Scale in Choice determines the level at which an individual decision-maker is represented and the temporal scope over which their optimization unfolds. Because this domain isolates a single agent whose environment does not respond to their actions, the scale must be sharply defined. Too fine a scale obscures the simplicity of the agent’s structure; too broad a scale invites forces—strategic influence or macro adjustments—that do not belong to solitary optimization and would break the logic of the domain.

A well-specified scale clarifies which details matter for the agent’s decision, which elements remain fixed as parameters, and how time should be represented when only one decision-making unit is present. Establishing scale in this way ensures that the analysis remains internally coherent and that the problem does not drift into the terrain of Interaction or Aggregation.
The Scale Framework provides the structure needed to articulate these requirements with precision.

---

THE SCALE FRAMEWORK

1. Ontological Scale (What exists at this level of analysis?)

At the ontological scale of Choice, the only analytically meaningful unit is a single decision-making agent. This agent is treated as a unified entity whose internal components—psychology, sub-motives, departments, workers, neurons—are not modeled as interacting parts. Likewise, no external agents are modeled as decision-makers whose actions alter the environment. The world at this scale consists of one agent facing a fixed environment.

The environment is real in the sense that it supplies parameters (constraints, available actions, information, payoffs), but it is not an active participant. It does not respond, adapt, or change as a result of the agent’s choices. At this scale, nothing else in the world is recognized as having agency.

The ontology of Choice is therefore composed of:

• one decision-making unit, indivisible for analytical purposes
• a fixed set of constraints and opportunities, not generated by interaction
• a structured but passive environment, not a strategic or evolving one

Nothing else “exists” in the analytic sense.
This is a one-entity world.

2. Resolution of Representation (How finely or coarsely are things described?)

In the domain of Choice, the representation must be fine enough to capture the structure of a single agent’s decision problem, and coarse enough to exclude any detail that would imply the presence or influence of other agents or system-level forces. The model focuses on what the agent sees, evaluates, and chooses from—nothing more.

Only the following details are represented explicitly:

• the agent’s feasible set, describing what actions are possible
• the agent’s preferences or objective function, describing how alternatives are evaluated
• the agent’s information, describing what is known when choosing
• the agent’s constraints, describing the limits imposed by resources, technology, or context

Everything beyond this is abstracted away.
There is no representation of:

• other agents’ decisions, expectations, or beliefs
• strategic or institutional structures
• aggregate states or evolving system conditions

The environment is rendered as a set of fixed parameters, not as an object with its own internal dynamics.

Choosing too high a resolution would incorrectly introduce additional agents, strategic channels, or system adjustments.
Choosing too low a resolution would obscure the internal logic of the agent’s choice—preferences, constraints, and available actions.

The correct resolution for this domain is therefore:

• sharp enough to model the agent’s internal decision structure
• blurred enough to remove any relational or systemic detail that does not belong to a one-agent world

This is the resolution at which the problem of Choice makes sense.

3. Temporal Scale (Over what timeframe do processes unfold?)

In the domain of Choice, temporal scale concerns the speed at which a single decision-maker’s evaluations and selections unfold. Because only one agent exists at this level of analysis and no external environment responds to their actions, time functions as an internal dimension of the agent’s problem rather than an arena of interaction or system-level adjustment.

The model represents time only insofar as it shapes the agent’s own decisions:

• Instantaneous choice, when the decision is made at a single moment.
• Single-period choice, when the agent selects among alternatives with fixed consequences.
• Intertemporal or multi-period choice, when the agent’s own future states matter but nothing outside the agent evolves in response.

The temporal scale of Choice must exclude:

• feedback loops driven by other agents (Interaction)
• adjustment paths governed by system-level dynamics (Aggregation)
• long-run trajectories defined by state variables the agent does not control

Within this domain, time determines:

• what counts as relevant change (only changes internal to the agent’s own sequence of choices),
• which transitions require explicit modeling,
• which can be treated as instantaneous,
• and how the agent’s decision process unfolds relative to their constraints and information.

If time is treated too finely, the model incorrectly implies internal microdynamics or continuous psychological processes that do not belong to this scale.
If treated too coarsely, important tradeoffs across moments—delay, risk, discounting, future constraints—disappear.

The correct temporal scale for Choice is therefore:

• narrow enough to reflect the agent’s isolated decision structure,
• flexible enough to model multi-period choices arising solely from the agent’s own state transitions,
• and strictly limited to time horizons that do not introduce interdependence or system evolution.

This is the speed at which reality moves for a one-agent world.

4. Interpretive Constraints (What conclusions are allowed at this scale?)

At the scale of Choice, the only valid inferences are those that follow from the internal structure of a single decision-maker’s problem. Because no other agents exist at this scale and no system-level variables evolve in response to the agent’s actions, the scope of legitimate conclusions is intentionally narrow. Any inference that relies on influence, coordination, feedback, propagation, or collective behavior lies outside this domain.

Permitted conclusions include statements about:

• how the agent evaluates alternatives given fixed information
• how constraints shape the agent’s feasible actions
• how preferences or objectives determine selection
• how the agent trades off options across time when the dynamics arise solely from the agent’s own internal state transitions
• how uncertainty affects the agent’s choices when it does not emerge from others’ actions or systemic evolution

Forbidden conclusions include statements that imply:

• reactions by other decision-makers
• strategic feedback, coordination, conflict, or cooperation
• institutions or rules that allocate outcomes across multiple agents
• adjustments or dynamics driven by changes at the system level
• aggregate effects or population-level patterns
• propagation of shocks or disturbances through a larger structure

These constraints are not aesthetic—they ensure logical coherence.
At this scale:

• no individual choice can be explained by referencing other agents, because no others exist
• no adjustment path can be invoked, because the environment does not change in response to the agent
• no collective phenomenon can be inferred, because the ontology does not contain collectives
• no system-level causation can be imported, because the scale excludes systems

Violating these constraints produces category errors—analyses that implicitly rely on Interaction or Aggregation while claiming to occupy the domain of Choice.

The interpretive rule is simple and strict:

Only conclusions that emerge from one agent optimizing within a fixed environment are valid at this scale.

Everything else belongs to another domain.

5. Canonical Scale Statement

The domain of Choice operates at the scale of a single decision-making agent whose environment is fixed and unresponsive to their actions. At this level, representation is limited to the agent’s own preferences, constraints, feasible actions, and information, with no additional detail concerning other agents or system-level variables. Time is modeled only insofar as it structures the agent’s internal sequence of decisions; no temporal adjustments arise from outside the agent. Valid inferences are restricted to conclusions that follow from the agent’s solitary optimization problem, excluding any reasoning that relies on cross-agent influence, institutional structures, or dynamic system behavior. This scale is the lawful operating level of Choice: a one-agent world defined entirely by the structure of its decision problem and insulated from the forces that govern Interaction or Aggregation.

---

Science Analysis Template

These are the structural patterns found across all Scientific Disciplines

1. Choice has its own scale ladder inside the social sciences.

Across disciplines, each science occupies a band of size, time, and organizational complexity. For Choice, the relevant ladders are:

• Physical size / abstraction:
  neural/biological mechanisms → psychological states → decision problem of a single agent → dyads/groups → organizations → institutions → systems.
• Economic organization:
  individual decision problem → strategic interaction (games, markets) → sectors/industries → whole economies.
• Social structure:
  person → group → organization → community → state → global order.

Choice lives exactly at the “single decision problem of a single agent” rung on all of these.

• It does not descend into the neural or biological implementation of choice (that is neuroscience, cognitive science).
• It does not climb up into group, market, or institutional behavior (that is Interaction and Aggregation).

At this scale, the only object that exists analytically is:

• one decision-making unit,
• facing a fixed menu of options, constraints, and information.

That is the ontological rung of Choice.

2. Scale transitions define the handoffs out of Choice.

Once you know where Choice sits on the ladder, you can specify precise exit conditions—the points where the problem must be handed off to another domain because the scale has changed.

There are three main transition directions:

1. Downward (into sub-personal detail).
   • If the analysis must explain how the agent forms preferences, processes information, or generates actions via underlying biological or cognitive mechanisms, the scale shifts downward to psychology, neuroscience, or cognitive science.
   • Choice does not model neurons, attention mechanisms, or subconscious processes; it assumes a coherent decision function.
2. Sideways / upward into Interaction.
   • The moment another agent’s decision must be represented explicitly—because that decision changes payoffs, constraints, or information—the problem leaves Choice.
   • As soon as “what I do depends on what you do” becomes central, the scale has transitioned to Interaction (games, markets, bargaining, institutions).
3. Upward into Aggregation & Dynamics.
   • If the relevant variables become aggregates or macro-states (e.g., population behavior, systemic risk, macroeconomic conditions), the scale has moved to Aggregation & Dynamics.
   • Individual decision problems can still exist in the background, but they no longer define the level of analysis.

These transitions are scale rules, not stylistic choices.
If you violate them—by talking about strategic responses or macro dynamics while claiming to be in Choice—you are simply in the wrong domain.

3. Choice’s time scale is the agent’s decision horizon, and nothing beyond that.

The general template insists that time scale be treated as precisely as spatial/organizational scale. For Choice, the temporal band is:

• Instant decisions:
  single-moment choices (e.g., “choose x or y now”) where time only indexes the moment of selection.
• Single-period problems:
  choices made at time t with consequences evaluated over a fixed horizon that is not itself modeled as evolving in response to the agent.
• Intertemporal decisions internal to one agent:
  savings vs. consumption over a lifetime, career planning, durable good purchase timing—so long as:
  • the “future” is part of the agent’s own problem, and
  • no other agents or system variables are evolving in a way that must be modeled explicitly.

Choice does not operate at time scales where:

• other agents are reacting and updating strategies (that requires Interaction’s temporal scale), or
• system-level variables are adjusting and propagating (that requires Aggregation’s dynamic scale).

In short:

• Time in Choice is the agent’s decision horizon, not the horizon of a market or a system.
• When the relevant “change” is anything other than the agent’s own planned sequence of actions, you have left the temporal window of Choice.

4. Choice uses formal tools that are scale-independent but anchored to the single-agent level.

The general template notes that formal sciences (logic, math) are not tied to physical scale but to abstraction scales. Choice leans heavily on these:

• preference relations, utility functions, objective functions;
• constraint sets, feasible regions;
• optimization operators, dynamic programming, Bellman equations.

These tools are scale-independent math, but in Choice they are anchored to a very specific abstraction scale:

• a single, unified decision function mapping states and constraints into actions.

This has consequences:

• The theorems (existence, uniqueness, optimality) only hold because the problem is posed at the single-agent level.
• As soon as you introduce other agents as strategic actors or system-level feedback, those same formal statements no longer describe the right object—you need game theory or dynamic system tools instead.

So Choice is not a formal science, but it is:

• a formalized empirical domain whose use of mathematics is locked to the single-agent abstraction scale,
• structurally incompatible with multi-agent or system-level applications of the same math unless those are explicitly redefined in the higher domains.

5. Choice’s organizational level distinguishes it from both biology and higher social sciences.

The template separates biology and social sciences by organizational level. Choice fits into this hierarchy as follows:

• Below Choice:
  • Neuroscience, physiology, and cognitive psychology explain how brains and bodies implement decision processes.
  • Their units are cells, circuits, modules—not economic agents.
• Choice itself:
  • The unit is the agent as a coherent chooser: a person, household, firm, or other entity treated as a single decision-making unit.
  • Internal complexity and external relations are suppressed; what matters is the structure of the decision problem.
• Above Choice:
  • Interaction: groups, markets, organizations, networks where multiple agents matter simultaneously.
  • Aggregation & Dynamics: communities, states, global systems where macro patterns, not individual choice structures, dominate.

This clarifies:

• why psychology and neuroscience can study “the same person” as Choice but at a different scale;
• why Interaction and Aggregation can also study “the same population” as Choice but at higher organizational levels;
• and why there is no contradiction—each works on a different rung of the ladder.

Choice’s scale is therefore organizationally minimal within the social sciences.

6. The full result: Choice has a sharply defined operational zone in the multi-scale map of science.

With the Scale template applied, Choice is pinned down along four axes:

• Size / organizational range:
  exactly one decision-making unit; no groups, no systems.
• Time range:
  the agent’s own decision horizon—instantaneous, single-period, or intertemporal—provided the dynamics are internal to the agent.
• Complexity level:
  just enough structure to define preferences, constraints, information, and feasible actions; no strategic networks, no macro-state equations.
• Handoff boundaries:
  • to Interaction at the first appearance of another modeled agent;
  • to Aggregation & Dynamics at the first appearance of macro-state variables as determinants of outcomes;
  • to sub-personal sciences at the first attempt to model implementation mechanisms inside the agent.

In the layered atlas of the sciences, this makes Choice:

• the atomic unit of decision theory,
• the anchor point beneath all higher social and economic structures,
• and the precise location where you must stop if you want to speak strictly about isolated optimization.