This kind of thing has always seemed redundant to me. We already have a perfectly good understanding of how decoherence causes effective collapse, by making bra

This kind of thing has always seemed redundant to me. We already have a perfectly good understanding of how decoherence causes effective collapse, by making branches of the wavefunction unable to interfere with each other. This follows automatically from the basic rules of quantum mechanics, without having to add any extra ingredients, and is absolutely necessary to get the predictions to come out right (e.g. in cosmology, where there were no conscious observers at all). This article proposes to add an extra, ad hoc collapse mechanism determined by a system's integrated information, but what's the point? At best, it will be a more complicated rephrasing of an effect that already happens automatically, in which case it is unnecessary by Occam's razor.