MCAT content library / Psychology & sociology / Problem solving, heuristics, and belief revision
A useful shortcut becomes a problem when the task changes or relevant information is left out. Identify the missing comparison.
About 17 minutes, including the worked examples and checks.
On this page
- Start with the problem’s structure
- Choose a procedure that fits the constraints
- Identify what the shortcut substitutes
- Worked probability: a matching feature is not the whole denominator
- Change the wording without changing the outcome
- Separate searching for support from keeping a discredited belief
- Compare confidence with the appropriate benchmark
- Check your reasoning
- Continue learning
- Sources
Start with the problem’s structure
A familiar solution can be efficient until the constraints change. A vivid example can guide a judgment until its frequency is mistaken for its memorability. AAMC groups problem-solving approaches, barriers, heuristics, overconfidence and belief perseverance within cognition.1 The task is to connect a proposed shortcut to the particular information it uses or neglects.
A useful explanation identifies the goal, the permitted operations and the stopping rule. “The participant is biased” supplies none of these. Even an incorrect answer may reflect misunderstood instructions, missing information or arithmetic error rather than the cognitive process named in an option.
Choose a procedure that fits the constraints
An algorithm specifies a procedure whose result follows under its stated conditions. A heuristic reduces the search or calculation using a shortcut. Trial and error tests possibilities, while working backward or setting intermediate goals can make a large problem manageable. A heuristic may be useful without guaranteeing the best solution.2
All cases and numbers below are original teaching examples. A small delivery robot follows a rule: move to the neighboring location nearest the destination. This works in an open grid. After a barrier is added, the route may first need to move away from the destination. Insisting on the old rule prevents exploration of that route. A complete search of a finite map, with already explored states recorded, can find an available route under the map’s stated movement rules.
A mental set is persistence with a familiar approach when a different representation or strategy is needed. Functional fixedness concerns treating an object as limited to its usual use.2 In a workshop puzzle, a rigid shipping tube might be usable as a roller. Presenting it as packaging could make that possibility less salient than presenting the same tube among loose construction materials. To test the explanation, vary presentation while preserving its physical usefulness.
Identify what the shortcut substitutes
Representativeness substitutes resemblance for a probability judgment. Availability uses how readily examples come to mind as a guide to frequency or likelihood. Anchoring gives an initial value disproportionate influence on a later estimate. These are distinguishable accounts, not interchangeable names for every error.3
Suppose a person overestimates equipment failures after hearing one memorable account. An availability explanation predicts a role for recall accessibility. A numerical anchoring explanation instead needs an initial value that pulls the estimate. Varying story vividness and an unrelated starting estimate separately would help distinguish them. A single high estimate cannot do that work alone.
Worked probability: a matching feature is not the whole denominator
A depot receives 1,000 parcels: 100 priority and 900 regular. A red stripe appears on 75% of priority parcels and 20% of regular parcels. The question is the chance that a randomly selected red-striped parcel is priority.
| Category | All parcels | Fraction with stripe | Striped parcels |
|---|---|---|---|
| Priority | 100 | 75% | 75 |
| Regular | 900 | 20% | 180 |
| Total | 1000 | 255/1000 | 255 |
The stripe is more common within the priority category: 75% versus 20%. Nevertheless, regular parcels are much more numerous. They contribute 900 × 0.20 = 180 striped parcels, compared with 100 × 0.75 = 75 priority parcels. Among the 255 striped parcels, the priority fraction is 75/255 ≈ 29.4%.
Answering “75%” reverses the condition. That number is the chance of a stripe given priority status, not priority status given a stripe. Answering “it looks like the priority pattern, so it is probably priority” also omits the category frequencies. The stripe is informative, yet most striped parcels in this depot are regular.
To test base-rate use experimentally, researchers could vary the proportions of priority and regular parcels while holding the two stripe probabilities fixed. A model that uses all supplied information predicts different answers. A response that remains fixed at 75% would be inconsistent with that model, although further checks would still be needed to distinguish a heuristic from confusion about the question.
Change the wording without changing the outcome
Framing concerns differences associated with how equivalent outcomes are described.4 Consider two descriptions of the same shipment policy: “90 of 100 packages arrive intact” and “10 of 100 packages are damaged.” If intact and damaged are the only outcomes, both statements describe the same distribution.
A controlled framing comparison holds the price, timing and underlying outcomes fixed and randomly varies the description. A choice difference would then concern the presentation, rather than a numerical advantage hidden in one option. Do not label two choices equivalent if one also changes the delivery time or the consequences of damage. That would change the decision itself.
Separate searching for support from keeping a discredited belief
Confirmation bias concerns favoring information that supports an existing belief.2 Belief perseverance concerns persistence after the original basis of a belief has been discredited.5 They can occur together, but they describe different points in an evidence process.
Imagine a fictional production line rated highly because of a performance chart. Investigators later establish that the chart was a randomly generated demonstration and contains no information about that line. Continuing to rely on that chart-derived rating, with no independent supporting evidence, illustrates perseverance. Seeking only favorable new reports illustrates selective information search. A rigorous comparison asks what evidence remained valid after the correction, rather than requiring every belief to reverse whenever anyone disagrees.
Compare confidence with the appropriate benchmark
Overconfidence can mean overestimating actual performance, placing oneself too highly relative to others, or expressing excessive precision. These measures need not move together.6 In a calibration task, suppose 100 predictions each receive 80% confidence and 60 turn out correct. Confidence exceeds observed accuracy by 20 percentage points for this set.
That result does not show where the person ranks among peers or whether the same pattern occurs on another task. Likewise, lower confidence after feedback does not prove improved calibration unless accuracy is also considered. Carry the measured quantity and its comparison into the conclusion: strategy versus available routes, conditional probability versus category counts, belief versus valid evidence, and confidence versus observed outcomes.
ACTIVE PRACTICE · CONDITIONAL PROBABILITY
Count the Whole Feature-Positive Group
The original depot contains 1,000 parcels, including 100 priority and 900 regular parcels. A red stripe appears on 75% of priority parcels and 20% of regular parcels. A random parcel is observed with a stripe.
Priority striped: of. Regular striped: of.
Given a stripe, are priority. Teal = priority among striped parcels; gold = regular among striped parcels.
The stripe probabilities are held fixed while prevalence changes. These are stipulated counts, not population estimates with sampling uncertainty.
Check your reasoning
Choose an answer before opening its feedback. Each option explains the reasoning behind that choice.
Check 1
A puzzle has a finite set of legal configurations. A familiar rule that always moves a piece closer to its goal enters a loop, although a solution exists that temporarily moves a piece farther away. Which procedure guarantees finding a solution under these conditions if allowed to finish?
A. Explore all reachable legal configurations systematically while recording which have already been explored.
Correct. A complete search of a finite reachable state space can examine the solution path without endlessly revisiting the loop. The guarantee depends on the stated finite map and legal-move rules.
B. Keep only moves that reduce the current distance to the goal, but alternate between equally near choices.
Review this choice. The solution requires a temporary increase in distance. The restricted move rule still excludes it.
C. Repeat the familiar sequence until the same configuration feels easier to recognize.
Review this choice. Recognition does not change the legal configurations or break the looping strategy.
D. Select the move used most often in earlier puzzles whenever the familiar rule gives two choices.
Review this choice. Past frequency is a heuristic. It does not guarantee exploring the route that the new constraint requires.
The principle to carry forward
A guarantee comes from the procedure and its assumptions, not from how familiar or intuitive a move seems.
Check 2
A collection contains 400 common components and 100 rare components. A marking appears on 80 common and 60 rare components. A component is selected uniformly from those with the marking. What is the probability that it is rare?
A. 20%
Review this choice. 100/500 is the prior fraction rare in the full collection. The selection is restricted to marked components.
B. Approximately 42.9%
Correct. The marked set contains 80 + 60 = 140 components. Of these, 60 are rare, so 60/140 ≈ 0.429.
C. 60%
Review this choice. 60/100 is the fraction marked among rare components. The question asks the reverse conditional proportion.
D. 75%
Review this choice. 60/80 compares rare marked components with common marked components. The denominator must include both kinds of marked component.
The principle to carry forward
Build the set specified by the condition, then count the requested category within it.
Check 3
Readers form a favorable judgment of a fictional workshop solely from a quality chart. They then receive and correctly understand proof that the chart was randomly generated and provides no evidence about the workshop. No independent quality information is supplied. They retain the chart-based judgment. Which account is most directly supported?
A. A rational update based on new independent evidence supporting workshop quality.
Review this choice. The stem supplies no independent supporting evidence. The original chart has lost its informational basis.
B. A selective search for additional favorable reports about the workshop.
Review this choice. No further search or new reports are described. That could support confirmation bias, but it is not the observed process here.
C. Persistence of the initial belief after the sole basis for it was discredited.
Correct. The crucial sequence is formation from one source, removal of that source’s evidential value, and retention of the same source-based judgment.
D. Proof of excessive confidence relative to the readers’ measured overall accuracy.
Review this choice. The stem provides neither confidence ratings nor an accuracy benchmark across judgments. It does not establish that calibration result.
The principle to carry forward
Match the inference to the evidence process actually observed; persistence and selective search are related but distinct.
Continue learning
Related lessons
- Attention under competing task demands
- Hypotheses, predictions, and competing explanations
- Psychological Measurement: What Does the Score Mean?
Browse psychology & sociology · MCAT content library
MCAT preparation and roadmap · Aim High, Score Higher
Sources
1. AAMC: Content Category 6B — Making sense of the environment
2. OpenStax Psychology 2e: Problem Solving
3. Tversky and Kahneman (1974): Judgment under Uncertainty: Heuristics and Biases
4. Tversky and Kahneman (1981): The framing of decisions and the psychology of choice
5. Ross, Lepper and Hubbard (1975): Perseverance in self-perception and social perception