I think that this (common) reaction probably illustrates a kind of over-correction among experienced problem-solvers: the knowledge that not everything is line

I think that this (common) reaction probably illustrates a kind of over-correction among experienced problem-solvers: the knowledge that not everything is linear (e.g.: if you add a bit to the area of a circle, by how much does the radius change?) leads to the assumption that nothing is linear (or, at least, nothing that would crop up as a math-circle problem).