What matters most
Core relationship: Record the observation first, then list explanations that could produce it. Confidence should rise when evidence is predicted by one explanation and unlikely under its rivals, not merely because more details were collected.
For inference under uncertainty without pretending certainty, begin with write the raw observation and finish with update proportionally. The sequence keeps the starting facts, intermediate result, and final limit visible instead of relying on memory.
A step-by-step approach
Write the raw observation
Avoid causal words inside the evidence statement. Capture the result before moving to list competing mechanisms. A written output shows whether the next deduction follows from the evidence or quietly adds a new assumption.
List competing mechanisms
Include measurement failure when the source could be wrong. Compare this output with the warning about one story becomes fact. If that failure still fits, revise the representation before using the result in another deduction.
Predict differing evidence
Ask what each explanation expects that another does not. Use predict differing evidence as a checkpoint. Name the input, the change made, and the conclusion that becomes available so another reader can reproduce the move.
Update proportionally
Strongly diagnostic evidence deserves a larger confidence change than compatible but common evidence. Finish by stating what this step establishes and what it leaves undecided. That boundary prevents a valid local conclusion from being stretched beyond the question.
Check against one story becomes fact
Questions for list competing mechanisms
- Can another reader reproduce write the raw observation from the stated information?
- Did you complete list competing mechanisms without reversing or enlarging the claim?
- Does the result survive the failure mode one story becomes fact?
- Which part remains uncertain after update proportionally?
This check is specific to the method above. A clear answer should expose the evidence chain and the point where a different fact would produce a different conclusion.
Example under pressure
Write the raw observation in the case
A sensor reports a sudden temperature drop. Possible explanations include a real environmental change, a moved sensor, power instability, or calibration drift. A second independent sensor separates these explanations better than another reading from the same device.
Why update proportionally matters
The case starts with write the raw observation, then uses list competing mechanisms to control the next inference. The conclusion is acceptable only after update proportionally. Changing any fact that supports those moves can change the answer, while an unrelated detail should not.
Common wrong turns
One story becomes fact
Plausibility is only the start of comparison. Return to write the raw observation and rebuild only from information the problem actually provides.
More of the same data
Repeated dependent readings may add little information. Compare the disputed move with predict differing evidence; the two should use the same definitions and direction.
False precision
A numerical confidence without a defensible basis can conceal uncertainty. Run the practice task while deliberately watching for false precision. Record the first point where the reasoning begins to drift.
Practice and review
Take one ambiguous event and write three explanations. For each, name one observation that would raise confidence and one that would lower it.
Continue after update proportionally
Use the probability reasoning guide for the broader method, then apply the idea in the probability calculator. Keep the written reasoning so the next attempt can improve a specific step rather than simply repeat the exercise.
Compare this method with Cognitive Bias in Puzzle Solving, then continue to Matrix Reasoning: Compare Row and Column Rules for a different reasoning task.