How to test the hypothesis of a positive correlation between group-level effects in brms model

Modeling
, ,
1

I know how to get the Bayes Factor for population-level effects, e.g.,
hypothesis(bayesmodel, "condition:tester > 0")

However, for a planned study I need to get the magnitude of the evidence for a correlation between two random effects in a brms GLMM. I do not know how to extract it. I know that I can get the credible interval, but in case that includes 0, I need to get the magnitude of the evidence for or against such a correlation. Does anyone know how I can get the Bayes factor for such a hypothesis? I would be super thankful

2

In case anyone els is wondering, I found the solution:
I extracted the name of the effect via posterior_summary:
posterior_summary(bayesmodel)
The name of the effect of interest in my case was cor_subject__condition:tester__condition:session, the “cor” in the beginning reflecting the class.
I was then able to test my hypothesis via:

hypothesis(bayesmodel, "subject__condition:tester__condition:session > 0", class = "cor")

I thus dropped the ‘cor_’ in the beginning and was able to test the effect then.