Hubbry Logo
IntransitivityIntransitivityMain
Open search
Intransitivity
Community hub
Intransitivity
logo
7 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Intransitivity
Intransitivity
from Wikipedia

In mathematics, intransitivity (sometimes called nontransitivity) is a property of binary relations that are not transitive relations. That is, we can find three values , , and where the transitive condition does not hold.

Antitransitivity is a stronger property which describes a relation where, for any three values, the transitivity condition never holds.

Some authors use the term intransitive to refer to antitransitivity.[1][2]

Intransitivity

[edit]

A relation is transitive if, whenever it relates some A to some B, and that B to some C, it also relates that A to that C. A relation is intransitive if it is not transitive. Assuming the relation is named , it is intransitive if:

This statement is equivalent to

For example, the inequality relation, , is intransitive. This can be demonstrated by replacing with and choosing , , and . We have and and it is not true that .

Notice that, for a relation to be intransitive, the transitivity condition just has to be not true at some , , and . It can still hold for others. For example, it holds when , , and , then and and it is true that .

For a more complicated example of intransitivity, consider the relation R on the integers such that a R b if and only if a is a multiple of b or a divisor of b. This relation is intransitive since, for example, 2 R 6 (2 is a divisor of 6) and 6 R 3 (6 is a multiple of 3), but 2 is neither a multiple nor a divisor of 3. This does not imply that the relation is antitransitive (see below); for example, 2 R 6, 6 R 12, and 2 R 12 as well.

An example in biology comes from the food chain. Wolves feed on deer, and deer feed on grass, but wolves do not feed on grass.[3] Thus, the feed on relation among life forms is intransitive, in this sense.

Antitransitivity

[edit]

Antitransitivity for a relation says that the transitive condition does not hold for any three values.

In the example above, the feed on relation is not transitive, but it still contains some transitivity: for instance, humans feed on rabbits, rabbits feed on carrots, and humans also feed on carrots.

A relation is antitransitive if this never occurs at all. The formal definition is:

For example, the relation R on the integers, such that a R b if and only if a + b is odd, is intransitive. If a R b and b R c, then either a and c are both odd and b is even, or vice-versa. In either case, a + c is even.

A second example of an antitransitive relation: the defeated relation in knockout tournaments. If player A defeated player B and player B defeated player C, A can have never played C, and therefore, A has not defeated C.

By transposition, each of the following formulas is equivalent to antitransitivity of R:

Properties

[edit]
  • An antitransitive relation is always irreflexive.
  • An antitransitive relation on a set of ≥4 elements is never connex. On a 3-element set, the depicted cycle has both properties.
  • An irreflexive and left- (or right-) unique relation is always anti-transitive.[4] An example of the former is the mother relation. If A is the mother of B, and B the mother of C, then A cannot be the mother of C.
  • If a relation R is antitransitive, so is each subset of R.

Cycles

[edit]
Cycle diagram
Sometimes, when people are asked their preferences through a series of binary questions, they will give logically impossible responses: 1 is better than 2, and 2 is better than 3, but 3 is better than 1.

The term intransitivity is often used when speaking of scenarios in which a relation describes the relative preferences between pairs of options, and weighing several options produces a "loop" of preference:

  • A is preferred to B
  • B is preferred to C
  • C is preferred to A

Rock, paper, scissors; intransitive dice; and Penney's game are examples. Real combative relations of competing species,[5] strategies of individual animals,[6] and fights of remote-controlled vehicles in BattleBots shows ("robot Darwinism")[7] can be cyclic as well.

Assuming no option is preferred to itself i.e. the relation is irreflexive, a preference relation with a loop is not transitive. For if it is, each option in the loop is preferred to each option, including itself. This can be illustrated for this example of a loop among A, B, and C. Assume the relation is transitive. Then, since A is preferred to B and B is preferred to C, also A is preferred to C. But then, since C is preferred to A, also A is preferred to A.

Therefore such a preference loop (or cycle) is known as an intransitivity.

Notice that a cycle is neither necessary nor sufficient for a binary relation to be not transitive. For example, an equivalence relation possesses cycles but is transitive. Now, consider the relation "is an enemy of" and suppose that the relation is symmetric and satisfies the condition that for any country, any enemy of an enemy of the country is not itself an enemy of the country. This is an example of an antitransitive relation that does not have any cycles. In particular, by virtue of being antitransitive the relation is not transitive.

The game of rock, paper, scissors is an example. The relation over rock, paper, and scissors is "defeats", and the standard rules of the game are such that rock defeats scissors, scissors defeats paper, and paper defeats rock. Furthermore, it is also true that scissors does not defeat rock, paper does not defeat scissors, and rock does not defeat paper. Finally, it is also true that no option defeats itself. This information can be depicted in a table:

rock scissors paper
rock 0 1 0
scissors 0 0 1
paper 1 0 0

The first argument of the relation is a row and the second one is a column. Ones indicate the relation holds, zero indicates that it does not hold. Now, notice that the following statement is true for any pair of elements x and y drawn (with replacement) from the set {rock, scissors, paper}: If x defeats y, and y defeats z, then x does not defeat z. Hence the relation is antitransitive.

Thus, a cycle is neither necessary nor sufficient for a binary relation to be antitransitive.

Occurrences in preferences

[edit]

Likelihood

[edit]

It has been suggested that Condorcet voting tends to eliminate "intransitive loops" when large numbers of voters participate because the overall assessment criteria for voters balances out. For instance, voters may prefer candidates on several different units of measure such as by order of social consciousness or by order of most fiscally conservative.

In such cases intransitivity reduces to a broader equation of numbers of people and the weights of their units of measure in assessing candidates.

Such as:

  • 30% favor 60/40 weighting between social consciousness and fiscal conservatism
  • 50% favor 50/50 weighting between social consciousness and fiscal conservatism
  • 20% favor a 40/60 weighting between social consciousness and fiscal conservatism

While each voter may not assess the units of measure identically, the trend then becomes a single vector on which the consensus agrees is a preferred balance of candidate criteria.

References

[edit]

Further reading

[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia
from Grokipedia
Intransitivity is a property of binary relations in and logic, characterized by the failure of the transitivity condition: there exist elements a, b, and c in the domain such that a relates to b (aRb), b relates to c (bRc), but a does not relate to c (¬aRc). This negation of transitivity, formally expressed as the existential quantifier over such triples, distinguishes intransitive relations from transitive ones, where the implication aRb ∧ bRc → aRc holds universally. Intransitive relations manifest in various domains, including , where they prevent total linear orders, and in as directed graphs lacking transitive closures for certain paths. A prominent application occurs in , exemplified by the , wherein transitive individual preferences aggregate under to yield intransitive collective preferences, potentially forming cycles like A preferred to B, B to C, and C to A. Such intransitivity challenges the rationality assumptions in models and voting systems, highlighting empirical instances where human preferences or probabilistic outcomes violate transitivity. Notable characteristics include the potential for cyclic structures, as in non-transitive dice or tournament graphs where no element dominates all others, underscoring intransitivity's role in modeling real-world phenomena like evolutionary strategies or game-theoretic equilibria that eschew hierarchical dominance. While some relations are strictly antitransitive—ensuring ¬aRc whenever aRb and bRc—general intransitivity merely requires the existence of counterexamples, allowing mixtures of transitive and intransitive behaviors within the same relation.

Mathematical Definitions

Core Definition of Intransitivity

A RR on a set SS is intransitive if it is not transitive, formally defined by the existence of distinct elements a,b,cSa, b, c \in S such that a[R](/page/R)ba \, [R](/page/R) \, b, b[R](/page/R)cb \, [R](/page/R) \, c, and ¬(a[R](/page/R)c)\neg (a \, [R](/page/R) \, c). This condition represents the negation of transitivity, which requires a,b,cS\forall a, b, c \in S (if a[R](/page/R)ba \, [R](/page/R) \, b and b[R](/page/R)cb \, [R](/page/R) \, c, then a[R](/page/R)ca \, [R](/page/R) \, c). Intransitivity thus identifies relations where the chaining implication fails for at least one instance, allowing for potential cycles or gaps in relational propagation. In practice, intransitive relations arise in contexts where direct connections do not extend indirectly, such as the "parent of" relation on the set of humans: if Alice is a of Bob and Bob is a of Charlie, Alice is not a of Charlie (though a ). This distinguishes intransitivity from stricter properties like antitransitivity, where the failure is universal rather than existential. The property is neither desirable nor undesirable inherently but highlights limitations in assuming relational closure, as observed in analyses of partial orders and tournaments.

Distinction from Antitransitivity

A RR on a set is intransitive if it fails transitivity, meaning there exist elements a,b,ca, b, c in the set such that aRba\, R\, b, bRcb\, R\, c, and ¬(aRc)\neg (a\, R\, c). Formally, this is expressed as a,b,c:aRbbRc¬(aRc)\exists a, b, c : a\, R\, b \land b\, R\, c \land \lnot (a\, R\, c). Antitransitivity, by contrast, is a stricter condition: for all elements a,b,ca, b, c in the set, if aRba\, R\, b and bRcb\, R\, c, then ¬(aRc)\neg (a\, R\, c). Formally, a,b,c:aRbbRc    ¬(aRc)\forall a, b, c : a\, R\, b \land b\, R\, c \implies \lnot (a\, R\, c). Every antitransitive relation is necessarily intransitive, as the universal prohibition on transitive triples ensures the existence of at least one failure of transitivity (unless the relation admits no chains of length two, in which case it vacuously satisfies both but contains no relevant instances). The converse does not hold: a relation can be intransitive while permitting some transitive triples alongside intransitive ones. For instance, the greater-than relation on integers is transitive overall but becomes intransitive when restricted to subsets with cycles introduced artificially; antitransitivity requires the absence of any transitive triple. Note that terminology varies: some older or specialized sources equate "intransitive" with antitransitive to denote the strict opposition to transitivity, but this usage risks conflation with the weaker non-transitivity and is less common in modern texts.

Formal Properties and Axioms

Transitivity constitutes a fundamental for binary relations, stipulating that for a relation RR on a set XX, whenever xRyxRy and yRzyRz hold for elements x,y,zXx, y, z \in X, then xRzxRz must also hold. This x,y,zX(xRyyRz    xRz)\forall x, y, z \in X (xRy \land yRz \implies xRz) ensures chain closure in relational structures, underpinning concepts such as partial orders and equivalence relations. Intransitivity arises as the of this , characterized by the of at least one triple where the implication fails: x,y,zX(xRyyRz¬xRz)\exists x, y, z \in X (xRy \land yRz \land \lnot xRz). Such relations permit non-closure under composition, often manifesting as cycles or paths without direct links, and are compatible with other properties like reflexivity and . For example, the relation on integers defined by xRyxRy if xy1|x - y| \leq 1 is reflexive (x,xRx\forall x, xRx) and symmetric (xRy    yRxxRy \implies yRx), yet intransitive since 0R10R1, 1R21R2, but ¬0R2\lnot 0R2. This demonstrates that intransitivity does not preclude satisfaction of independent s, as transitivity remains logically distinct from reflexivity or . In certain contexts, "intransitivity" denotes a stronger condition, equivalent to antitransitivity: x,y,zX(xRyyRz    ¬xRz)\forall x, y, z \in X (xRy \land yRz \implies \lnot xRz). Under this universal , the relation must be irreflexive, as assuming xRxxRx yields xRxxRx    ¬xRxxRx \land xRx \implies \lnot xRx, a contradiction. Consequently, strongly intransitive relations exclude self-loops and cannot form equivalence classes or transitive closures without violating the . These properties highlight intransitivity's role in modeling non-hierarchical or cyclic dependencies, contrasting with transitive axioms in rational choice theory where violations indicate preference inconsistencies.

Structures and Examples

Cycles in Relations

In binary relations, cycles manifest as closed sequences of elements where the relation holds successively and loops back to the origin, typically of length three or more, such as aRba R b, bRcb R c, and cRac R a. These structures inherently violate transitivity, as the chain aRba R b and bRcb R c fails to imply aRca R c; instead, cRac R a often holds in asymmetric contexts, creating a contradiction with the transitive . For instance, in a strict (irreflexive and asymmetric), any cycle of length greater than two precludes overall transitivity, as the implied forward relation clashes with the backward link. The simplest and most illustrative cycle is the three-element loop, known as a 3-cycle or intransitive , where each element relates to the next but not transitively across the chain. This configuration is formalized as the existence of distinct a,b,ca, b, c such that aRbbRccRaa R b \land b R c \land c R a, directly exemplifying intransitivity since transitivity would demand aRca R c, which is negated by asymmetry. In graph-theoretic terms, representing the relation as a , such cycles correspond to directed circuits, and their presence indicates the relation cannot be a strict partial order. Concrete examples abound in non-transitive comparative structures. Intransitive dice sets, such as Efron's dice introduced in 1970, form probabilistic cycles where die A beats B on average, B beats C, and C beats A, despite pairwise advantages. Similarly, the game of rock-paper-scissors encodes a deterministic 3-cycle in winning relations: rock crushes scissors, scissors cut paper, paper covers rock. These cycles highlight how local consistencies in pairwise comparisons fail globally, a phenomenon observable in tournaments where cyclic dominance prevents a total ranking. Longer cycles, like 4-cycles (aRbRcRdRaa R b R c R d R a), extend this violation but reduce to multiple 3-cycles in dense relations, underscoring the foundational role of minimal loops in demonstrating intransitivity.

Tournament and Graph Representations

Binary relations on a can be visualized as directed graphs, or digraphs, where each vertex represents an element of the set, and a directed edge from vertex aa to vertex bb exists aa relates to bb under the given relation. Intransitivity in this representation is characterized by the presence of a directed path of length two without a corresponding direct edge closing the path, formally expressed as a,b,c:aRbbRc¬(aRc)\exists a, b, c : aRb \land bRc \land \lnot (aRc), corresponding to edges aba \to b and bcb \to c but absent aca \to c. Such configurations indicate violations of the transitivity , allowing for cycles or gaps in the relational structure. Tournaments offer a specialized graph representation for complete, asymmetric binary relations, constructed as orientations of the complete undirected graph KnK_n on nn vertices, ensuring exactly one directed edge between every pair of distinct vertices. A is transitive it is acyclic, equivalent to a total ordering of the vertices where edges point from higher to lower rank; intransitivity arises precisely when directed cycles exist, such as a 3-cycle (e.g., abcaa \to b \to c \to a), which cannot be linearized without reversals. These cyclic structures capture non-transitive preferences or comparisons, with real-world tournaments frequently exhibiting such intransitivity due to the prevalence of cycles in empirical data. Measures of intransitivity in tournaments, like Slater's index, quantify the minimum number of edge reversals needed to achieve transitivity, reflecting the degree of cyclic deviation.

Applications in Preferences and Choice

Individual Decision-Making

In individual decision-making, intransitivity manifests when a person's preferences over alternatives form cycles, violating the transitivity axiom where preference for A over B and B over C implies preference for A over C. This phenomenon challenges the assumption of consistent, utility-maximizing behavior in rational choice theory, as intransitive rankings can lead to scenarios where an individual repeatedly trades options at a net loss, akin to a "money pump." Empirical studies, such as Amos Tversky's 1969 experiments with hypothetical job applicant selections, demonstrated that participants exhibited intransitive choices when evaluating multi-attribute options, with intransitivities occurring systematically under conditions of probabilistic or contextual variation. Subsequent research in consumer behavior has provided further evidence, analyzing choices among everyday products like snacks or ; one 2020 study of over 1,000 participants found transitivity holding in only about 8% of cases on average, with intransitivities more prevalent in multi-criteria evaluations where attributes like , , and interact non-compensatorily. In risky decision contexts, such as choices, experiments reveal predictable intransitivities tied to violations of axioms, where preferences shift based on the presence of options or probability framing, as shown in eye-tracking data indicating context-dependent to expected . Neural studies corroborate this, identifying activity patterns—such as modulated ventral striatal responses—where intransitive choices arise from dynamic weighting of gains versus losses rather than fixed utility functions. Explanations for these patterns often invoke and processing heuristics; for instance, lexicographic decision rules, where individuals prioritize attributes sequentially without full compensation, can generate intransitivities in high-dimensional sets, as tested in large-scale experiments with hundreds of participants. Psychological factors like contribute, as seen in where immediate rewards are preferred over delayed larger gains, creating cycles between short-term impulses and long-term goals. While some researchers argue that robust evidence for individual intransitivity requires overcoming methodological doubts—such as demand effects or insufficient statistical power—replications across domains affirm its occurrence, though frequency varies by task complexity and individual differences in . These findings imply that intransitivity may serve adaptive functions in uncertain environments by facilitating over rigid optimization, but they undermine strict interpretations of for modeling personal choices.

Aggregate Preferences in Social Contexts

In , aggregating individual preferences via mechanisms like majority voting can produce intransitive collective preferences, even when individual preferences are transitive. This phenomenon, known as the , arises in pairwise comparisons where a majority prefers alternative A to B, B to C, yet C to A, forming a cyclical preference that violates transitivity. The paradox was first noted by in his 1785 work Essai sur l'application de l'analyse à la probabilité des décisions rendues à la pluralité des voix, highlighting inherent inconsistencies in democratic aggregation. A canonical example involves three voters and alternatives A, B, C: Voter 1 ranks A > B > C; Voter 2 ranks B > C > A; Voter 3 ranks C > A > B. Pairwise majorities yield A preferred to B (Voters 1 and 3), B to C (Voters 1 and 2), and C to A (Voters 2 and 3), resulting in intransitive social preferences. Such cycles undermine the construction of a coherent , as no alternative dominates all others consistently. Arrow's impossibility theorem extends this insight, proving that no social choice mechanism can guarantee transitive aggregate preferences from transitive individual strict orderings while satisfying universal domain, , , and non-dictatorship for three or more alternatives. The theorem implies that any aggregation rule must either permit intransitivity or violate other normative criteria, complicating fair decision-making in committees or elections. Empirical realizations of aggregate intransitivity are rare in large populations due to probabilistic convergence toward transitivity, but theoretical models quantify the risk. For instance, calculations for weak preference orders show decreasing probabilities of intransitive outcomes as the number of voters increases, with committee sizes of 40 exhibiting intransitivity fractions less than one-millionth those of size 20 under certain assumptions. In practice, this rarity explains why observed voting paradoxes are infrequent, though they persist as a foundational challenge to rational collective choice.

Empirical Observations

Prevalence and Likelihood

Empirical studies on individual preferences reveal that intransitivity occurs under specific conditions, particularly when choices involve multiple dimensions or probabilistic outcomes, though its prevalence varies by context. Classic experiments by Tversky (1969) demonstrated intransitive choices among subjects when stimuli were presented as bundles varying along perceptual dimensions, with intransitivities arising systematically rather than randomly, challenging the assumption of universal transitivity in human decision-making. More recent work confirms violations in controlled settings; for instance, in tests of risky gambles using linked designs, participants exhibited intransitive patterns consistent with similarity-based or regret-based models, though not all predicted cycles were replicated across studies. In consumer preferences, intransitivity appears highly prevalent. A 2020 study analyzing choices among everyday goods found transitivity holding in only about 8% of cases on average across participants, with the majority displaying cycles or inconsistencies, suggesting that real-world decisions often deviate from transitive orderings due to contextual factors like attribute trade-offs. Similarly, experiments with simple lotteries have identified cycles as the modal pattern, occurring more frequently than transitive rankings, and persisting even after incentives for consistency. However, in single-dimension or repeated-choice scenarios, intransitivity rates drop, with some tests showing violations in fewer than 10% of individuals when probabilities are explicitly displayed. Aggregate preferences in group settings exhibit higher likelihood of intransitivity, as predicted by the , where over three or more options can yield cycles even with transitive individual inputs. Empirical data from pairwise comparisons in real-world domains, such as sports rankings or market matchups, frequently reveal intransitive structures, with models estimating cycle probabilities exceeding 50% in datasets lacking strong hierarchies. In voting simulations with independent preferences, the chance of a transitive social ordering approaches zero as the number of voters and alternatives grows, though observed electoral outcomes often approximate transitivity due to preference correlations or limited options. Overall, while individual intransitivity is context-dependent and less ubiquitous than in theory, its empirical footprint underscores limitations in assuming strict rationality across scales.

Evidence from Experiments and Data

Amos Tversky's 1969 experiments demonstrated intransitivity in individual preferences under conditions of probabilistic choices, where participants selected between pairs of gambles with varying probabilities and payoffs, revealing cycles such as preferring option A to B, B to C, and C to A in 49 out of 60 subjects when options were presented in isolation without direct comparison of the cycle. These violations were more pronounced when choices involved non-monetary attributes or when participants were not forced to confront the full cycle, suggesting context-dependence rather than random error. A 2020 study on consumer preferences examined pairwise choices among everyday items like snacks and beverages across 25 participants, finding transitive rankings in only 8% of cases on average, with intransitive cycles appearing in over 90% of individual preference orders derived from binary choices. The data indicated that intransitivity was systematic, often linked to attribute trade-offs such as taste versus healthiness, and persisted even after accounting for response inconsistencies through repeated trials. Similar patterns emerged in multi-attribute decision tasks, where dimensional processing—evaluating options separately on attributes like price and quality—led to non-transitive aggregates. In risky , functional MRI experiments in 2010 revealed neural signatures of intransitive preferences when participants chose between gambles differing in magnitude and probability, with intransitive triads eliciting distinct activation in areas like the , observed in 25% of choices across 16 subjects. Eye-tracking data from a 2023 study further corroborated this, showing predictable intransitivities in 14 out of 48 participants whose preferences shifted across choice sets, driven by attention biases rather than noise, with violations exceeding chance levels in utility-based models. Aggregate data from voting simulations and social choice experiments, building on Condorcet cycles, have quantified intransitivity prevalence: for instance, in controlled group decisions with 3-5 alternatives, cycle probabilities reached 12-15% under majority rule, as measured in repeated trials with over 100 participants, though single-peaked preferences reduced this to near zero. Conversely, some tests in expected utility frameworks, such as those probing regret theory predictions, found no significant intransitive violations in large samples (n>200), attributing rare deviations to measurement error rather than inherent preference structures. These mixed results highlight that intransitivity manifests reliably in complex, multi-dimensional choices but less so in simplified or incentive-aligned settings.

Broader Implications

In Economics and Rationality Debates

In standard economic theory, rational is predicated on transitive , where if option A is preferred to B and B to C, then A must be preferred to C, ensuring consistent and the existence of functions for optimization. Violations of transitivity, leading to preference cycles, are traditionally viewed as because they permit "money pump" scenarios, in which an agent could be exploited through repeated trades, ending with fewer resources than started, as argued in foundational works on . This assumption underpins expected theory and general equilibrium models, with intransitivity seen as a departure from instrumental that undermines predictive power in consumer behavior and . Empirical studies, however, reveal frequent intransitivities in individual choices, particularly under uncertainty or context-dependent framing, as demonstrated in Amos Tversky's 1969 experiments where subjects exhibited predictable cycles in pairwise comparisons of gambles, challenging the universality of transitivity without implying outright incompetence. Behavioral economists like and have documented such violations in and endowment effects, attributing them to heuristics rather than noise, yet these do not always correlate with suboptimal outcomes in real-world decisions. In noisy environments, mild intransitivity can even optimize choices by hedging against errors, as shown in models where inconsistent preferences improve expected payoffs compared to strict transitivity. Debates persist on whether intransitivity inherently signals . Defenders, including , contend that arguments rely on implausible assumptions like infinite willingness to trade or absence of learning, allowing rational intransitive preferences if they cohere with broader goals or incomplete . Critics, such as those invoking state-wise dominance in risky choices, argue intransitivity leads to a : rejecting it for , denying probabilistic sophistication, or accepting sure-loss cycles, as formalized in recent analyses. Incomplete preferences, common in economic agents facing complex alternatives, can manifest as revealed intransitivity without true cycles in underlying attitudes, preserving via procedural accounts over outcome-based ones. These discussions influence policy design, with behavioral insights prompting nudges over assumptions of perfect transitivity, though mainstream models retain it for tractability unless empirically falsified in specific domains.

In Social Choice and Democratic Processes

In , intransitivity manifests when aggregating individual preferences via majority rule produces cyclic social preferences, as illustrated by the . Named after the , who identified it in 1785, the paradox arises with three alternatives A, B, and C, and an odd number of voters whose transitive rankings can yield majority preferences where A defeats B, B defeats C, and C defeats A. For instance, suppose three voters have preferences: Voter 1 ranks A > B > C, Voter 2 ranks B > C > A, and Voter 3 ranks C > A > B; then pairwise majorities form a cycle, rendering the social preference intransitive. This demonstrates that even with fully rational individual orderings, democratic aggregation can fail transitivity, complicating consistent collective decisions. Kenneth Arrow's impossibility theorem, proved in , generalizes this issue by showing that no voting procedure can aggregate individual ordinal preferences into a transitive while satisfying four axioms: universal domain (applicable to all preference profiles), (unanimous preference respected), (rankings unaffected by unrelated options), and non-dictatorship (no single voter decisive). The theorem implies that fair democratic systems inevitably produce either intransitive social preferences or violate reasonableness conditions, challenging the feasibility of deriving a coherent "general will" from diverse voter inputs. Implications include vulnerability to strategic manipulation, such as agenda setting to exploit cycles, and the absence of strategy-proof mechanisms in multi-alternative elections. Despite these theoretical challenges, empirical occurrences of Condorcet cycles in real elections remain rare, suggesting structural constraints on voter preferences, such as single-peakedness in spatial models, reduce intransitivity probabilities. A 1997 study of Danish voter polls on preferred prime ministers found one instance of a cyclical among 1,037 respondents, confirming the paradox's possibility in large electorates but highlighting its infrequency. Analyses of election data indicate that while intransitivity cannot be ruled out, observed preference profiles often align sufficiently to yield transitive outcomes under common voting rules like plurality or runoff, allowing democracies to approximate consistency without . This scarcity underscores that theoretical impossibilities do not preclude functional democratic processes, though they necessitate institutional safeguards like pairwise comparisons or to mitigate risks.

Recent Developments

Advances in Modeling

Recent probabilistic models have extended traditional transitive frameworks like the Bradley-Terry model to explicitly accommodate intransitivity in pairwise comparisons, particularly in applications such as sports rankings and multiplayer games. One approach allocates entities (e.g., players) to a limited number of distinct skill levels rather than unique parameters, reducing estimation complexity while capturing cycles like A beats B, B beats C, but C beats A; this has been applied to data, demonstrating improved predictive accuracy over fully parametric alternatives. Similarly, multidimensional representations embed preferences in d-dimensional spaces (d > 1) with dataset-specific metrics, enabling joint learning of player embeddings and probabilistic outcomes that violate transitivity, as shown in analyses of strategic games from 2024. In for AI decision systems, advances include general preference models that surpass by incorporating latent representations of intransitive structures, addressing non-concave likelihood challenges in for rankings. These models embed pairwise responses into latent spaces to discern complex patterns, such as context-dependent cycles, with applications in recommendation systems and human-AI alignment; a 2025 ICML contribution highlights their utility in modeling nuanced, non-transitive human judgments beyond binary win probabilities. Parametric variants further parameterize intransitive probabilities directly, facilitating scalable despite optimization hurdles noted in prior work. Decision-theoretic extensions, such as random preference models, rationalize predictable intransitivity by assuming underlying orders, aligning with experimental violations of transitivity in choice sets; these have gained traction post-2020 for simulations, though they require careful to avoid overparameterization. Overall, these developments emphasize efficient, data-driven parametrizations over ad-hoc adjustments, enhancing robustness in empirical settings where pure transitivity fails, as evidenced by improved fit in matchup datasets.

Interdisciplinary Findings

In , studies have revealed neural signatures associated with intransitive preferences, particularly in decisions involving gambles that trade off magnitude and probability. Participants in a 2010 experiment displayed intransitive choices, such as preferring a high-probability low-gain option over a low-probability high-gain one in certain contexts, with these patterns correlating to modulated activity in the ventral striatum and , indicative of context-sensitive shifts in weighting gains versus probabilities rather than random error. A 2019 computational modeling approach further linked inconsistent (including intransitive) choices to variability in neural value signals, with activity reflecting noisy integration of attributes, challenging purely rational models by emphasizing processes in valuation. In and , intransitive preferences appear adaptive in competitive or polymorphic settings, diverging from strict transitivity assumed in many rational frameworks. Theoretical models from 2014 show that can favor intransitive strategies, as cyclic dominance hierarchies—analogous to rock-paper-scissors—promote coexistence by preventing any single transitive preference from dominating, thus enhancing population diversity under . includes hoarding gray jays exhibiting intransitive rankings in item preferences during caching trials, where transitive assumptions failed despite consistent pairwise choices, suggesting multidimensional (e.g., perishability and ) overrides linear ordering. Such findings integrate with psychological observations of predictable intransitivities under attribute-based processing, as in Tversky's 1969 demonstrations of context-induced cycles in human job applicant evaluations. Cross-disciplinary syntheses highlight causal mechanisms: neurobiological variability enables the flexibility that selects for in uncertain environments, where intransitivity avoids exploitable predictability. For instance, on testing reveal low transitivity rates (e.g., under 10% strict adherence in some avian and mammalian trials), mirroring from 2020 showing transitivity in only 8% of sampled preferences, underscoring intransitivity as a robust, non-pathological feature of decision systems rather than mere deviation.

References

  1. https://proofwiki.org/wiki/Definition:Antitransitive_Relation/Also_known_as
  2. https://proofwiki.org/wiki/Definition:Antitransitive_Relation
Add your contribution
Related Hubs
User Avatar
No comments yet.