As neuroscience gains the power to record thousands of neurons at once and compare biological brains to artificial ones, it finds itself rich in tools but poor in consensus — more than thirty competing methods now claim to measure neural similarity, yet the field has not agreed on what similarity truly means. This is not merely a technical dispute; it is a philosophical one, echoing the oldest challenge in comparative science: whether resemblance is a surface pattern or a window into shared mechanism. The kidney researchers of the 1960s did not stop at correlation — they used it to reveal how