Reasoning

What is the difference between deductive and inductive reasoning in terms of structure and strength of conclusions?

Deductive reasoning moves from general principles to specific conclusions. If the premises are true and the argument is valid, the conclusion is necessarily true (logically certain). Inductive reasoning moves from specific observations to general conclusions. Conclusions are probabilistic — well-supported but never guaranteed. Science is fundamentally inductive: we observe patterns and draw general conclusions, but new evidence can always overturn them.

What is syllogistic reasoning? What errors do participants make, and how could they be avoided?

A syllogism has two premises and a conclusion (e.g. All ants are insects / All insects are animals / All ants are animals). Participants judge whether the conclusion follows logically. Errors arise from: (1) belief bias — accepting logically invalid conclusions that sound believable, or rejecting valid ones that sound odd; (2) misunderstanding logical terms like 'some' (logical vs everyday use); (3) phrasing of premises. Errors can be avoided by applying the formal rules of logic — treating premises as abstract symbols rather than meaningful content, and asking only whether the conclusion follows from the premises regardless of real-world truth.

What is conditional reasoning? Explain modus ponens and modus tollens and why one is harder.

Conditional reasoning tests whether evidence supports or refutes an 'if P then Q' statement. Modus ponens: If P then Q / P is true / Therefore Q is true — most people find this straightforward and valid. Modus tollens: If P then Q / Q is false / Therefore P is false — this is logically valid but much harder. People struggle because it requires reasoning backwards from a negated consequent, which is less intuitive. Many people incorrectly affirm the consequent (if Q, then P) instead.

What is the Wason Card Task and how does context affect performance?

Participants see four cards (e.g. E, K, 4, 7) and must choose which to turn over to test the rule 'If a card has a vowel on one side, it has an even number on the other.' Most people make the confirmation error — they turn over the vowel card and the even number card, instead of the vowel and the odd number card (the falsifying cards). Performance improves dramatically when the rule is given a real-world social context (e.g. checking if underage drinkers have alcohol), because people can reason naturally about permission and cheating (Cosmides & Tooby) — System 2 is engaged more effectively.

How does the scientific method use inductive reasoning, and what does Wason's Number Sequence Task reveal about hypothesis testing?

Inductive reasoning is used when moving from observations to a general hypothesis or theory — the point at which specific data are used to draw broader conclusions. In Wason's task, participants see '2, 4, 6' and try to find the rule. Most generate confirming sequences (even numbers, +2), demonstrating confirmation bias. The most productive strategy is falsification — generating sequences that would disprove your hypothesis. This is more efficient because a single disconfirming result eliminates a hypothesis, while confirmations never prove a rule definitively.

Describe Fugelsang et al.'s experiment on inductive causal reasoning and summarise the results pattern.

Participants read vignettes as horticulturists trying to identify what causes flowers to bloom. Degree of belief (plausibility of the cause: sunlight = high belief; red pots = low belief) and degree of covariation (how often cause and effect co-occurred) were manipulated. Participants rated causal power. Under low covariation, high-belief vignettes received far higher ratings — prior belief dominated. Under high covariation, the gap between high and low belief narrowed — strong evidence reduced the influence of prior belief. So belief matters most when covariation evidence is weak, and covariation can override belief when it is high.

What is the dual-process model? Name, describe, and contrast the two systems.

Proposed by Kahneman & Frederick. System 1: fast, automatic, implicit, associative, high capacity, emotionally charged, independent of cognitive ability — not necessarily rational; the source of biases and heuristics. System 2: slow, effortful, explicit, rule-governed, flexible, low capacity — rational, shows individual differences. System 1 is used first and more often; System 2 evaluates and can override System 1. Together they explain bounded rationality — we can reason logically, but not perfectly or always. The dual-process model best describes how we actually reason in real life.