On dataset tensions and signatures of new cosmological physics

TL;DR

This paper critically examines the use of dataset tensions, like the Hubble tension, as evidence for new physics, emphasizing the importance of Bayesian model comparison and cautioning against conclusions based solely on tension reduction.

Contribution

It introduces a Bayesian framework that clarifies how dataset tensions relate to model probabilities, highlighting that tension alleviation alone does not confirm new physics.

Findings

Standard tension metrics are incomplete without data co-dependence terms.

Reducing dataset tensions does not necessarily favor extended models.

Claims of new physics based solely on tension alleviation are unreliable.

Abstract

Can new cosmic physics be uncovered through tensions amongst datasets? Tensions in parameter determinations amongst different types of cosmological observation, especially the `Hubble tension' between probes of the expansion rate, have been invoked as possible indicators of new physics, requiring extension of the $\Lambda$CDM paradigm to resolve. Within a fully Bayesian framework, we show that the standard tension metric gives only part of the updating of model probabilities, supplying a data co-dependence term that must be combined with the Bayes factors of individual datasets. This shows that, on its own, a reduction of dataset tension under an extension to $\Lambda$CDM is insufficient to demonstrate that the extended model is favoured. Any analysis that claims evidence for new physics {\it solely} on the basis of alleviating dataset tensions should be considered incomplete and suspect. We describe the implications of our results for the interpretation of the Hubble tension.