I can just try to explain what part of this I'd find valuable to research.

All that speculation - mine's, GP's, OP's - hinges on one assumption: That something like hereditary visual detectors in the brain exist.

I.e. that there are structures in the brain that have "weights" for large eyes, or cat features or spider features, etc etc - and that those weights are not learned by the individual, but are somehow "hardcoded" and passed down the germ line - which would allow them to be "learned" through evolution of the species.

As a programmer and with my hobbyist understanding of molecular biology, I'd see this as a pretty remarkable hypothesis. Right now, I don't see how this could possibly work: The brain and even the eyes of every person are different, so how could such a detector be "reconstructed" on a cellular level for an individual who has never seen a spider?

It would also raise interesting follow-up questions, both if it were confirmed or disproven:

If it were confirmed, does this mean there are encoded bits of visual information in the DNA? Could we decode them somehow and get "photographs" from prehistoric or even pre-human times? (Or well, less photographs and more something like the "eigenfaces" of face detectors) Are there more such hardwired circuits we didn't know yet? Are there similar circuits for other senses or for higher-level areas in the brain?

On the other hand, if it were disproven, we'd have to rethink situations where we take the existence of such hardwired stimuli almost for granted, like in sexual imagery.

The cat stuff itself has no predictive value, but it points into directions that could deliver it.