[xg15]

Yeah, I'm not from the field, so I'm probably missing the cause of that negativity?

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.

[throwup238]

Here’s where it helps to have at least some background in biology: what you propose is not a radical hypothesis. In fact I’d be hard pressed to find a single neuroscientist who disagrees with that hypothesis except in its most nuanced details.

We have no idea how that information is transmitted from generation to generation but we have enough animal behavior research to know that many animals instinctually identify visual cues like predators pretty much from the moment they are born. We also have decent evidence that the inheritance may not be entirely genetic in origin, because nearly identical populations in different locales may have wildly different behaviors (like the animals in the Galapagos islands, who aren’t afraid of humans because they’ve had no predators).

The three main candidates for how this information is transmitted are: genetics, epigenetics, and embryonic development. The latter two fields are still in their infancy but that leaves more room for just so stories.

Always beware of anyone using evolutionary biology to make an argument about the development of species.

[xg15]

That sounds entirely reasonable and also extremely exciting. I didn't know there is so much research on that topic going on. Thank you for explaining that.

> Always beware of anyone using evolutionary biology to make an argument about the development of species.

From what I understand so far, I can just say that I dislike arguments of the form "feature X provides evolutionary advantage Y to the species, and that explains why the species evolved it".

In fact, it doesn't explain anything: Being able to shoot laser beams out of ones eyes would provide a large evolutionary advantage, but I still don't expect any future children of mine to spontaneously develop that ability - because there is no feasible way how the body could change to realize that ability, how that change would be transmitted from the parents or how it would even develop in the first place.

I think it could still be a useful shorthand if you already know that an aspect of the body is influenced by evolution, to explain the "direction" this feature took.

But I think I see what you mean, there is a risk of getting caught up in "it could have happened like this" speculation that is not grounded in any reality anymore.

[throwup238]

> But I think I see what you mean, there is a risk of getting caught up in "it could have happened like this" speculation that is not grounded in any reality anymore.

Almost but not quite. Evolutionary laser beams can be discounted out of hand but when we’re talking about a feature (phenotype) that already exists, it is somewhat grounded in reality because we have incontrovertible proof that it can evolve. That opens it up to wild speculation about convoluted evolutionary paths and, since 500+ millions of years of natural selection is nothing if not convoluted, very many of them sound very reasonable.

The added complication is that most of the evidence doesn’t exist anymore or is so degraded our current technology can’t recover the data. Figuring out the “direction” of evolution is nearly impossible because we have few intermediate genetic samples for organisms beyond ~500-1000kYa. The science of extracting DNA from recent fossils is also in its infancy.