Quantum Decision Theory

The Quantum Architecture of Choice

From neurons to Hilbert spaces to the edge of spacetime, a map of every framework that claims to explain why you do what you do.

Begin the Descent

Decision Space Visualizer

Click the field below. Three competing possibilities exist in superposition, until observation collapses them into a single outcome.

|A⟩ |B⟩ |A+B⟩
Collapsed
The wave function resolved
Click to collapse the superposition

I. Classical Behavior

Before quantum cognition, there was the ABC model, the bedrock of behavioral psychology. Every action is a chain: Antecedent triggers Behavior, which produces a Consequence that feeds back into the loop.

Operant Conditioning

Skinner's framework: reinforcement strengthens behavior, punishment weakens it. The environment sculpts the agent. No internal states needed, just stimulus and response frequencies.

P(B|S) = f(reinforcement_history)

Bounded Rationality

Simon's insight: humans don't maximize utility, they satisfice. The agent uses heuristics, truncates search trees, and settles for "good enough." Rational, but within limits.

U(x) ≈ argmax{U | cost(search) < threshold}

Prospect Theory

Kahneman and Tversky's Nobel contribution: losses loom larger than gains. The value function is concave above a reference point and convex below it. Decisions are relative, not absolute.

V(x) = x^α if x ≥ 0, -λ(-x)^β if x < 0

These frameworks capture enormous predictive power. But they all share an assumption: that the decision state is definite before it is measured. Quantum cognition challenges exactly this.

II. The Quantum Mind

In the 2000s, researchers noticed something odd: human judgment violates classical probability. Conjunction fallacies, order effects, disjunction effects, patterns that cannot arise from any classical probability distribution. But they arise naturally from quantum probability.

Hilbert Space Model of Cognition

Instead of placing beliefs on a classical probability simplex, map them onto a Hilbert space. A person's cognitive state is a state vector |ψ⟩. A question is a projector. The probability of a "yes" answer is:

P(yes) = ⟨ψ| P_yes |ψ⟩ = ||P_yes |ψ⟩||²

Crucially, projectors don't commute: P_A P_B ≠ P_B P_A. The order in which you ask questions changes the answer: exactly as observed in human surveys.

POVMs

Positive Operator-Valued Measures generalize projectors. They allow "fuzzy" measurements, partial observations that don't fully collapse the state. Perfect for modeling ambiguous social judgments.

Interference

The Linda Problem's conjunction fallacy: P(feminist & bank teller) > P(bank teller). Classically impossible. But in quantum probability, the cross-term interference between "feminist" and "bank teller" amplitudes adds probability mass to the conjunction.

Framework Comparator

See the same decision, "Should I take this job offer?", through five lenses.

Behavioral View

The decision is a function of reinforcement history. Past job switches that led to positive outcomes (higher salary, praise) increased the probability of "accept." Past negative outcomes (social disruption, regret) decreased it. The agent's internal states are a black box.

P(accept) = Σ w_i · R_i(past_switches)

Cognitive View

The agent builds a mental model: a decision tree with nodes for salary, culture, commute, growth. Each branch has a subjective utility weighted by estimated probability. Bounded rationality means the tree is pruned. System 1 (fast heuristics) pre-filters; System 2 (slow deliberation) evaluates the shortlist.

U(accept) = Σ p_i · u(outcome_i) - cost(deliberation)

Quantum View

Before deliberation, the agent is in a superposition of "accept" and "reject." Thinking about salary projects the state onto one basis; thinking about family projects it onto another. These bases are incompatible: they don't share eigenstates. The order of consideration creates interference that biases the final measurement (the decision).

|ψ⟩ = α|accept⟩ + β|reject⟩, |α|² + |β|² = 1

Algorithmic View (AIXI)

The optimal agent compresses all past observations into the shortest program that predicts them (Kolmogorov complexity), then acts to maximize expected future reward over all computable environments, weighted by their algorithmic probability. The job decision is a single step in an infinite-horizon optimization.

a* = argmax_a Σ_e 2^(-K(e)) · V(a, e)

Pre-Geometric View

There is no "agent" and no "job." Both are patterns in the Ruliad, the entangled limit of all possible computations. What appears as a "decision" is a particular foliation of rulial space, a cross-section where observer and environment momentarily crystallize. Constructor theory asks: which transformations are possible, regardless of dynamics?

{agent, environment} ⊂ Ruliad / ~observer

III. Universal Agency

What if we could build the mathematically optimal agent? Marcus Hutter's AIXI is exactly that, a formal definition of perfect intelligence that is also perfectly uncomputable.

AIXI: The Platonic Agent

AIXI combines Solomonoff induction (the optimal predictor) with sequential decision theory (the optimal actor). At each time step, it:

1. Enumerates all computable hypotheses about the environment.
2. Weights each by 2-K(h): shorter programs are more probable.
3. Selects the action maximizing expected lifetime reward across all hypotheses.

a*_t = argmax_{a_t} Σ_{e_t} ... Σ_{e_m} [Σ_{k} 2^{-K(q_k)} · Σ_{i=t}^{m} r_i]

AIXI is the ceiling of intelligence: no agent in any computable environment can do better. Its uncomputability is the price of universality.

Kolmogorov Complexity

The length of the shortest program that produces a string. It's the mathematical formalization of "simplicity", and it's uncomputable. We can only approximate it, which is what compression algorithms (and human intuition) do.

Solomonoff Induction

The optimal way to predict the next bit of a sequence: weight every consistent program by its algorithmic probability. It converges on the true distribution faster than any other predictor. It's also uncomputable, a recurring theme.

Computable Approximations

AIXItl truncates the horizon and limits program length. MC-AIXI uses Monte Carlo tree search. These lose optimality guarantees but become runnable. Modern LLMs can be seen as massively parallel, approximate Solomonoff predictors.

IV. Beyond Spacetime

The deepest frameworks dissolve the boundary between agent and environment entirely. Here, "choice" is not something an agent does: it's a pattern that the universe exhibits.

Layers of Agency

Click each layer to descend from familiar neuroscience into increasingly alien territory.

1
Neurons & Circuits
Spike trains, synaptic weights, dopamine prediction errors. The hardware. Billions of neurons firing in patterns that somehow produce the sensation of choosing. At this level, "decision" means a population of neurons reaching a threshold.
2
Psychology & Heuristics
Prospect theory, anchoring, framing effects. The software running on the neural hardware. Useful, predictive, but incomplete, these models describe deviations from rationality without explaining why the deviations have the specific structure they do.
3
Quantum Cognition
Hilbert spaces, non-commuting observables, interference. Not "quantum woo", a rigorous mathematical framework that explains the structure of cognitive biases. Order effects and conjunction fallacies fall out naturally from the geometry of the state space.
4
Algorithmic Agents
AIXI, Kolmogorov complexity, Solomonoff induction. Here the agent is defined purely by its computational relationship to its environment. No neurons, no psychology, just a program that compresses observations and maximizes reward. Intelligence as compression.
5
Holographic Screens
The holographic principle: all information in a volume of space is encoded on its boundary. Markov blankets formalize the boundary between agent and environment. The "agent" is whatever is inside the blanket; the "world" is outside. The blanket is the interface, perception and action are boundary conditions, not internal states.
6
The Ruliad
Wolfram's Ruliad: the entangled limit of all possible computations. Every possible universe, every possible observer, every possible decision, all woven into a single irreducible structure. What we call "physics" is a particular sampling of the Ruliad by an observer with bounded computational resources. Agency itself dissolves into rulial geometry.

Animated Markov Blanket

The boundary between self and world. Sensory states flow in; active states flow out. Internal states are shielded from the environment, they can only "see" the world through the blanket.

ENVIRONMENT MARKOV BLANKET INTERNAL states S S A A sensory active "The boundary is the agent.", Karl Friston

Constructor Theory

David Deutsch's framework shifts physics from dynamics to possibility. Instead of asking "what happens next?", it asks "which transformations are possible and which are impossible?" An agent is a constructor: a system that can cause a transformation and retain the ability to cause it again.

The Ruliad

Stephen Wolfram's ultimate structure: the entangled limit of all possible computations applied to all possible initial conditions. Physics, mathematics, and consciousness are different samplings of this single object. Observer-reality boundaries are foliations of rulial space.

Holographic Principle

All information in a 3D volume can be described by data on its 2D boundary. The Markov blanket of an agent is the agent, a holographic screen encoding everything the internal states need. Consciousness may be a boundary phenomenon.

V. What Kind of Agent Are You?

Five questions. Five agent types. Where do you fall on the spectrum from classical reinforcement learner to non-dual pattern in the Ruliad?

1. When making a big decision, you typically:

List pros and cons, pick the winner
Build a mental model and simulate outcomes
Notice the answer depends on what you think about first
Try to find the simplest explanation that predicts everything
Realize the "decision" and the "decider" are the same process

2. What is consciousness?

An emergent behavior from neural reinforcement patterns
An internal world-model that includes a model of itself
A measurement process that collapses cognitive superpositions
The shortest program that predicts its own output
A particular cross-section of the Ruliad

3. Free will is:

An illusion, you are your reinforcement history
Compatibilist, your model-based deliberation is "you"
Undetermined until measured, genuine indeterminacy exists
A computational concept: the agent's output is uncomputable from outside
A category error, there's no separate "will" to be free

4. The relationship between mind and reality is:

Reality shapes behavior through consequences
The mind builds models of reality; they interact
Mind and reality are complementary observables
Reality is a hypothesis weighted by its complexity
There is no boundary, both are patterns in the same structure

5. What would change your mind about everything?

Overwhelming counter-evidence from experiments
A more elegant model that explains more with less
Discovering that measurement order changes physical reality
A proof that my current framework has higher Kolmogorov complexity than an alternative
Seeing that "my mind" and "everything" were never separate

Share your result, the architecture of choice starts with understanding your own.