quantum algorithms:
In its standard formulation, quantum physics is a probabilistic theory of nature, since it provides only probabilities (in fact probability amplitudes) for values of observable in experiment. This is in contrast to classical mechanics, which (on classical pure states) predicts events with certainty.
However, on very small scales classical mechanics predicts the wrong events, while quantum mechanics and quantum field theory predicts the right probabilities.
On the other hand, it is well known that upon “coarse graining”, which means after averaging over certain details, also classical mechanics induces a probabilistic theory of nature, namely statistical mechanics/thermodynamics.
Therefore it is natural to speculate that maybe also quantum physics is the statistical mechanics of a more refined theory of as yet unseen “degrees of freedom of nature” on small scales, which does predict events with certainty on very small scales, and which reduces to quantum mechanics on larger scales as one averages over these unseen new degrees of freedom.
These “unseen degrees of freedom” are usually called hidden variables, and a theory which is non-probabilistic but designed to reproduce quantum mechanics as its statistical coarse grained theory are called hidden variable theories. (These are hence one potential interpretation of quantum mechanics.)
There have been various attempts to construct such hidden variable theories. However, there are also theorems about the characteristic properties of quantum mechanics which assert (under some (natural) assumptions, of course) that there cannot be a hidden variable theory.
These theorems are
One well-developed attempt to construct a hidden variable theory is Bohmian mechanics; this makes hidden variables out of the entire wavefunction and violates the assumption of locality.
Barandes 2026 argues that the deBroglie-Bohm pilot wave theory is best understood as a hidden Markov stochastic process model, and that the configuration-space wave function, which serves as the pilot wave for the theory, is optimally interpreted as consisting of a set of latent variables for that hidden Markov model. Hidden Markov models are just a formal way of representing stochastic processes whose dynamics are non-Markovian as Markov processes by adding latent variables (hidden variables).
Hidden variables (and their non-existence) are already mentioned in:
p. 172 in: Mathematische Grundlagen der Quantenmechanik, Springer (1932, 1971) [doi:10.1007/978-3-642-96048-2]
p. 198 in: Mathematical Foundations of Quantum Mechanics Princeton University Press (1955) [doi:10.1515/9781400889921, Wikipedia entry]
Textbook account:
See also:
The original article on Bell's theorem:
The original article on the Kochen-Specker theorem:
Discussion of hidden variable theories in terms of quantum logic:
The view of a pilot wave theory as a hidden Markov model:
Last revised on March 4, 2026 at 16:28:50. See the history of this page for a list of all contributions to it.