---
source: 'OpenContext: Open problems in generalized contextuality'
url: 'https://open-context.jan-bulling.com/problems/noncontextual-switch'
number: 2
title: Noncontextual SWITCH
status: open
statusSince: '2026-10-03'
published: true
posedBy:
  - Rafael Wagner and John H. Selby
posed: '2026-10-03'
updated: '2026-11-18'
agenda:
  - classicality
  - framework
tags:
  - Indefinite causal order
  - Quantum SWITCH
  - Higher-order operations
  - Epistemically restricted models
contributions: 2
comments: 2
license: 'CC BY 4.0, unless the entry says otherwise'
---

# Problem 2: Noncontextual SWITCH

From [OpenContext: Open problems in generalized contextuality](https://open-context.jan-bulling.com/problems/noncontextual-switch).

- **Status:** open since 3 Oct 2026
- **Posed by:** Rafael Wagner and John H. Selby, 3 Oct 2026
- **Last updated:** 18 Nov 2026
- **Research agenda:** I. Reconceptualization of classicality; II. Towards a better framework
- **Tags:** Indefinite causal order, Quantum SWITCH, Higher-order operations, Epistemically restricted models

## Statement

Let $\mathcal{S}$ be the quantum SWITCH, i.e., the higher-order operation that takes two channels $A,B:\mathcal{B}(\mathcal{H})\to\mathcal{B}(\mathcal{H})$, with Kraus representations $\{\mathsf{A}_i\}_i$ and $\{\mathsf{B}_j\}_j$, respectively, and outputs 

$$
\mathcal{S}(A,B)[\sigma\otimes\rho] := \sum_{ij} \mathsf{S}_{ij}\,(\sigma\otimes\rho)\,\mathsf{S}_{ij}^{\dagger},
$$
with 
$$
\mathsf{S}_{ij} := \ket{0}\!\bra{0}\otimes \mathsf{B}_j\mathsf{A}_i + \ket{1}\!\bra{1}\otimes \mathsf{A}_i\mathsf{B}_j . 
$$
Fix a prepare-transform-measure fragment $\mathbb{PTM}_{\tt SWITCH}$ of quantum theory that contains: 

(i) Coherent control states in the computational basis;

(ii) non-singleton sets of input channels $\{A\}$, $\{B\}$ together with the resulting channels $\{\mathcal{S}(A,B)\}$; and 

(iii) effects sufficient to evaluate a causal witness that certifies indefinite causal order for this fragment. 


Does this fragment admit of a generalized noncontextual model?

## Background

The quantum SWITCH is a higher-order operation $\mathcal{S}$ that takes two channels $A,B:\mathcal{B}(\mathcal{H})\to\mathcal{B}(\mathcal{H})$ and outputs a channel on a control qubit together with the system,
$$
\mathcal{S}(A,B)[\sigma\otimes\rho] := \sum_{ij} \mathsf{S}_{ij}\,(\sigma\otimes\rho)\,\mathsf{S}_{ij}^{\dagger},\qquad \mathsf{S}_{ij} := \ket{0}\!\bra{0}\otimes \mathsf{B}_j\mathsf{A}_i + \ket{1}\!\bra{1}\otimes \mathsf{A}_i\mathsf{B}_j ,
$$
where $\{\mathsf{A}_i\}_i$ and $\{\mathsf{B}_j\}_j$ are Kraus representations of $A$ and $B$. Take $A$ and $B$ unitary for simplicity. If the control $\sigma$ is in a computational-basis state the channels are applied in a definite order; if it is in a mixed state, in a probabilistic mixture of the two orders; but if it is coherent with respect to the computational basis, the SWITCH appears to implement the two orders in superposition. Such scenarios are said to exhibit indefinite causal order.

A causally “unscrambled” formulation of quantum theory faces an evident obstacle in settings where causal order appears indefinite, and even the broader question of whether generalized contextuality is a meaningful notion of classicality there is hard; some attempts to formalize such a notion exist [9, 10]. Framing indefinite causality within operational probabilistic theories, or within process theories, is itself a major line of active research, which makes it hard to pose near-term questions that do not rely on frameworks still being developed.

There is nevertheless a simple sense in which the question can be tackled with existing tools: for fixed $A$ and $B$, the element $\mathcal{S}(A,B)$ is an ordinary quantum channel, and the classicality content of channels is well understood.

## Relevance

Indefinite causal order is a significantly puzzling effect allowed by quantum theory, one that calls traditional ways of reasoning about causality into question [1, 2, 3, 4], and framing it within operational probabilistic theories, or more generally within process theories, is a major, ongoing line of research [12, 13]. Even whether generalized contextuality is a meaningful notion of classicality there is a hard question, with only a few attempts to formalize such a notion [9, 10]. Most questions one could ask depend on frameworks still under construction. This one does not: for fixed inputs the SWITCH produces ordinary quantum channels, for which an overwhelmingly large literature and toolbox on generalized noncontextuality exists, so the question is well posed today while pointing at the harder one.

The problem traces to Goal II, Towards a better framework. A causally “unscrambled” formulation of quantum theory, in the sense of causal-inferential theories [11], faces an evident obstacle in settings where causal order appears indefinite, and the quantum SWITCH is the simplest and most studied such setting.

## Significance

A generalized noncontextual model is, by construction, a classical model in which all nonclassical features arise from epistemic restrictions alone. If such a model exists for the quantum SWITCH, the puzzling features of indefinite causal order would have an explanation of this kind, which would clarify the role that epistemically restricted toy models can play in interpreting phenomena that seem to lack a causally ordered explanation. If no such model exists, the quantum SWITCH becomes a concrete target for the extended framework that Goal II seeks. Either way, the answer may spark progress toward the much harder question of whether the map $\mathcal{S}$ itself can be considered classical in a sense akin to generalized noncontextuality.

## Research agenda

The long-term goals of the research agenda that this problem traces to.

### I. Reconceptualization of classicality

Determine whether generalized noncontextuality captures the appropriate boundary between classical and nonclassical explanation, the thesis being that it offers a broad, formal, and meaningful framework for what it means for a process in Nature to be “classical”.

### II. Towards a better framework

Develop a formalism that generalizes ontological models while cleanly separating causal (ontic) from inferential (epistemic) elements, in which a suitably adapted notion of noncontextuality can be formulated.

## References

1. G. Chiribella, G. M. D'Ariano and P. Perinotti, Transforming quantum operations: quantum supermaps, EPL 83, 30004 (2008)
2. G. Chiribella, G. M. D'Ariano, P. Perinotti and B. Valiron, Quantum computations without definite causal structure, Phys. Rev. A 88, 022318 (2013)
3. [P. Taranto, S. Milz, M. Murao, M. T. Quintino and K. Modi, Higher-order quantum operations (2025)](https://arxiv.org/abs/2503.09693)
4. [F. Costa, G. Rubino, C. Branciard, Č. Brukner and M. T. Quintino, Indefinite quantum causality (2026)](https://arxiv.org/abs/2606.19438)
5. M. Araújo, C. Branciard, F. Costa, A. Feix, C. Giarmatzi and Č. Brukner, Witnessing causal nonseparability, New J. Phys. 17, 102001 (2015)
6. C. Branciard, M. Araújo, A. Feix, F. Costa and Č. Brukner, The simplest causal inequalities and their violation, New J. Phys. 18, 013008 (2015)
7. D. Ebler, S. Salek and G. Chiribella, Enhanced communication with the assistance of indefinite causal order, Phys. Rev. Lett. 120, 120502 (2018)
8. M. Araújo, P. A. Guérin and A. Baumeler, Quantum computation with indefinite causal structures, Phys. Rev. A 96, 052315 (2017)
9. [R. Kunjwal and O. Oreshkov, Nonclassicality in correlations without causal order (2025)](https://arxiv.org/abs/2307.02565)
10. S. Shrapnel and F. Costa, Causation does not explain contextuality, Quantum 2, 63 (2018)
11. [D. Schmid, J. H. Selby and R. W. Spekkens, Unscrambling the omelette of causation and inference: the framework of causal-inferential theories (2021)](https://arxiv.org/abs/2009.03297)
12. [J. H. Selby, M. E. Stasinou, M. Wilson and B. Coecke, Generalised process theories (2025)](https://arxiv.org/abs/2502.10368)
13. M. Wilson, G. Chiribella and A. Kissinger, Quantum supermaps are characterized by locality, Quantum 10, 2013 (2026)

## Progress and history

- 3 Oct 2026, Posed: Entry created from Problem II of the companion paper.
- 17 Nov 2026, Advance: Independent verification: On the existing solution to the problem as by A,B and C ([OpenContext 2.1](https://open-context.jan-bulling.com/contribution/2.1), by Rafael Wagner)
- 18 Nov 2026, Posed: Reformulation: Reformulation of the problem ([OpenContext 2.2](https://open-context.jan-bulling.com/contribution/2.2), by Rafael Wagner)

## Contributors

Everyone who posed or proposed the problem, edited its text, or added a result or reference, in the order they first contributed.

- **Rafael Wagner and John H. Selby:** Posed the problem
- **Rafael Wagner:** Independent verification: On the existing solution to the problem as by A,B and C (2.1); Reformulation: Reformulation of the problem (2.2)

## Contributions

Material submitted by researchers and approved by the moderators, oldest first.

### 1. Independent verification: On the existing solution to the problem as by A,B and C

- **Date:** 17 Nov 2026
- **Credit:** Rafael Wagner
- **AI use:** no AI assistance
- **Permanent link:** https://open-context.jan-bulling.com/contribution/2.1
- **Votes:** 0 (0 up, 0 down)

I have revised this and agree with everything the text has said.

### 2. Reformulation: Reformulation of the problem

- **Date:** 18 Nov 2026
- **Credit:** Rafael Wagner
- **AI use:** no AI assistance
- **Permanent link:** https://open-context.jan-bulling.com/contribution/2.2
- **Votes:** 0 (0 up, 0 down)

What if the switch doesn't switch well?

## Discussion

Comments approved by the moderators, oldest first, each followed by its replies.

### Rafael Wagner, 18 Nov 2026

This question shall have two posers Rafael Wagner (me) and John H. Selby (another moderator)

### Rafael Wagner, 18 Nov 2026

Increadible that there are no problems under revision yet!

## How to cite

```bibtex
@misc{opencontext-problem-2,
  author        = {Rafael Wagner and John H. Selby},
  title         = {{Noncontextual SWITCH}},
  howpublished  = {OpenContext: Open problems in generalized contextuality, Problem 2, version 1},
  year          = {2026},
  month         = oct,
  date          = {2026-10-03},
  url           = {https://open-context.jan-bulling.com/contribution/2},
  urldate       = {2026-11-20},
  note          = {Problem posed 3 October 2026; status: open, last updated 18 November 2026}
}
```

## Working on this problem

- A solution must resolve the problem **as stated**, by the criterion fixed in the statement. A proof of a weakened or modified statement is a partial result, not a solution.
- Check the contributions above before starting, so as not to repeat existing work.
- Submit results at https://open-context.jan-bulling.com/problems/noncontextual-switch, with the extent of AI use and the names of the humans who take responsibility for them. AI-generated material counts as progress only after human scrutiny.
- Cite the problem and the people who posed it. The full rules: https://open-context.jan-bulling.com/guidelines
