The study didn't test for mathematical ability. It tested for numeracy which is completely different.
The study even shows that most of the variance(34%) is explained by fluid reasoning and working-memory. I'd wager that fluid reasoning is strongly correlated with mathematical ability.
In my opinion this paper disproves it's stated hypothesis but doesn't realize it.
Many years ago I read a study that said English majors were more likely to succeed as programmers than any other major.
I don't think people really appreciate how hard it is to write well and to understand all the rules in a language. I suspect that if you can master that in English, you can use it in any language.