Rational Choice Model¶
Model an actor's choice as the maximal feasible option under an attributed evaluation, to explain or predict behavior.
Core Idea¶
A rational choice model explains or predicts an actor's behavior as if the actor selects a maximal feasible alternative under an attributed evaluation. The analyst must identify the actor, the available options, the actor's modeled ordering of them, and the rule that turns that ordering into a predicted act. This is an explanatory model, not proof that the person consciously calculates a number or that the prediction is accurate.[ref-83bf50809d8b][ref-67c7263ecca4]
“Maximal” may identify a set of tied choices rather than one action. A particular model must say enough about its ordering and feasible menu for a prediction to exist. Cyclic rankings can lack a maximal alternative even on a finite menu; an open-ended menu can lack an attained best option. Complete and transitive preferences are one common modeling specification rather than an observed fact about every actor.[^ref-faa6e6695f8d]
Scope of Application¶
Lancaster models a consumer choosing goods whose activities produce valued characteristics under a budget. Becker models a potential offender weighing uncertain consequences of an offense against other uses of time and resources. The shared structure is a situated actor, feasible alternatives, attributed evaluation and behavioral choice rule. Lancaster's deterministic case does not require risk-weighted expected utility; Becker's uncertain case uses it.[ref-83bf50809d8b][ref-67c7263ecca4]
Riker and Ordeshook apply a rational calculus to voting or abstention. Only the original publisher extract and an accessible note were inspected here, so this entry does not infer their full voting equation or an empirical parameter estimate.[^ref-26f0faae2618]
Clarity¶
Separate preference, choice rule, and behavioral explanation. Ranking two goods says what a consumer prefers; a budget and maximal-choice rule specify which feasible bundle the model predicts; checking a purchase against that prediction tests whether the model describes behavior. A mismatch can arise from the menu, attributed evaluation, rule, or the model's fit, rather than establishing that a person violated a universal rationality law.[^ref-83bf50809d8b]
The model does not morally endorse an actor's goal. Becker's economic explanation of offending does not approve the offense. It also need not assert that a person literally wrote down utility values or solved an optimization program.[^ref-67c7263ecca4]
Manages Complexity¶
A choice setting may involve many prices, sanctions, probabilities and nonmaterial goals. Five questions organize it: who chooses, what can they choose, how are the options compared, what counts as a maximal choice, and what behavior follows? Lancaster's goods-to-characteristics map and Becker's probability of conviction fill in different case-specific details.[ref-83bf50809d8b][ref-67c7263ecca4]
This shortcut has a limit. If an analyst changes the attributed preferences after every observed choice, almost any act can be rationalized. For a testable prediction, declare the menu, evaluation and tie or existence convention before using the act as evidence. That is a modeling diagnostic inferred from the structure, not a quantified result of these studies.
Abstract Reasoning¶
Suppose a price change is followed by a different purchase. Update the feasible menu and ask whether the model predicts a different maximal characteristics bundle. If it does not, inspect the assumed characteristics map and preference ordering before drawing a conclusion about the buyer. For an offense decision, ask whether a change in conviction probability or punishment alters the modeled comparison with other uses of resources; no universal numerical deterrence effect follows from the setup alone.[ref-83bf50809d8b][ref-67c7263ecca4]
The live Preference Prime is an internal component: the actor-relative ordering is necessary for the maximal-choice rule. The model adds a situated actor, feasibility and a behavioral prediction. Its typed edge records Rational Choice Model → Preference as strict part_of composition, with the parent inside the child. An ordering can exist without predicting someone's conduct.[ref-83bf50809d8b][ref-67c7263ecca4]
Knowledge Transfer¶
Carry the role test from consumer demand to crime or bounded political participation, while rebuilding the actor, menu, evaluation and choice rule for each domain. Do not copy Lancaster's product-characteristics transformation into Becker's offense model or assume that every case needs Becker's risk probabilities. Optimization may implement a numerical case, but an ordinal ordering need not supply a universal numerical objective. The named rational choice model remains a specialist account of actors' behavior on this evidence.[ref-83bf50809d8b][ref-67c7263ecca4][^ref-26f0faae2618]
Example¶
Consumer characteristics: In Lancaster's account, a consumer selects a feasible bundle of goods or activities. Those activities produce characteristics, over which the model assigns utility. Budget and transformation relations constrain the menu; the model predicts a maximizing bundle under its own assumptions. The actor is the consumer, the evaluation concerns characteristics, and the output is a consumer-demand prediction. No uncertain lottery or population aggregate is needed for this instance.[^ref-83bf50809d8b]
Potential offense: In Becker's account, a person compares the expected utility of offending with other uses of time and resources. Conviction probability and punishment affect this case's uncertain ranking. The actor is the potential offender, feasible alternatives include offending and not offending, and the maximal-choice rule predicts behavior from the modeled comparison. Later aggregate offense totals are a further step, not part of the individual's choice by definition.[^ref-67c7263ecca4]
Relationships to Other Abstractions¶
Current abstraction Rational Choice Model Domain-specific
Parents (1) — more general patterns this builds on
-
Rational Choice Model is part of Preference Prime
An actor's evaluator-relative ordering is an identity-bearing part of a rational choice model.
Hierarchy path (1) — routes to 1 parentless root
- Rational Choice Model → Preference
Neighborhood in Abstraction Space¶
Rational Choice Model sits in a sparse region of the domain-specific corpus (95th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Unclustered & Miscellaneous (2551 abstractions)
Nearest neighbors
- Generalized Semi-Infinite Programming — 0.79
- Interval Contractor — 0.78
- Feasible Region — 0.78
- Nonelementary Problem — 0.77
- Goal Modeling — 0.77
Computed from structural-signature embeddings · 2026-10-08
Not to Be Confused With¶
Preference orders options but does not itself predict a situated actor's conduct. Optimization finds a best option under constraints but need not explain human behavior. Expected Utility supplies a probability-weighted ranking for some uncertain cases, not Lancaster's deterministic consumer instance. Normative Rationality judges reasons and responsiveness to evidence; a successful as-if prediction does not prove those standards. Methodological Individualism may aggregate actor-level explanations into a social one, whereas this entry can model one actor.[ref-83bf50809d8b][ref-67c7263ecca4]
References¶
[^ref-83bf50809d8b]: Kelvin J. Lancaster, “A New Approach to Consumer Theory,” Journal of Political Economy 74, no. 2 (1966): 132–157, https://doi.org/10.1086/259131. Full original journal scan at https://bpb-us-e1.wpmucdn.com/sites.psu.edu/dist/c/13885/files/2014/07/Lancaster1966_A-New-Approach-to-Consumer-Theory.pdf, especially §§III–IV, printed pp.135–137 (PDF pp.5–7). [^ref-67c7263ecca4]: Gary S. Becker, “Crime and Punishment, An Economic Approach” (title punctuation transcribed as a comma for the reference binder; the original prints a colon), Journal of Political Economy 76, no. 2 (1968): 169–217, https://doi.org/10.1086/259394. Full original journal scan at https://laws21.classes.ryansafner.com/readings/Becker-1968.pdf, especially §II, printed pp.176–179 (PDF pp.9–12). [^ref-26f0faae2618]: William H. Riker and Peter C. Ordeshook, “A Theory of the Calculus of Voting,” American Political Science Review 62, no. 1 (1968): 25–42, https://doi.org/10.2307/1953324. Original publisher extract and accessible note 7 at https://www.cambridge.org/core/journals/american-political-science-review/article/abs/theory-of-the-calculus-of-voting/F32A6F556750C8207E920CE9D0850A2E; full body not inspected. [^ref-faa6e6695f8d]: Nolan McCarty and Adam Meirowitz, “The Theory of Choice,” in Political Game Theory (Cambridge University Press, 2007), https://doi.org/10.1017/CBO9780511813122.002. Publisher summary at https://www.cambridge.org/core/books/abs/political-game-theory/theory-of-choice/87D6B965B51CAB5FF5E33A74BF4FF1E2; full chapter not inspected.