It has nothing to do with the definition of math. If you put people in an MRI and give them random problems then you’ll notice a distribution in the patterns of brain activity that are distinct. Some of these groupings we’ll call “math like” because the problems are more like mathematics than linguistics. And in the other distribution we’ll call them “word like” because the problems fall into the category we associate with word problems.

The fact that we see separate patterns of brain activity is the interesting part not any semantic argument about what you or mathematicians feel should be defined as being math or not.