Music and maths

Because I am in some respects both a mathematician and a musician, I often get people asking me whether mathematical ability and musical ability are correlated. Well, I expect they are, since most pairs of intellectual abilities are positively correlated to some extent. But are maths and music more significantly correlated than other pairs of subjects?

I did a piece of small-scale research into this many, many years ago. I tested a class of children aged 9–10 years, using a range of mathematics achievement tests and a set of tests of musical ability. Overall, the scores for the two subjects did correlate positively – but it was only a moderate level of correlation. But I also worked out all the correlations between maths and the different aspects of musical ability. I found that there was hardly any correlation between mathematics and the vertical relationship in music, like recognising intervals and hearing chords. The strongest correlation with maths attainment was in the ability to recognise horizontal relationships in music, such as lengths of notes and rhythmic patterns.

This is consistent with my own experience, since I am pretty adept at sight reading (on the trumpet) and handling rhythmic patterns. But, as any member of the choir that I conduct will tell you, I can't sing in tune! That's why they prefer me to conduct rather than to sing: I do less damage that way. I do quite a bit of musical arrangement and some composition, mainly for use at my church. But I tend to approach this more analytically than intuitively, which probably reflects my mathematical cast of mind.

I did not attempt to replicate my small-scale study or to get it published, because, frankly, like most purely correlational research it is pointless. If you do establish a correlation between two variables it is not really anything that you can make use of. You can't deduce, for example, that doing more music – or specifically more work on rhythm in music – would make someone better at maths, or vice versa.