Agreed. The two arcs illusion shows "seeing" isn't even a very good ruler, let alone ray tracer. And there are lots of illusions that demonstrate in the right c

Agreed. The two arcs illusion shows "seeing" isn't even a very good ruler, let alone ray tracer. And there are lots of illusions that demonstrate in the right context, our eyes aren't even all that good at judging completely different colours. We don't instinctively know that sound and light are very similar things, but at very different wavelengths. We know we can let go of magnetic letters on the fridge, and they won't fall off. That doesn't mean our bodies instinctively understand the quantum interaction that makes that happen. Yet there is solid math used to model how strong the magnet needs to be in order to hold the plastic letter in place on a vertical metallic plane. The non-mathematical person can detect whether a magnet is too strong or too weak for the purpose without at all understanding the math or science behind it. Math is a good predictor of things we can model with it. That doesn't mean humans in general necessarily understand the model in a mathematical context. We have expectations based on experimentation (or training) and observation, which is a totally different thing.