Thinking isn’t
a ladder.
It’s a loop. ✳
AI can make the output.
You make the judgment.
A lab to investigate your thinking. A gym to practice it. Make your questions, decisions, evidence, and revisions visible.
Beyond the pyramid.
Into the loop.
Keep the cognitive operations. Release the fixed sequence. Put the problem, human agency, and curiosity at the center.
Pyramid → release → recursive loop. An original animated adaptation inspired by the UnBlooms Cognitive Fitness Gym. The starting pyramid represents the revised taxonomy, not Bloom’s original drawing.
Train the thinking.
Show the work.
Choose a short exercise. You can work with your own draft, a peer’s work, or AI output. Each rep ends with a decision you can explain.
Reps count recorded decisions, not mastery.
Your decision trail
0 recorded repsYour trail stays in this browser when storage is available. No automatic grading. Download a copy to keep it.
Four movements.
Infinite ways through.
Start where the problem asks you to. Return when evidence changes your mind. Reflection travels with you.
At the center: the problem.
Throughout: agency + metacognition.
Don’t grade
the output.
Grade the thinking.
But build the infrastructure first.
A polished product can conceal fragile understanding. Decisions, rejected ideas, verification, and revisions reveal what a final answer alone cannot.
UnBlooms builds on Bloom’s taxonomy while moving beyond the familiar pyramid. Learning is recursive, contextual, and anchored in human agency and curiosity.
How did we get here? +
1956 / Shared vocabulary
Bloom and collaborators classify educational objectives, with Evaluation following Synthesis.
2001 / The revision
Anderson and Krathwohl revise the categories into verbs, placing Create after Evaluate.
The AI moment
If a tool can produce the artifact, what evidence shows a person learned?
UnBlooms / A recursive response
Put the problem at the center. Make human judgment visible throughout.
Same loop.
Different worlds.
Good thinking depends on the discipline. Explore an illustrative activity, then adapt it to your learning goal.
Does AI deepen learning
or automate curiosity?
Select what matters. This guide supports a design decision; your context and institutional policy matter.
How deep is the critique?
The Critical AI Evaluation Scale offers five lenses—not a fixed ladder. Select one to explore evidence of thinking.
Your thinking.
Made visible.
Capture a real decision. Notes stay in this browser when storage is available. Download a copy to keep them.
Follow the checkpoints in any order. Each lights up as you make your reasoning visible.
Ready when you are.What would change
your mind?
Keep the loop open.