Is practice sufficient for expertise?

July, 2010

A study of sight-reading ability in pianists confirms the importance of many hours of practice, but also suggests that working memory capacity makes a difference.

A new study challenges the popular theory that expertise is simply a product of tens of thousands of hours of deliberate practice. Not that anyone is claiming that this practice isn’t necessary — but it may not be sufficient. A study looking at pianists’ ability to sight-read music reveals working memory capacity helps sight-reading regardless of how much someone has practiced.

The study involved 57 volunteers who had played piano for an average of 18.6 years (range from one to 57 years). Their estimated hours of overall practice ranged from 260 to 31,096 (average: 5806), and hours of sight-reading practice ranged from zero to 9,048 (average: 1487 hours). Statistical analysis revealed that although hours of practice was the most important factor, nevertheless, working memory capacity did, independently, account for a small but significant amount of the variance between individuals.

It is interesting that not only did WMC have an effect independent of hours of practice, but hours of practice apparently had no effect on WMC — although the study was too small to tell whether a lot of practice at an early age might have affected WMC (previous research has indicated that music training can increase IQ in children).

The study is also too small to properly judge the effects of the 10,000 hours deliberate practice claimed necessary for expertise: the researchers did not advise the number of participants that were at that level, but the numbers suggest it was low.

It should also be noted that an earlier study involving 52 accomplished pianists found no effect of WMC on sight-reading ability (but did find a related effect: the ability to tap two fingers rapidly in alternation and to press a computer key quickly in response to visual and acoustic cues was unrelated to practice but correlated positively with good sight-readers).

Nevertheless, the findings are interesting, and do agree with what I imagine is the ‘commonsense’ view: yes, becoming an expert is all about the hours of effective practice you put in, but there are intellectual qualities that also matter. The question is: do they matter once you’ve put in the requisite hours of good practice?