Lab
Search Space Contraction
Complex identification problems resolve through sequential observations that divide the remaining possibility space — efficiency depends on choosing tests that maximize eliminated candidates per effort expended.
Then check the pattern
Why does a single coarse observation often narrow a search more than multiple precise measurements?
Coarse observations are less prone to measurement error and deliver more reliable results A binary partition removes half the possibility space; precision only distinguishes among what survives that cut Precise measurements require specialized training while coarse observations can be done by anyone Initial observations establish context that makes later precision meaningful
Answer: A binary partition removes half the possibility space; precision only distinguishes among what survives that cut. Dividing the remaining field in half eliminates 50% of candidates in one step. Measuring a detailed property only differentiates among survivors of earlier cuts — you must first reduce the space to where that precision matters. Option D sounds compelling but confuses sequencing strategy with elimination efficiency.
When does further investigation stop adding value?
When remaining candidates produce identical consequences for the decision being made When the research budget allocated to the question has been exhausted When peer reviewers agree the conclusion has been adequately supported When the lead investigator declares sufficient confidence in the result
Answer: When remaining candidates produce identical consequences for the decision being made. If all remaining options lead to the same action — same treatment protocol, same risk assessment, same operational response — additional narrowing becomes cosmetic. Investigation serves decisions, not abstract completeness. Budget constraints (B) are real but external to the logic of when narrowing matters.
What makes an observation worthless for narrowing?
The measurement tool's precision falls below the threshold needed to detect differences The feature being measured appears identically in all candidates still under consideration The observation contradicts results from earlier, more reliable tests The observed value falls outside the range predicted by any remaining theory
Answer: The feature being measured appears identically in all candidates still under consideration. An observation that finds the same property in every remaining option eliminates zero candidates — it adds information but changes nothing about what might be true. Measurement imprecision (A) can blur boundaries, but a feature present in all survivors is useless even if measured with infinite precision.
What governs which test gets performed next?
Tests proceed from least invasive to most invasive to preserve sample integrity Standard protocols dictate measurement order to ensure reproducibility across studies The ratio of candidates eliminated to observation cost determines priority Available equipment constrains which measurements are technically feasible at that moment
Answer: The ratio of candidates eliminated to observation cost determines priority. A cheap test that halves the field gets performed before an expensive test that distinguishes only two finalists. You maximize eliminated options per unit of effort — whether that effort is time, money, or sample consumption. Option D describes a constraint, not a selection principle.
Why do some possibility spaces never contract to a single answer?
Measurement tools lack the resolution to distinguish candidates below a certain similarity threshold The cost of the next test exceeds the benefit of knowing which finalist is correct Natural variation within the correct category exceeds the difference between candidate categories Remaining candidates occupy positions in conceptual space where no defining test exists yet
Answer: Measurement tools lack the resolution to distinguish candidates below a certain similarity threshold. Tools impose a resolution floor — if candidates differ only in ways no available test can detect, contraction stops at 'one of these few'. This is a harder limit than cost-benefit tradeoffs (B) or natural variation (C), both of which allow decision rules like 'close enough'. Option D describes knowledge gaps, not contraction mechanics.
← All labs