Domain is the foundation of the entire Science Analysis Template. For Interaction, it defines the conceptual world in which multiple agents confront one another through markets, strategic environments, incentives, contracts, and institutional rules. It specifies what counts as an interaction, what kinds of agents and structures populate the system, how their choices influence one another, and which simplifying assumptions make multi-agent analysis tractable. Before equilibrium can be defined, before incentives can be compared, before mechanisms can be engineered, the science must first declare its Domain: the boundaries that separate strategic interdependence from solitary choice and macro-level aggregation; the scale at which multiple agents and institutions operate; the entities that constitute markets, games, and mechanisms; the properties that describe their incentives and information; and the assumptions under which strategic behavior can be coherently analyzed. By fixing these commitments up front, the Domain of Interaction provides the structural framework that all later reasoning—market models, game theory, mechanism design, institutional analysis—must respect, ensuring that every result fits within a unified conception of how agents influence one another through rules, strategies, and equilibria.
Interaction (Markets, Strategy & Mechanisms) – Domain – SAT
| Element | Interaction (Markets, Strategy & Mechanisms) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Scope Category | 1.1 Scope of the Domain | 1.2 Ontological Commitments | 1.3 State-Variables | 1.4 Admissible Idealizations | 1.5 Domain Assumptions | 1.6 Internal Coherence Requirements | |||||||
| Sub-Item | Interaction – Boundaries | Interaction – Scale | Interaction – Entities | Interaction – Properties | Interaction – Categories | Interaction – Variables | Interaction – Parameterization | Interaction – Simplifications | Interaction – Validity Conditions | Interaction – Structural Assumptions | Interaction – Implicit Commitments | Interaction – Consistency | Interaction – Compatibility |
1.1 Scope of the Domain
Scope of the Domain defines what Interaction is allowed to study and the level at which it explains multi-agent strategic behavior. Boundaries specify when a situation truly involves interdependence—when each agent’s payoff, constraints, or opportunities are shaped by the actions and expectations of others. Scale fixes the resolution at which these relationships are represented: the level of markets, games, institutions, or mechanisms where strategic adjustments, incentive responses, and equilibrium formations occur. Together, Boundaries and Scale carve out the legitimate terrain of this science: the world where choices collide, information asymmetries matter, rules structure incentives, and outcomes emerge from the strategic interplay of multiple agents rather than from isolated optimization or aggregate macro forces.
Boundaries specify exactly which phenomena count as genuine strategic interaction and which do not. Interaction is concerned with environments in which an agent’s outcomes depend on the actions, expectations, or information of other agents, transmitted through markets, incentives, contracts, or institutional rules. Anything outside this condition—such as isolated optimization or system-wide macro behavior—falls beyond its scope. Clear boundaries prevent the concept from collapsing into either solitary decision-making or aggregate dynamics, preserving the distinct terrain of multi-agent interdependence. They signal to researchers and readers alike that Interaction focuses on the structure and consequences of strategic linkages among agents, ensuring that its explanations are interpreted within the correct scientific framework and not mistaken for micro-level choice or macro-level evolution.
Scale specifies the level of granularity at which strategic interdependence is analyzed. In Interaction, the relevant scale is neither the isolated individual nor the macroeconomic aggregate, but the domain where multiple decision-makers, institutions, and rules jointly determine outcomes. This includes markets with many participants, bilateral or multilateral games, bargaining environments, networks of exchange, auction platforms, and mechanisms that coordinate or discipline behavior. Temporal scale may range from single-shot encounters to repeated or dynamic strategic processes, and informational scale concerns what agents know, observe, or believe about one another. Being explicit about scale ensures that the analytical tools—equilibrium concepts, incentive constraints, mechanism design principles—are applied at the appropriate resolution and that conclusions are not misinterpreted as statements about solitary optimization or emergent macro dynamics.
1.2 Ontological Commitments
Ontological Commitments specify what Interaction assumes is real within the world of strategic interdependence: the agents who act, the information they possess, the incentives that shape their choices, the institutional structures that channel their behavior, and the mechanisms through which their actions affect one another. These commitments establish the fundamental building blocks of the interactive domain—strategic players, actions, payoffs, information sets, market rules, and institutional constraints—and determine how explanations must be framed. They clarify which elements constitute the strategic environment, how those elements are described, and what kinds of relationships or dependencies are considered meaningful. By anchoring the theory in a well-defined ontology, Interaction ensures that its models, measurements, and equilibria are organized around entities and properties that actually matter for multi-agent behavior, and that its results remain coherent across different market forms, games, and mechanisms.
Entities are the fundamental objects and actors that Interaction presupposes as real within a strategic environment. In this domain, the primary entities are not isolated decision-makers but strategic agents whose choices affect—and are affected by—the choices of others. Alongside agents, the ontology includes actions, payoff structures, information sets, signals, strategic environments (such as markets, bargaining games, auctions, or networks), and institutional rules that govern how interactions unfold. These entities form the building blocks of analysis: they determine what can be chosen, what can be observed, what incentives exist, and how outcomes arise from interdependent behavior. Clarifying these entities is essential because strategic research must know precisely which objects it is modeling—who interacts, what they can do, what they know, and how the rules connect them. An explicit ontology also prevents confusion between levels of analysis: for instance, whether a price is treated as an exogenous parameter or an endogenous outcome, or whether a mechanism is an entity with its own structure rather than an emergent feature of agent behavior. Interaction proceeds by treating these agents, actions, information structures, and mechanisms as the core constituents of its conceptual world.
Properties are the characteristics or attributes of the entities that matter for strategic interdependence. For agents, these include preferences, beliefs, risk attitudes, information endowments, strategic sophistication, and incentive sensitivities. For actions and strategies, relevant properties include feasibility, observability, commitment structure, and whether choices are discrete, continuous, or contingent on others’ behavior. For mechanisms and institutions, properties include allocation rules, incentive compatibility, information requirements, enforcement capacity, and the degree to which they shape or restrict the strategic space. These attributes are critical because they determine how agents interact, how incentives operate, and how outcomes emerge from the interplay of multiple decision-makers. Identifying the key properties ensures that research focuses on the strategic dimensions that actually drive behavior—what agents know, what they can do, how much their actions matter for others, and how rules structure their possibilities. In analytical terms, specifying properties enables the quantitative modeling of payoff functions, information partitions, strategy sets, and equilibrium conditions, and allows researchers to formulate hypotheses about how changes in one property—such as information precision, competitive intensity, or enforcement strength—alter strategic outcomes.
Categories are the classification schemes that organize the entities and properties of strategic environments into meaningful types. Interaction requires a taxonomy that distinguishes different kinds of agents (buyers, sellers, bidders, regulators, competitors, coalition members), different kinds of strategic environments (markets, bargaining games, auctions, matching systems, networks), and different classes of actions (prices, bids, quantities, signals, commitments). It also includes categories for information structures (complete, incomplete, asymmetric), institutional architectures (mechanisms, contracts, rules), and outcome concepts (equilibrium types, allocation rules, payoff structures). These conceptual groupings make strategic complexity tractable by placing similar phenomena into coherent buckets—allowing researchers to know whether they are analyzing competition or cooperation, signaling or screening, mechanism design or market equilibrium. Defining categories is essential because the structure of Interaction depends on whether an object is treated as an agent, an environment, a rule, or a strategic function; misclassifying these leads to incoherent models. Clear categories ensure that when theorists invoke a concept like “market,” “game,” or “incentive,” they are referring to a well-defined class with consistent properties, enabling precise communication and stable theoretical architecture.
1.3 State-Variables
State-Variables define how Interaction represents the evolving condition of a strategic environment. They track the features of agents, markets, and mechanisms that change over time or across informational states—such as beliefs, strategies, price vectors, allocations, signals, and institutional states. Parameterization determines how these variables are encoded: the structure of strategy spaces, the representation of information partitions, the dimensions of payoff functions, and the mathematical form of market or mechanism rules. Together, state-variables and their parameterizations translate the ontology of Interaction into a quantifiable framework, allowing strategic behavior, informational changes, and institutional effects to be analyzed with precision and coherence.
Variables in Interaction are the measurable or definable quantities that capture the evolving state of a strategic environment. They represent properties of agents, information structures, actions, and institutional settings that can change and that determine how interdependent decisions unfold. These include strategy profiles, price vectors, bids, quantities, beliefs, information signals, contract terms, allocation outcomes, and any institutional parameters that structure strategic possibilities. Each variable must correspond to something that can be observed, inferred, or calculated within the interaction: what agents know, what they choose, what the mechanism reveals, and how payoffs are determined. Identifying these state-variables is essential because they encode the strategic condition of the system at any moment—what actions are available, what beliefs are held, what constraints are active, and what rules govern behavior. By tracking these variables, researchers can quantify how incentives shift, how expectations evolve, how markets adjust, and how mechanisms produce outcomes. They form the operational bridge between Interaction’s conceptual ontology (agents, actions, information, rules) and its formal analysis through equilibrium, optimization, and mechanism design.
Parameterization determines how the variables of a strategic environment are formally represented so they can be analyzed with precision. It specifies the structure of strategy spaces (continuous, discrete, mixed), the encoding of information (partitions, signals, belief vectors), the mathematical form of payoff functions, and the rules that define market or mechanism operation (allocation functions, pricing rules, incentive constraints). It also involves choosing the appropriate level of detail—whether strategies are parameterized as full contingent action plans or as simplified choices; whether information is represented through full Bayesian belief distributions or through coarse signals; whether prices and quantities are modeled continuously or discretely.
A good parameterization captures the essential degrees of freedom of strategic behavior while avoiding unnecessary complexity. It determines what aspects of the environment are treated as fixed parameters (institutional rules, mechanism design features, exogenous shocks) and which are endogenous outcomes (equilibrium prices, allocations, strategy choices). Poor parameterization can obscure incentive effects, misrepresent information flow, or improperly constrain equilibrium analysis. Effective parameterization ensures that the strategic dynamics are faithfully represented, that comparisons across environments remain coherent, and that the model isolates the forces genuinely responsible for interaction outcomes.
1.4 Admissible Idealizations
Admissible Idealization specifies which simplifications Interaction is allowed to make in order to analyze strategic behavior effectively. Simplified models reduce the complexity of multi-agent environments to their essential incentive structures, limiting attention to the actions, beliefs, and institutional rules that drive interdependence. These idealizations include assumptions about rationality, common knowledge, equilibrium formation, information structure, and institutional clarity. Limit conditions identify where such abstractions reliably capture strategic dynamics and where they break down—such as in environments with bounded rationality, unstable expectations, opaque rules, or severe informational frictions. Together, admissible idealizations formalize the acceptable distance between real-world strategic complexity and the representations used to study markets, games, and mechanisms, ensuring that Interaction remains tractable without detaching from the phenomena it seeks to explain.
Simplified models in Interaction are deliberate abstractions that strip away the full complexity of real strategic environments in order to isolate the core forces that generate interdependence. They idealize agents, information, and institutions so that strategic behavior can be analyzed with clarity rather than drowned in detail. Examples include assuming perfectly rational and payoff-maximizing agents, common knowledge of rationality, complete or precisely structured information partitions, frictionless markets, or mechanisms that operate with ideal enforcement and zero transaction costs. These simplifications make it possible to derive equilibria, characterize incentive effects, and understand how rules shape outcomes without having to model every psychological nuance, institutional imperfection, or informational asymmetry of real-world settings.
Admissible idealization means such simplified models are permitted as long as they reveal the essential incentive mechanisms at work and do not distort the logic of strategic interaction. They serve as conceptual tools that highlight how coordination, conflict, signaling, screening, competition, and institutional design operate at their core. As in all scientific practice, these simplified models are not literal descriptions of reality; they are approximations that allow tractable reasoning about multi-agent behavior. Their usefulness depends on judicious application and on clear awareness of the assumptions being made and the contexts where those assumptions no longer hold.
Every simplified model of strategic interaction has a domain of validity—conditions under which its assumptions faithfully capture the incentives at work, and circumstances where those assumptions cease to be credible. Limit conditions specify these boundaries. For example, models that rely on perfect rationality, common knowledge, or complete information break down when agents face severe cognitive limits, when beliefs are unstable or systematically biased, or when information is fragmented, costly, or strategically manipulated. Likewise, idealized competitive market models fail when market power, frictions, or institutional constraints dominate behavior; simplified mechanism-design models fail when enforcement is weak, participation is uncertain, or agents cannot fully commit to strategies.
Recognizing these limit conditions is essential for scientific rigor: it identifies when classical game-theoretic or market-equilibrium abstractions can be trusted and when richer modeling of cognition, institutions, or information flow is required. It also guides theoretical development—regions where idealizations fail often reveal where new mechanisms, equilibrium concepts, or behavioral refinements are needed. Stating limit conditions prevents the overextension of Interaction models beyond the environments they were meant to describe and clarifies the scope within which their insights remain reliable.
1.5 Domain Assumptions
Domain Assumptions articulate the background commitments Interaction takes for granted when analyzing strategic environments. Structural assumptions specify the fundamental stances that shape how games, markets, and mechanisms are modeled—such as whether agents act rationally, whether information structures are well-defined, whether institutions enforce rules reliably, and whether strategic behavior unfolds in discrete or continuous spaces. Implicit commitments capture the unspoken defaults embedded in the field, such as the expectation that agents form coherent beliefs about others, that incentives guide behavior, and that equilibrium concepts meaningfully describe strategic outcomes. Together, these assumptions form the unseen scaffolding that governs how Interaction interprets multi-agent phenomena and what kinds of explanations it considers acceptable, ensuring that different models within the domain operate under a coherent conceptual and methodological framework.
Structural Assumptions in Interaction are the deep, domain-wide commitments about how strategic environments function—assumptions that are rarely tied to a single model but permeate the entire analytical approach. They include stances about agent rationality (perfect, bounded, or rule-based), how beliefs are formed and updated (Bayesian or otherwise), whether interactions unfold in continuous or discrete action spaces, whether information is static or dynamically revealed, and whether institutions enforce rules reliably or only probabilistically. These assumptions shape the mathematical and conceptual form of strategic models: perfect rationality and common knowledge support equilibrium analysis, continuous action spaces invite calculus-based optimization, discrete strategy sets lead to combinatorial game structures, and well-defined information partitions enable Bayesian reasoning.
Such assumptions matter because Interaction relies heavily on the architecture of incentives, beliefs, and rules; if the underlying stance is wrong or only approximately true—for example, if agents lack the foresight assumed by classical equilibrium concepts or if enforcement institutions behave erratically—then the resulting models can misrepresent strategic dynamics. Making these structural commitments explicit allows the field to question, refine, or replace them as evidence and theory evolve, just as shifts in other sciences forced revisions to foundational assumptions about determinism, continuity, or randomness. In Interaction, these foundational choices determine what counts as a strategic explanation and anchor the entire modeling enterprise.
Implicit Commitments in Interaction are the unstated assumptions and inherited conventions that make strategic modeling possible but often go unacknowledged. Researchers typically presume, without explicitly stating, that agents can form coherent expectations about others; that incentives reliably influence behavior; that strategic reasoning, whether perfect or bounded, is meaningfully structured rather than arbitrary; that institutions function with enough stability to shape outcomes; and that equilibrium concepts—Nash, subgame-perfect, Bayesian, competitive—capture something real about how multi-agent systems settle. These commitments also include assumptions about communication, observability, rule-following, and the interpretability of signals, all of which underlie core models but are rarely spelled out.
These tacit beliefs shape how Interaction frames problems, what counts as a valid explanation, and which mechanisms or equilibria are considered plausible. They remain largely invisible until challenged—such as when behavioral evidence contradicts classical rationality, when institutions fail to enforce rules, or when strategic behavior departs systematically from equilibrium predictions. Making these implicit commitments explicit allows the field to evaluate their scope, revise them when necessary, and understand why different modeling traditions (e.g., classical game theory, behavioral strategy, experimental economics, mechanism design) sometimes talk past one another. Examining these hidden assumptions clarifies the conceptual foundations of Interaction and prevents its theoretical structure from resting on untested or outdated defaults.
1.6 Internal Coherence Requirements
Internal Coherence Requirements ensure that Interaction operates as a unified, logically consistent science rather than a patchwork of incompatible models. Consistency requires that its foundational concepts—strategic agents, incentives, beliefs, information structures, mechanisms, and equilibrium notions—do not contradict one another across different applications. Compatibility demands that entities, variables, assumptions, and modeling tools integrate into a single workable framework: payoff functions must align with strategy spaces, information structures must match the equilibrium concept employed, and institutional rules must be modeled in ways that support, rather than undermine, strategic reasoning. Together, these coherence requirements impose the discipline needed for Interaction to produce interpretable and reliable explanations of multi-agent behavior, ensuring that different strands of the field—market theory, game theory, mechanism design, bargaining, and network analysis—fit within the same conceptual architecture.
Consistency requires that all components of Interaction—its definitions of agents, strategies, information, incentives, mechanisms, and equilibrium concepts—fit together without contradiction. A strategic model cannot assume agents have fixed beliefs in one part and then require changing beliefs in another without specifying how; it cannot define a mechanism’s rules one way and apply them differently in equilibrium analysis; it cannot treat information as symmetric in one section and asymmetric in another unless the distinction is formalized. Terms like “strategy,” “signal,” “market,” or “equilibrium” must be used with stable meanings across the field, or the resulting theory collapses into incoherence.
More subtly, theoretical predictions from different layers of Interaction must align. For example, a model of market competition cannot rely on behavioral assumptions that contradict the assumptions underpinning the mechanism design used to evaluate regulatory interventions. Internal consistency ensures that the domain’s complex network of strategic reasoning, equilibrium logic, and institutional detail forms a single, logically integrated framework. Without consistency, at least one part of the theory must be wrong—and the strategic explanations that depend on it lose all credibility. Maintaining consistency requires explicit definitions, disciplined mathematical formulation, and care in extending models so new components reinforce rather than undermine the established architecture.
Compatibility requires that every component of Interaction—its agents, strategies, information structures, incentives, mechanisms, equilibrium concepts, and domain assumptions—integrates into a unified explanatory framework. It is not enough for the pieces to avoid contradiction; they must fit together in a way that allows strategic environments to be modeled and interpreted coherently. The entities defined in one part of the theory (agents, markets, rules) must align with the relationships and laws specified elsewhere (best responses, incentive constraints, equilibrium conditions). Variables such as beliefs, prices, allocations, or strategy profiles must meaningfully connect to the payoff structures and institutional rules that determine outcomes. If a concept cannot be related to others—if, for example, a mechanism includes rules that never influence incentives, or a belief variable is introduced but never affects strategic behavior—the framework becomes fragmented rather than unified.
Compatibility also demands harmony between domain assumptions and analytical tools. Assumptions about rationality, information, or enforcement must match the evidence used to justify equilibrium modeling or mechanism design. A theory that models agents as perfectly rational while simultaneously invoking behavioral deviations without formal integration is incompatible; a theory that assumes complete information but uses equilibrium concepts requiring uncertainty is likewise incoherent.
A compatible Interaction framework ensures that all components reinforce one another, producing a cohesive system where models, methods, and interpretations interlock. This coherence strengthens explanatory power, enables connections across different market forms and strategic settings, and prevents the domain from devolving into isolated sub-theories that cannot speak to each other.