It’s a paper about replicability, meta-analyses, whether the scientific method can be trusted.

They say publication bias is a big problem, and for that reason we see absurd results in the literature. Like the way meta-analyses of ganzfeld studies show find a 32% rate of correctly guessing ‘transmitted’ images, instead of the 25% rate you’d get from chance.

To show how the published literature is different from real experimental results (to prove their point about publication bias), they do eight ganzfeld studies… and get a 32% hit-rate.

Discussion here: https://psi-encyclopedia.spr.ac.uk/articles/delgado-romero-and-howard-telepathy-experiments