does how many foundation model layers are required for a particular competency correlate with evolution

in the platonic representation hypothesis and what it says about the world, I argue that this layered similarity of representations between different large models is a result of the fact that evolution built up our model of the world layer by layer
this hypothesis would imply that when a particular competency arose evolutionarily should be monotonically related with how many layers of these foundation models are required to perform this task
e.g. one might expect something like "contains an animal" or "has something scary" to be predictable with fewer layers than something like number sense