CogniqOS
Subscribe

TRAP #010 ·VOL I The Individual · Judgment & Evaluation Errors

Bounded Rationality

Discovered by Herbert A. Simon, 1947

Our brains don't optimise - they satisfice and call it a day. (Evidence: Strong) [Foundation concept]
Category: BiasEvidence: ReplicatedUniversality: HighTier 1Type: Mechanismⓘ what these mean

This is a Lexicon entry: the mechanism, when it shows up, and the countermeasure. The full correction, with real-world cases and the audit prompt, hasn't been written yet.

Definition

The idea that human decision-making is limited by available information, time and cognitive capacity, leading us to settle for "good enough" rather than optimal solutions.

When it shows up

Any real decision under time, information or cognitive-capacity limits: choosing a restaurant, hiring someone after four interviews, picking a medical treatment with a 20-minute consultation. Anywhere optimisation is impossible. Modern context: A product team given 200 customer feedback data points makes a worse feature prioritisation decision than a team given the top 10 - the larger dataset activates search strategies that find patterns in noise. High-stakes instance: Policy design - regulations written to be technically correct but cognitively complex produce systematically lower compliance than equivalent rules written to match the satisficing heuristics of the regulated population. Failure mode: The mind selects the first option that crosses an acceptability threshold (satisficing) rather than the best option. The decision is 'good enough' by definition because that is the stopping rule but systematically leaves value on the table. Estimated cost: High - often irreversible where this bias dominates unchecked decisions.

Failure mode

The mind selects the first option that crosses an acceptability threshold (satisficing) rather than the best option. The decision is 'good enough' by definition because that is the stopping rule but systematically leaves value on the table.

Countermeasure

Accept satisficing for low-stakes choices; it is rational. For high-stakes choices, explicitly extend the search: set a minimum number of options to evaluate before stopping or a minimum time before deciding. When correction costs more than the bias: Almost always: satisficing is often the correct strategy, not a flawed one. The cost of searching for the optimal solution in large solution spaces exceeds the benefit for most decisions. Bounded rationality is not a failure mode - it is an efficient adaptation. Correct only when the decision is high-stakes, irreversible and the cost of sub-optimality is large.

Related traps

Also connected in the map4 more, locked

Credit & first seen

DiscoveredSimon, H.A. (1947). Administrative Behaviour. Macmillan. Satisficing formalised in Simon, H.A. (1955).A behavioural model of rational choice.Quarterly Journal of Economics 69(1), 99-118. See also Simon, H.A. (1956).'Rational choice and the structure of the environment.'Psychological Review 63(2), 129-138. source ↗

Also revealed

Want the full correction for Bounded Rationality?

This entry has the mechanism and countermeasure. The full write-up - real cases, sources, the audit prompt - comes later.

Entry #10 of 631 in The Lexicon · see the full Lexicon