The hypothesis is specifically about developing eyesight. We don’t have any reason to think that factors in adult eyesight have any relevance to infant development, so this is irrelevant.

This explains why we care that premature babies have worse eyesight: there's a link to P-fat levels. This strengthens the causal connection between P-fat and developing eyesight, which supports the hypothesis.

All this tells us is that eyesight isn’t totally genetic. However, we’re specifically trying to strengthen the connection between P-fat and eyesight. This doesn’t do that for us, so it's irrelevant.

We don’t care what babies prefer, because we don't know if that has anything to do with how they develop eyesight. Without a more explicit connection, their preferences aren't relevant.

If this means there's no further development after birth, then this weakens the scientists’ hypothesis, because the postnatal difference between formula and mother’s milk wouldn’t be relevant. And we still don't know what role P-fat has in gestation.