Skip to content
Tutorials Center
My learning

Topics

Lesson 35 of 39 · Entanglement, Measurement and Decoherence

The Measurement Problem and Interpretations

Understand the quantum measurement problem, simulate projective measurements and the quantum Zeno effect, and compare the main interpretations fairly.

Quantum mechanics predicts the statistics of experiments with spectacular precision. Yet there is a question about what the theory says is happening that physicists still debate. It is called the measurement problem, and this lesson states it carefully, shows the measurement rules at work in simulations, and gives a fair map of the main interpretations. None of them is established by experiment as the one right answer, and this lesson does not claim one.

Two rules that do not fit together

Textbook quantum mechanics uses two kinds of dynamics.

  1. Unitary evolution. Between measurements the state obeys the Schrödinger equation iħ ∂ψ/∂t = Hψ. It is deterministic, linear and reversible.
  2. Projective measurement (the Born rule and collapse). When an observable A with eigenstates |a_i⟩ is measured on state |ψ⟩, the result is a_i with probability p_i = |⟨a_i|ψ⟩|², and afterwards the state is |a_i⟩. This is random, nonlinear and irreversible.

The measurement problem is that measuring devices are made of atoms, so they should also obey rule 1. If they do, then a spin-1/2 particle in (|↑⟩ + |↓⟩)/√2 that interacts with a detector evolves, by linearity, into

Text
(|↑⟩ + |↓⟩)/√2 ⊗ |ready⟩  →  (|↑⟩|"up"⟩ + |↓⟩|"down"⟩)/√2

an entangled superposition of two outcomes. But we always see one outcome. When and how does the superposition become a single result? Rule 2 does not say what counts as a "measurement" or why it is different from other interactions. This is the problem: one theory with two incompatible rules, and no precise criterion for when to use which.

Do not confuse the measurement problem with the claim that "consciousness causes collapse" or that observation needs a human. Most physicists who work on the problem do not hold that view. Detectors, photographs and any record that can influence the world are enough for the empirical rules to apply.

The rules in action

Two simulations of projective measurements on a spin-1/2 particle show the structure of rule 2. The first reproduces the "sequential Stern-Gerlach" experiment: measure σ_z, then σ_x, then σ_z again. After the first σ_z measurement a result of + is certain to be repeated by an immediate second σ_z measurement. But a σ_x measurement in between destroys that memory, since σ_x and σ_z do not commute.

Python Sequential spin measurements: z, then x, then z again

Press Run (or Ctrl + Enter) to see the result here.

Program input · what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The quantum Zeno effect

Rule 2 has a striking consequence. Imagine a two-level system that is driven so that, left alone, its state vector rotates from |0⟩ to |1⟩ through a total angle of π/2 (a complete flip). Now measure whether it is still in |0⟩ after each of N equal steps. In each step the vector turns by π/(2N), so the probability of being found in |0⟩ is cos²(π/(2N)), and after a successful measurement the state is projected back onto |0⟩. Over all N measurements the survival probability is

Text
P_survive = [cos²(π/(2N))]^N  →  1   as N → ∞

Frequent observation freezes the evolution: the "watched pot never boils". The effect has been observed with trapped ions, cold atoms and optical systems. Every interpretation below reproduces it, though they describe it differently (for example as decoherence caused by the repeated interaction with the apparatus).

Python Quantum Zeno effect: survival probability versus number of measurements

Press Run (or Ctrl + Enter) to see the result here.

Program input · what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

(For N = 1 the single measurement happens after the full rotation, and the survival probability is cos²(π/2) = 0.)

A fair map of the interpretations

Every approach below agrees with standard quantum mechanics on all experiments done so far, except that objective-collapse theories predict small deviations that experiments try to find. The choices differ in what they say exists and how they treat the two rules.

Copenhagen and related "operational" views. Quantum states encode what we can predict about experiments, and measurement outcomes in classical terms are primitive. Bohr, Heisenberg and their followers differ on details. The strength is practical clarity; the weakness is that the boundary between the quantum system and the classical measuring apparatus is left vague.

Many-worlds (Everett). There is no collapse. Rule 1 holds always, so the entangled superposition above is real: the observer, too, is in a superposition, and each branch contains an observer who sees one definite result. Strengths: the theory is simple and uses only the Schrödinger equation. Open problems: what the probabilities mean when all outcomes occur, and why the Born rule |ψ|² is the right weight. Many-worlds supporters argue that decoherence picks out the branches; critics find the derivation of probabilities unsatisfying.

Bohmian mechanics (de Broglie-Bohm, pilot wave). Particles have definite positions at all times, guided by the wave function through a deterministic equation. The Born rule follows from a statistical assumption about initial conditions. The theory is explicitly nonlocal, in line with Bell's theorem, and it is hard to extend to relativistic quantum field theory in a natural way.

Objective collapse (GRW, CSL, Penrose proposals). Rule 1 is modified: a small, random, nonlinear term makes superpositions of large objects collapse spontaneously, while leaving small systems alone. This is the only family that gives different predictions, and experiments with large molecules, optomechanics and spontaneous radiation emission have been putting limits on the collapse parameters. So far no deviation from standard quantum mechanics has been found.

Epistemic and relational views (QBism, relational quantum mechanics). The quantum state expresses an agent's degrees of belief, or a relation between systems, rather than an objective property of one system. They dissolve the problem by redefining what the state is, which others consider a high price.

What most physicists do. In daily work the practical recipe (unitary evolution plus Born rule, with decoherence explaining why interference disappears) is used without needing to decide among these views. Surveys of physicists show a wide spread of opinions; there is no consensus, and no experiment currently distinguishes the unmodified interpretations from each other.

Check your understanding

Quick check · 4 questions

of 4 answered You scored / 4 · perfect!

  1. 1. What is the measurement problem?

    Rule 1 (linear, deterministic) and rule 2 (random collapse) seem to conflict, and the theory gives no sharp definition of a 'measurement'.
  2. 2. In the sequential experiment z, x, z, the probability of the third result being + (given the first was +) is about:

    After the σ_x measurement the state is an eigenstate of σ_x, for which σ_z is equally likely to give + or -.
  3. 3. In the quantum Zeno effect with N measurements, the survival probability [cos²(π/2N)]^N tends to:

    For large N, cos²(π/2N) ≈ 1 - (π/2N)², so the product is about 1 - π²/(4N), which approaches 1.
  4. 4. Which interpretation family predicts deviations from standard quantum mechanics that experiments can test?

    Collapse models add a small nonlinear stochastic term, so they differ from standard theory in principle; many-worlds and Bohmian mechanics reproduce the standard predictions.

Summary

  • Quantum mechanics uses unitary evolution plus projective measurement with the Born rule, and does not define sharply when to switch rules: the measurement problem.
  • Sequential spin measurements show that noncommuting observables disturb each other; repeated measurements of the same observable are repeatable.
  • Frequent measurements can freeze evolution (quantum Zeno effect).
  • Main interpretations: Copenhagen/operational, many-worlds, Bohmian mechanics, objective collapse, and epistemic/relational views. Only collapse models differ in predictions.
  • There is no experimental consensus on which interpretation, if any, is right.

Next: Decoherence and the Quantum-to-Classical Transition