Briefs · research
Hypothesis Stress-Test
Attack a hypothesis to find where it would break before you test it.
You walk away with
A list of the assumptions, confounds and the experiment that would falsify it.
Decidi convenes
🌌 Stephen Hawking📈 The Statistician🔎 The Deep Researcher🔬 The Data Skeptic😈 The Devil's Advocate🛡️ The Steelman
Recommended level: Deep — The newest, most capable models — for when being wrong is expensive.
What the council debates
Stress-test this hypothesis before I invest in testing it. Try to break it. THE HYPOTHESIS: [state it precisely — what you predict and why] THE REASONING: [the logic and any prior evidence behind it] HOW YOU PLAN TO TEST IT: [the experiment or study design, if any] Debate: 1. Is the hypothesis precise and falsifiable, or too vague to test. 2. The assumptions it rests on, and which are shakiest. 3. The confounds and alternative explanations for any expected result. 4. What evidence would actually falsify it (not just confirm it). 5. The weakest link in the proposed test design. 6. The cheapest experiment that would most quickly support or kill it. FINAL SYNTHESIS: - The assumptions the hypothesis depends on, ranked by fragility. - The most likely alternative explanation to rule out. - The single experiment that would most efficiently test it, and the result that would falsify it.