1 Psychology and human development1.6 Thought and higher cortical functions
1.6 Thought
2 min read1 figure
Written by Resident Doctor Committee · Published
Concepts and reasoning
Prototype theory (Rosch) holds that a concept is organised around a typical member, so robins are recognised as birds faster than penguins.
Deduction derives conclusions that must follow from true premises, while induction draws probable general conclusions from observations that stay open to revision.
Algorithms and heuristics
An algorithm guarantees a solution when its steps are followed, as in long division, but it may need impractical time or information.
A heuristic is a rule of thumb that saves effort, and Tversky and Kahneman described the predictable biases these shortcuts produce.
Satisficing (Simon) is choosing the first acceptable option when time, information and capacity are limited, a state called bounded rationality.
Heuristics
Heuristic | Rule used | Error it produces |
|---|---|---|
Availability | Ease of recall | A vivid or recently met case makes a diagnosis seem more likely or common than it is |
Representativeness | Resemblance to a prototype of the category | A patient who does not match the stereotype is discounted, and base rates are overlooked |
Anchoring and adjustment | Starting from an initial value or first diagnosis | Too little revision as contrary evidence arrives |
Availability turns on how easily cases come to mind, whereas representativeness turns on how closely this case resembles the typical one.
Framing effect
The framing effect is when logically identical information produces different choices depending on its wording.
The frame may set a gain against a loss, or a success rate against a failure rate.
Only the wording changes: the facts and probabilities stay the same.
Other biases and fallacies
Bias | What happens |
|---|---|
Base rate neglect | A positive test is read as near-certain disease, ignoring how common the condition is in the group tested (low pre-test probability) |
Confirmation bias | A clinician favouring a diagnosis seeks, notices and weights supporting evidence, and discounts or reinterprets contrary evidence |
Diagnostic momentum | An early label is carried forward by later clinicians and grows more certain through repetition, not new evidence |
Conjunction fallacy | A combination of events is judged more probable than one of its components |
Gambler's fallacy | People believe an independent random event's probability changes with preceding outcomes, so a long-absent outcome seems 'due' |
Anchoring is one clinician revising too little, whereas diagnostic momentum is a label passed between clinicians and hardened by repetition.
Confirmation bias actively selects and weights evidence, whereas anchoring is a failure to move far enough from a starting point.
Clinicians counter representativeness and base rate neglect by stating the population frequency, considering alternatives and seeking evidence that would disprove the leading diagnosis.
Cognitive models
Overgeneralisation is a cognitive distortion in Beck's cognitive model in which one isolated setback is turned into a broad rule and applied to unrelated situations.
The domino effect in depression is a self-feeding spread of negative appraisal from one area of life to more and more areas, which feeds rumination.
Naive, folk or commonsense psychology is the everyday, non-scientific way people explain and predict behaviour by ascribing beliefs, desires and intentions.