Questions: Reasoning Biases and Systematic Errors in Logic

Question 1

A physician suspects a patient has a rare autoimmune condition and orders a battery of tests that could confirm it. Several tests come back positive, reinforcing her suspicion. She doesn't order tests that could rule out more common conditions with similar symptoms. This behavior best illustrates:

• A. Base-rate neglect — she is ignoring how rare the autoimmune condition is
• B. Belief bias — she is judging the validity of her diagnosis based on its plausibility
• C. Confirmation bias — she is seeking confirming evidence while ignoring disconfirming evidence
• D. The representativeness heuristic — she is matching the patient's symptoms to her prototype of the disease

Answer: (C)

Confirmation bias involves preferentially seeking, attending to, and interpreting information that confirms an existing belief, while neglecting disconfirming evidence. The physician is not passively exposed to confirming information — she is actively selecting it. While base-rate neglect and representativeness may also play a role in the clinical scenario, the described behavior (selecting confirming tests, not ordering disconfirming tests) is the defining signature of confirmation bias.

Question 2

In Wason's selection task, participants are shown four cards and asked which cards to flip to test the rule 'If a card has a vowel on one side, it has an even number on the other side.' Most people choose the vowel card and the even-number card. What does this pattern reveal about human reasoning?

• A. People correctly apply deductive logic to find the most informative cards
• B. People seek confirming evidence (the vowel card) and ignore the logically required disconfirming evidence (the odd-number card)
• C. People use the representativeness heuristic to match cards to the rule's structure
• D. People exhibit belief bias by favoring conclusions that seem mathematically plausible

Answer: (B)

The logically correct cards are the vowel (could falsify if it has an odd number) and the odd-number card (would falsify the rule if it has a vowel on the other side). The even-number card cannot falsify the rule — finding a vowel there merely confirms it, but finding a consonant tells you nothing. Most people choose the confirming even-number card instead of the falsifying odd-number card, demonstrating confirmation bias: the tendency to seek evidence that could verify rather than falsify the hypothesis.

Question 3

A student learns about confirmation bias in a psychology course. This knowledge alone is sufficient to prevent confirmation bias from affecting the student's future reasoning.

• T. True
• F. False

Answer: False

Knowing about a bias does not automatically suppress it, because reasoning biases reflect fast, automatic System 1 processes that operate before reflective System 2 scrutiny is applied. The initial biased judgment is formed before conscious awareness intervenes. Effective debiasing requires changing the decision environment — making disconfirming evidence salient, requiring explicit consideration of alternatives — not just intellectual knowledge of the bias's existence.

Question 4

Belief bias causes people to judge a logically valid argument as invalid when its conclusion contradicts their prior beliefs, even if the premises logically entail the conclusion.

• T. True
• F. False

Answer: True

This is the defining feature of belief bias. In studies presenting syllogisms with unbelievable but logically valid conclusions (e.g., 'All mammals can walk; whales are mammals; therefore whales can walk'), participants frequently judge the argument as invalid because the conclusion conflicts with world knowledge. They are substituting a fast semantic judgment about the believability of the conclusion for the slower logical evaluation of whether the conclusion follows from the premises.

Question 5

Why are reasoning biases described as 'systematic' rather than 'random,' and what does this distinction reveal about their origin?

Model answer: Systematic means the errors occur predictably, in specific directions, across different people and contexts — not randomly or idiosyncratically. The direction of error is informative: confirmation bias always favors confirming evidence, the representativeness heuristic always overweights surface resemblance, belief bias always favors believable conclusions. This predictability reveals that the biases reflect the structure of fast associative cognitive processes that evolved for efficient judgment under uncertainty. They are not failures of intelligence but adaptations that work well in familiar domains yet produce characteristic errors when applied to formal logic, probability, and statistics.

The distinction matters because if biases were random, they would cancel out and be less practically significant. Their systematic nature means they produce consistent, predictable distortions — in medical diagnosis, legal judgment, financial decisions — that cannot be corrected by averaging over many observations or hoping noise cancels. Understanding the direction and source of the bias is what enables targeted debiasing strategies.