← All mental models

Reusable cognitive primitives

Optimization

Searching a space of options for the best one under constraints — following a gradient of 'better' toward a maximum or minimum.

Where it appears

Dependencies & synergies

Derived from the graph’s real structure — 12 concepts across 17 disciplines carry this pattern. Every figure is a count, not a score.

Reach across disciplines

How many concepts carrying this pattern are seen through each discipline.

Patterns that travel with it

Other thinking patterns that recur on the same concepts — the more shared concepts, the more often they co-occur.

The shape of its reasoning

What kinds of relationships the carrying concepts form — the relational signature of the pattern.

  • Teaching link23 edges
  • Dependency13 edges
  • Kind & structure8 edges
  • Cause & effect8 edges
  • Analogy & transfer5 edges
  • Evidence1 edges
  • Explains & models1 edges
Furthest-reaching carriers

Concepts where this pattern does the most cross-disciplinary work — each reaches disciplines beyond its own.

  • OptimizationreachesAlgebraAlgorithmsArtificial IntelligenceBiologyComputer ScienceData ScienceDecision TheoryDesignEnvironmental EngineeringEthicsIndustrial EngineeringMachine LearningMathematicsOperations ResearchPhysicsPolitical ScienceProject ManagementPublic PolicyStatisticsSustainability ScienceSystems Science
  • Natural selectionreachesAnthropologyArtificial IntelligenceAstrobiologyBioinformaticsBiotechnologyCell BiologyComparative LiteratureComputer ScienceDemographyEarth & Space SciencesEcologyGeneticsGeologyMachine LearningMedicineMicrobiologyMolecular BiologyPaleontologyPublic HealthSociologyStatistics
  • Cost–benefit analysisreachesBiologyBusinessDesignEngineeringEthicsPolitical SciencePublic AdministrationPublic PolicySystems Engineering
  • Gradient descentreachesBusinessEconomicsEngineeringMathematical ModellingStatisticsSystems Engineering
  • Dynamic programmingreachesDiscrete MathematicsLogicMathematicsSoftware EngineeringTheory of Computation
  • Greedy algorithmreachesDiscrete MathematicsLogicMathematicsSoftware EngineeringTheory of Computation

Foundation of the pattern

Carrying concepts that others in the same pattern build on (they depend on or follow from these).

  1. Fitness 1 other concept build on it

Examples across disciplines

Machine Learning

Gradient descent nudges parameters downhill to minimise error.

Biology

Natural selection 'searches' for traits that raise reproductive fitness.

Economics

A firm choosing output to maximise profit under a budget.

Physics

Light takes the path of least time; systems settle at least energy.

How this pattern travels

Interpreted from where the pattern recurs in the atlas — structural transfer and coverage, not a claim it is universally the "best" model.

  • Recurs across 17 disciplines over 12 carrier concepts — a broadly transferable pattern.

    structural · Structural recurrence in the atlas — a pattern is a reasoning lens, not a law.

  • 3 of its 12 carrier concepts are themselves cross-disciplinary connectors.

    structural · Structural recurrence in the atlas — a pattern is a reasoning lens, not a law.

  • Explicit "where it breaks" notes exist for 8 of 12 annotated assignments.

    curated · Curated boundary annotations — absence is a representation gap, not evidence the model has no limits.

  • Currently dark in the atlas: no origin recorded.

    atlas representation · Describes the current Thinking OS representation, not the model itself.

Coverage matrix

How these concepts distribute across domains and concept families — real counts, not a score.

SystemsInformationMatterChangePatternsEnergyLifeScaleStructureDecisionNetworksWavesCausalityComputationEarthSpaceOptimizationNumberProbabilitySecurityThresholdsConstraintsNatural sciencesFormal sciencesEngineeringMedicine & healthSocial sciencesHumanitiesProfessionalArtsInterdisciplinary211341

Knowledge timeline

Real, sourced key dates of these concepts.

3 events
18201830184018501817 · Comparative advantage1858 · Natural selection1859 · Natural selection

The statistical fingerprint

How the 12 concepts that exercise this pattern distribute — from the graph, not a ranking.

Disciplinary fingerprint

Carrier concepts under each illuminating lens.

How settled its carriers are

Epistemic status of the concepts that exercise this pattern.

  • Established12 · 100%

Where the model breaks

This pattern is a reasoning lens, not a law. Here is where it stops helping:

  • Adaptation: Evolution "optimises" locally and without foresight — it climbs the nearest peak, not the global one, and the fitness landscape itself keeps shifting.
  • Comparative advantage (reasoning tool): The optimisation assumes stable, comparable preferences, mobile resources and full information — it ignores adjustment cost, distribution and who bears the loss.
  • Cost–benefit analysis: Optimising net benefit assumes everything worth counting can be priced on one scale — it silently down-weights the incommensurable and the irreversible.
  • Equal temperament (reasoning tool): The optimization sacrifices the pure whole-number ratios of just intonation — only the octave stays acoustically pure.
  • Fitness (reasoning tool): Fitness is defined post hoc by realised reproductive success, not a pre-set target the organism aims at.
  • Gradient descent (mechanism): Finds a local optimum along the given objective; a wrong loss, bad conditioning or a non-convex surface leaves it stuck far from the best solution.
  • Greedy algorithm: Greedy is optimal only for problems with optimal substructure / matroid structure; elsewhere it is a heuristic that can be arbitrarily far from best.
  • Natural selection (analogy): Evolution has no objective function and no optimizer — 'maximises fitness' is a descriptive analogy. Selection is myopic, can be trapped on local peaks, and the fitness landscape itself shifts with the environment.

3 of 12 explained assignments cite a source; the rest are editorial interpretations. 1 are flagged for review. None is externally validated.

Common misconception

The best local option is the best overall. Greedy, downhill-only search can get stuck in a local optimum.

Try a transfer challenge

You have seen this model in one place. Where else could it apply — and where would the analogy break?

Comparative advantage
The intuition

Ask: what is being maximised or minimised, and what limits the search? Beware settling in a 'local' best that isn't the global one.

The reach of this pattern
  • Recurs across 17 disciplines.
  • 34 concepts exercise this thinking pattern.
  • 8 of them explicitly note where the model breaks down.

Derived from the graph — a pattern is a reasoning lens, not a law.

How to recognise it

What quantity is being driven to a maximum or minimum here, under what constraint — and could the search be stuck in a 'local best' that isn't the true one?

Keep this question handy when you meet something new — it helps you notice the pattern, not just name it.