence and your actual accuracy.
Strategy monitoring — tracking whether your approach is actually working — is a core metacognitive regulation skill. MindFrame's session analytics show which thinking patterns are improving across weeks.
When working memory is overwhelmed by anxiety, thinking quality degrades. Systematic metacognitive practice builds resilient thinking habits that hold under pressure.
How MindFrame helps
Catch the thinking errors that lead to wrong exam answers — overconfidence, availability bias, confirmation bias. Recognising a pattern in your own reasoning is the first step to correcting it.
Know when you're ready vs when you're fooling yourself. The direct antidote to the illusion of knowing — you rate confidence before seeing the result, so you build an accurate internal compass.
Make better choices under time pressure. Structured practice in applying clear reasoning when options are ambiguous — useful for essay planning, case studies, and MCQ elimination.
Understand how you actually think through problems. A few minutes of structured self-examination that builds the habit elite learners use to compound their improvement week over week.
Quantify and track your foresight over time. Make predictions, revisit them, and measure how well your confidence matched reality — the skill at the core of any high-stakes assessment.
What your scores actually mean
Imagine taking practice tests and rating your confidence on each answer — say, 60% certain, 90% certain, 40% certain. Your Brier Score measures how well those confidence ratings matched your actual results. A score of 0 is perfect. A score of 1 is maximally wrong.
Most untrained students sit around 0.20–0.35. They're more overconfident than they think — they feel 90% sure about things they only get right 60% of the time. Every MindFrame session updates your Brier Score, so you always know exactly where you stand.
Fitting it into your routine
Research shows distributed practice is 2–4× more effective than massed studying. MindFrame is designed for short daily sessions (about 4–5 minutes) that add up over weeks — not another app demanding your evenings.
1 Calibration Lab challenge before your first study block
Primes your self-monitoring before you encounter new material
1–2 Bias Hunter challenges during a 5-minute rest
Exposes the thinking errors most likely to surface in your subject area
1 Reflection Sprint after your last study session
Consolidates what you learned and flags what to revisit tomorrow
MindFrame vs flashcard apps
Common questions
The Brier Score measures how accurate your probability forecasts are: (stated probability − actual outcome)², averaged. 0 is perfect; 0.25 is what you get by always saying 50%. Top forecasters in the Good Judgment Project held scores around 0.10–0.12. MindFrame shows yours after every session so you can see whether your self-assessment is getting sharper.
We have not yet measured MindFrame's effect on its own users, so we do not promise a timeline. Published calibration research shows feedback on confidence versus accuracy can improve calibration with deliberate practice. Your results page shows your Brier Score and Calibration Error after every session and your trend over time, so you can see for yourself whether they move.
A standard session takes about 4–5 minutes; quick sessions take 2–3. You answer a short set of challenges, rate your confidence on each, and optionally explain your reasoning. Then you get your scores and one concrete next step.
MindFrame trains the thinking process, not subject knowledge: noticing when you are overconfident, checking assumptions, updating on evidence. Whether those habits carry over to a given subject is something you can watch in your own calibration trend; we do not claim a measured transfer effect yet.
All 19 training modes with no paywall on sessions, every game, core analytics, calibration scores, the Cognitive Fingerprint, core badges and 3 AI coach messages a day. No credit card required.
Every training mode and game is already free, which covers most students. If you're using MindFrame for a course, a research study or a nonprofit programme and need Season Passes for a group, email [email protected].
No credit card required. All 19 training modes included. About 4–5 minutes a session.
Create your free account →Already training? Read the research behind MindFrame