What Is Decoherence, and Does It Solve the Measurement Problem?

Quantum mechanics lets a system sit in a superposition of alternatives: an electron spin that is both “up” and “down” in a definite, measurable combination. Yet tables, cats and lab pointers never look like that. The standard modern explanation for why is decoherence. It is real, it is measured, and it is often oversold. Our character Decoherex is the part of the story that is solid; Collapsea is the part that is still argued over.

What is decoherence, in one paragraph?

No real object is perfectly isolated. Air molecules bounce off it, thermal photons scatter from it, and its own internal parts interact with each other. Each such interaction can carry away a little information about the object’s state. When the environment “records” which alternative the system is in, the system and environment become entangled, and the interference between the alternatives is no longer visible in measurements made on the system alone. Wojciech Zurek’s review summarises it directly: decoherence is caused by interaction with the environment, which monitors certain observables and destroys interference between the corresponding “pointer states” (Zurek 2003, Rev. Mod. Phys.^[1]).

What does that look like in the maths?

Take a two-state system in the superposition |ψ⟩ = a|0⟩ + b|1⟩, where a and b are complex amplitudes with |a|² + |b|² = 1. Its density matrix ρ = |ψ⟩⟨ψ| has diagonal entries |a|² and |b|² (the probabilities) and off-diagonal entries ab* and a*b (the “coherences” that produce interference).

Now let the system interact with an environment that starts in state |E⟩ and ends up correlated with it:

(a|0⟩ + b|1⟩)|E⟩  →  a|0⟩|E₀⟩ + b|1⟩|E₁⟩

If we look only at the system (mathematically, take the partial trace over the environment), the off-diagonal entries are multiplied by the overlap ⟨E₁|E₀⟩. When the two environment states are nearly distinguishable, that overlap is nearly zero, and the system’s density matrix looks almost diagonal: like a list of probabilities, with the interference terms gone from local view. Nothing has been deleted; the coherence has spread into correlations with the environment, where it is practically impossible to recover.

How fast does it happen?

For large objects, absurdly fast. The Stanford Encyclopedia of Philosophy gives a standard illustration: a speck of dust of radius 10⁻⁵ cm floating in air has interference between spatially separated components suppressed down to a width of about 10⁻¹³ cm after about a microsecond of exposure to air (SEP, Decoherence^[2]). That is why superpositions of everyday things are not just hard to see but effectively unseeable. For a single, well-shielded atom or photon the timescale can be long enough to do experiments, which is exactly what quantum-optics labs exploit.

Has anyone actually watched decoherence happen?

Yes. In 1996 Michel Brune, Serge Haroche and colleagues prepared a microwave field in a high-Q cavity in a superposition of two classically distinct phases — a small “Schrödinger cat” built from a few photons — using Rydberg atoms passing one at a time. They then watched the superposition turn into a statistical mixture as it unfolded, which their paper describes as direct insight into a process at the heart of quantum measurement (Brune et al. 1996, PRL^[3]). Haroche shared the 2012 Nobel Prize in Physics with David Wineland “for ground-breaking experimental methods that enable measuring and manipulation of individual quantum systems” (NobelPrize.org 2012^[4]). The Nobel press release states the practical problem plainly: single particles lose their quantum properties as soon as they interact with the outside world (Nobel 2012 press release^[5]).

In 2003 Klaus Hornberger, Markus Arndt, Anton Zeilinger and colleagues sent fullerene molecules through a matter-wave interferometer and raised the pressure of the background gas step by step. Interference fringes faded as the gas pressure rose, quantitatively matching decoherence theory for collisions (Hornberger et al. 2003^[6]). This is decoherence with a dial on it: more collisions, more “which-path” information leaked, less interference.

Who came up with the idea?

Suppression of interference appears in many early papers, but the SEP credits H.-D. Zeh with laying the modern foundation of decoherence as a subject in the early 1970s (Zeh 1970, Found. Phys. 1, 69; Crossref DOI 10.1007/BF00708656), with Zurek’s papers of the early 1980s equally influential (SEP^[2]). Zurek’s 2003 review introduced many readers to “einselection” — environment-induced superselection — the idea that the environment picks out which states of a system are stable enough to look classical (Zurek 2003^[1]).

So does decoherence solve the measurement problem?

Not by itself. This is the most common overstatement, and the primary literature is clear about it. The measurement problem has two parts: why we never see a superposition of results, and why a measurement yields a definite single result at all. Decoherence answers the first. It does not answer the second. The SEP entry puts it exactly: decoherence explains why we do not observe superpositions of measurement results, but it does not explain why we observe measurement results in the first place; decoherence as such does not provide a solution to the measurement problem unless it is combined with a foundational approach such as Bohm, Everett, GRW collapse models, or a Copenhagen-style view (SEP §2.1^[2]).

The reason is visible in the equation above. After decoherence, the full state of system plus environment is still a superposition, a|0⟩|E₀⟩ + b|1⟩|E₁⟩. A diagonal reduced density matrix looks like “either 0 or 1 with these probabilities”, but it was obtained by ignoring the environment, not by one outcome actually happening. Maximilian Schlosshauer’s review describes the implications for foundations as a matter of great controversy and works through how each interpretation uses decoherence differently (Schlosshauer 2005^[7]). That question is open. We label it open.

What decoherence is — and is not

• Is: ordinary Schrödinger evolution of system plus environment; experimentally observed; the main engineering enemy of quantum computers; the reason everyday objects look classical.
• Is not: a conscious observer effect; a new law that replaces quantum mechanics; a proof that wavefunction “collapse” is physical; a settled answer to which interpretation is right.

For the full measurement story, from the Born rule to the interpretations, read our pillar What Is Quantum Measurement, Really?. For the question people usually search first, see Does Observing a Particle Really Change It?.

Where this lives in the shop

Decoherex is the leak: the environment quietly carrying away which-path information. Collapsea stands for the definite outcome that decoherence alone does not deliver, and Schrödy is the superposition that started the argument. They sit together in Quantum Physics T-Shirts.

FAQ

Is decoherence the same as wavefunction collapse?

No. Decoherence follows from the standard Schrödinger equation applied to system plus environment. Collapse, if it is physical, would be something extra. Decoherence makes the world look as if collapse happened, without settling whether it did.

Does a person have to look for decoherence to happen?

No. Any interaction that records which-alternative information in the environment does it: a stray photon, a gas molecule, a vibration in the apparatus.

Why does decoherence matter for quantum computers?

Qubits must keep their superpositions and entanglement long enough to compute. Every unwanted interaction with the surroundings leaks information and destroys interference, which is why quantum hardware is shielded, cooled and error-corrected.

References

1. W. H. Zurek (2003), “Decoherence, einselection, and the quantum origins of the classical,” Rev. Mod. Phys. 75, 715 (arXiv:quant-ph/0105127; DOI 10.1103/RevModPhys.75.715). https://arxiv.org/abs/quant-ph/0105127
2. G. Bacciagaluppi, “The Role of Decoherence in Quantum Mechanics,” Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/entries/qm-decoherence/
3. M. Brune, E. Hagley, J. Dreyer, X. Maître, A. Maali, C. Wunderlich, J. M. Raimond & S. Haroche (1996), “Observing the Progressive Decoherence of the ‘Meter’ in a Quantum Measurement,” Phys. Rev. Lett. 77, 4887. https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.77.4887
4. NobelPrize.org, The Nobel Prize in Physics 2012 (Haroche, Wineland). https://www.nobelprize.org/prizes/physics/2012/summary/
5. NobelPrize.org, Press release: The Nobel Prize in Physics 2012. https://www.nobelprize.org/prizes/physics/2012/press-release/
6. K. Hornberger, S. Uttenthaler, B. Brezger, L. Hackermüller, M. Arndt & A. Zeilinger (2003), “Collisional decoherence observed in matter wave interferometry,” Phys. Rev. Lett. 90, 160401 (arXiv:quant-ph/0303093). https://arxiv.org/abs/quant-ph/0303093
7. M. Schlosshauer (2005), “Decoherence, the measurement problem, and interpretations of quantum mechanics,” Rev. Mod. Phys. 76, 1267 (arXiv:quant-ph/0312059; DOI 10.1103/RevModPhys.76.1267). https://arxiv.org/abs/quant-ph/0312059

Written by Pixelated Physics. Every factual claim is linked to the source we checked; points that are still debated are labelled open.