Superdeterministic Approaches

Superdeterministic approaches seek an underlying explanation of quantum statistics in which hidden variables and measurement settings are statistically dependent. This changes a premise of Bell’s theorem. Determinism alone does not imply that dependence, and writing down a setting-dependent distribution is not yet a complete predictive theory.

Required background. Born rule: Translate a specified state and measurement into outcome probabilities.

Measurement independence is a statistical condition

In a Bell experiment, let $\lambda$ describe hidden variables and $x,y$ the settings. Measurement independence is

$$\mu(\lambda\mid x,y)=\mu(\lambda).$$

A model can retain local response functions while replacing this by a setting-dependent measure:

$$p(a,b\mid x,y)= \int d\lambda\,\mu(\lambda\mid x,y) p(a\mid x,\lambda)p(b\mid y,\lambda).$$

Bell’s usual common-measure derivation no longer goes through. This is an evasion by changing assumptions, not a mathematical counterexample to the theorem.

Why determinism is not enough

A deterministic universe can still have ensembles in which settings and relevant hidden variables are statistically independent. Conversely, stochastic models can violate that independence. The decisive issue is the joint distribution and its physical origin, not whether human choices are metaphysically free.

A common past cause can correlate settings and system variables. A scientific model must quantify those correlations and explain why the experimental randomizers, preparation procedures, and records have the observed behavior.

A deliberately unexplanatory construction

For each setting pair, let $\lambda=(a,b)$ have distribution $\mu(\lambda\mid x,y)=p_{\rm target}(a,b\mid x,y)$. Local response functions read out the stored components. This reproduces any chosen target distribution, including quantum data.

The construction demonstrates how unconstrained setting dependence weakens a no-go theorem. It explains almost nothing: the desired answer has been placed directly in the hidden distribution. A productive theory needs restrictions that determine distributions across new experiments, rather than a separate fit for each case.

Experimental accountability

Changing setting generators, timing, and source separations tests broad classes of ordinary correlations and closes particular loopholes. Such tests do not logically exclude every conceivable model with arbitrary global setting dependence.

That logical freedom does not make all superdeterministic proposals untestable. A concrete dynamics may predict deviations, memory effects, or specific correlation scales. Those predictions must be stated before examining the data and compared with ordinary systematic alternatives.

Locality, probability, and classicality

A proposal may aim for local underlying dynamics, but it must still reproduce quantum no-signaling, state preparation, interference, and measurement statistics. Explaining Bell violations alone does not resolve the measurement problem or derive Born probabilities.

Its appeal is retaining some classical explanatory commitments by revising independence assumptions. Its costs include justifying the required distributions, avoiding arbitrary adjustment, and supplying an intervention account that remains useful for experiments. Criticisms should identify those costs without relying only on the word “conspiracy.”

Exercises

Explain why separate distributions $\mu_{xy}$ prevent directly averaging a pointwise CHSH bound over one common ensemble.

Solution

The standard bound combines four correlation terms evaluated under the same $\mu(\lambda)$. With different $\mu_{xy}$, the four averages weight hidden states differently, so they cannot be combined into one integral of the usual bounded expression without additional constraints. The missing common measure is the exact mathematical issue.

References

• M. J. W. Hall, “Local deterministic model of singlet state correlations based on relaxing measurement independence,” Physical Review Letters 105, 250404 (2010), doi:10.1103/PhysRevLett.105.250404.
• S. Hossenfelder and T. N. Palmer, “Rethinking superdeterminism,” Frontiers in Physics 8, 139 (2020), doi:10.3389/fphy.2020.00139.