MindFrame
For professionals

The thinking skill your career runs on

Consultants, analysts, lawyers, and clinicians who train their metacognition make fewer costly errors and communicate uncertainty with more accuracy. Four minutes a day. Measurable in six weeks.

Free plan available  ·  15 training modes  ·  620 challenges

The evidence

Why metacognition training is different

These figures come from independent peer-reviewed research — not from MindFrame. The effect sizes are why this is worth four minutes of your day.

g = 0.63
Training effect on real-world performance
Durable across months. Measured in controlled independent studies — not vendor reports. Larger than tutoring, class size reduction, or most professional development interventions.
de Boer et al., 2018 — 67 studies
+22%
Performance improvement from structured reflection
10 minutes of structured reflection after a task improves subsequent performance by 22%. This is the cognitive mechanism that MindFrame systematises into daily practice.
Di Stefano et al., 2016
~40%
Error rate when professionals say they're 99% certain
Overconfidence is not a character flaw — it is a measurable calibration error. Professionals who track and train their confidence accuracy make materially fewer high-stakes mistakes.
Fischhoff et al., 1977

Full evidence base with study links →

Who trains with MindFrame

Your profession, your specific blind spots

📋

Consultants & strategists

You advise clients under uncertainty. Your credibility depends on knowing when you actually know.

  • Calibrated confidence on recommendations
  • Confirmation bias detection before the final deck
  • Structured approach to conflicting data signals
📈

Finance & investment professionals

Every thesis you hold reflects your current beliefs. Most investors don't measure how accurate those beliefs actually are.

  • Brier Score tracking on your own predictions
  • Overconfidence detection on high-conviction positions
  • Pattern recognition across your historical errors
⚖️

Lawyers & legal advisors

Legal judgment requires accurate probability estimation under pressure. Most legal training does not teach you to measure this.

  • Calibration on outcome probability assessments
  • Anchoring and availability bias training for case framing
  • Structured second-opinion thinking habits
⚕️

Doctors & clinical professionals

Diagnostic overconfidence is one of the most studied failure modes in medicine. It responds to training.

  • Overconfidence reduction in diagnostic judgment
  • Base-rate neglect training for symptom presentation
  • Structured uncertainty quantification under time pressure
🔬

Analysts & researchers

Confirmation bias quietly shapes which evidence you weight. Metacognition training surfaces what you're missing.

  • Confirmation bias detection in data interpretation
  • Calibrated uncertainty on model outputs
  • Accuracy self-assessment before presenting findings
🧭

Executives & decision-makers

The higher the stakes, the more your blind spots compound. Accurate self-assessment is the rarest executive skill.

  • Accurate self-assessment of your own reasoning quality
  • Calibrated uncertainty communication to boards and investors
  • Tracking your decision quality over time — not just outcomes

How it works

A session takes four minutes

1

Four minutes of structured challenge

Each session presents 5–20 calibration challenges drawn from the modes most relevant to your identified weaknesses. You answer, rate your confidence, and articulate your reasoning.

2

AI scores your reasoning, not just your answer

MindFrame's scoring engine evaluates outcome accuracy, reasoning quality (via Claude), and calibration (how well your confidence matched your actual accuracy). Three layers, one score.

3

Your Metacognitive Fingerprint builds over time

Every session adds a data point to your Cognitive Fingerprint — a radar showing your relative strength across five dimensions. Your weakest areas drive your next session's challenge selection.

4

Patterns and gaps surface automatically

The analytics engine tracks your calibration trend, most common bias patterns, and improvement velocity over time. No self-reporting — the data is derived from your session performance.

Why MindFrame

Built for professional-grade thinking

🎯

Measures what you actually care about

Your calibration score is the ratio between your confidence and your accuracy — the metric that determines whether your professional judgment is trustworthy. MindFrame is built around this number.

🔬

Evidence-based, not gamified trivia

MindFrame is not Duolingo for thinking. Every challenge is grounded in the cognitive science of metacognition. Effect sizes are cited from independent research, not vendor studies.

🔒

Private by default

Your reasoning text, session answers, and coaching conversations are private. Fingerprint sharing is opt-in. Your employer sees nothing unless you choose to share.

📊

Compounds with consistent practice

Metacognitive ability improves most between sessions 3–20. A four-minute daily session over six weeks produces measurable calibration improvement — this is the window where it clicks.

Not a course. Not a workshop. A practice.

Metacognitive ability does not improve from a one-day bias workshop. It compounds through consistent, feedback-rich practice — the same way physical fitness responds to progressive training. MindFrame gives you the structured reps, the performance tracking, and the AI coach to make it stick.

Common questions

FAQ for professionals

How is this different from an online course on cognitive bias?

Passive awareness of a bias reduces its effect by less than 5% (Morewedge et al., 2015). Active training combining recognition, labelling, and correction reduces it by up to 29%. MindFrame is the active training, not the lecture.

How much time does it take?

A standard session takes 3–5 minutes. You can configure session length (5, 10, or 20 challenges). The product is designed around the constraint that professionals do not have 30 minutes per day for self-improvement.

Is my data visible to my employer?

No. Your session data, reasoning text, and coaching conversations are private to you. If you join a team plan, your manager sees aggregate anonymised team-level data — never individual session content.

How do I know it's working?

Your calibration score — the gap between your confidence and your accuracy — is tracked across every session. A declining calibration error over 4–6 weeks is the measurable signal that training is working.

What is the Metacognitive Fingerprint?

A radar profile built from your session history across five dimensions: calibration, reasoning quality, adaptation, reflection, and outcome accuracy. It updates every session and can be shared as a public link.

Start training today

Free plan includes all core modes. No credit card. The calibration you build in the first six weeks is yours permanently.

15 training modes  ·  620 challenges  ·  AI coach included on Pro