# Bell Correlations as Selection Artefacts

**Authors:** Huw Price, Ken Wharton  
**arXiv:** [2309.10969](https://arxiv.org/abs/2309.10969)  
**Date:** 2023-09-19  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2309.10969v1/

## Key points

- Huw Price and Ken Wharton propose an explanation for correlations observed in Bell experiments
- They argue that these correlations can arise as a special type of selection artefact
- Their explanation challenges the notion of nonlocality and reconciles it with relativity
- The authors update their proposal by demonstrating its application in a real Bell experiment
- They avoid the explicit postulate of retrocausality present in the previous version of their proposal
- Alice's influence over the initial state of actual experiments does not require retrocausal effects, but rather necessitates that she does not rule out such influence in hypothetical scenarios where Initial Control is missing
- Bell correlations can emerge as a selection artefact in real experiments by considering this possibility
- The argument extends beyond the specific case of BIG-∨ and suggests enlarging the model to explain more ways in which Bell correlations can be preselection artefacts
- Physically realistic versions may require mixing result types to align with experimental constraints
- Two paths to potential realization of retrocausality are discussed: backward-in-time signaling and time-symmetry breaking
- This paper presents an updated explanation for Bell correlations based on selection artefacts rather than nonlocality or retrocausality
- Empirical support is provided through a real Bell experiment
- Insights into how this explanation can be applied to more complex models are offered.

## Summary

In their paper titled "Bell Correlations as Selection Artefacts," Huw Price and Ken Wharton propose an explanation for the correlations observed in Bell experiments. They argue that these correlations can arise as a special type of selection artefact, without the need for direct spacelike causality or influence. This explanation challenges the notion of nonlocality and offers a novel way to reconcile it with relativity. The authors update their earlier proposal in two main aspects. Firstly, they demonstrate its application in a real Bell experiment, providing empirical evidence for their explanation. Secondly, they avoid the explicit postulate of retrocausality, which was present in the previous version of their proposal. The key idea behind their argument is that Alice's influence over the initial state of actual experiments does not require retrocausal effects. Instead, it simply necessitates that she does not rule out such influence in hypothetical scenarios where Initial Control is missing. By considering this possibility, they show how Bell correlations can emerge as a selection artefact in real experiments. The authors emphasize that their argument extends beyond the specific case of BIG-∨ (a simplified model used for illustration purposes). They suggest that enlarging the model will provide more ways in which Bell correlations can be explained as preselection artefacts. In discussing physically realistic versions of BIG-∨, Price and Wharton acknowledge that mixing result types may be necessary to align with experimental constraints. They also address the concept of retrocausality and highlight two paths to its potential realization: one involving backward-in-time signaling and another involving time-symmetry breaking. Overall, this paper presents an updated explanation for Bell correlations based on selection artefacts rather than nonlocality or retrocausality. The authors provide empirical support for their proposal through a real Bell experiment and offer insights into how this explanation can be applied to more complex models.