I think it is at least plausible that some fairly complex image recognition could be genetically determined. Indeed, the classic example of the seagull proves that pre-programming such circuitry is possible in animals with significantly simpler brains than mammals. I don't think it's such a wild leap that more complicated image recognition could be possible at birth.
At any rate, it's not even necessary that such image recognition be completely formed at birth. It can develop along with the rest of the visual system and other systems that are tuned for sexual attractiveness. (I'm ignoring breasts-for-food-purposes, the recognition of which is largely determined by tactile stimuli, mediated by the trigeminal nerve).
At any rate, it's not even necessary that such image recognition be completely formed at birth. It can develop along with the rest of the visual system and other systems that are tuned for sexual attractiveness. (I'm ignoring breasts-for-food-purposes, the recognition of which is largely determined by tactile stimuli, mediated by the trigeminal nerve).