25 October 2011

Grit Not Enough

Deliberate practice (DP) occurs when an individual intentionally repeats an activity in order to improve performance. The claim of the DP framework is that such behavior is necessary to achieve high levels of expert performance. The proponents of the framework reject evidence that suggests that other variables are also necessary to achieve high levels of expert performance, or they claim that the relationship between those variables and expert performance is mediated by DP. Therefore, the DP framework also implies that DP is sufficient to achieve high levels of expert performance. We test these claims by reviewing studies on chess expertise. We found strong evidence that abundant DP is necessary (but not sufficient) and estimated that the minimum requirement to achieve master level is 3,000 hours of DP. We also review evidence showing that other factors play a role in chess skill: general cognitive abilities, sensitive period, handedness, and season of birth.

From Guillermo Campitelli1 and Fernand Gobet "Deliberate Practice Necessary But Not Sufficient," 20 Current Directions in Psychological Science vol. no. 5 280-285 (October 2011) (Hat tip to Dienekes).

As the press release for the study notes:

In one survey of chess players in Argentina, Campitelli and Gobet found that, indeed, practice is important. All of the players that became masters had practice at least 3,000 hours. “That was not surprising,” he says. There is a theory in psychology that the more you practice, the better you’ll do in areas like sports, music, and chess. “But the thing is, of the people that achieved the master level, there are people that achieved it in 3,000 hours. Other people did, like, 30,000 hours and achieved the same level. And there are even people that practiced more than 30,000 hours and didn’t achieve this.”

Campitelli and Gobet concluded that practice is necessary to get to the master level—but it’s not enough. . . . [Also] about 90 percent of the general population is right-handed, only about 82 percent of adult chess players are right-handed.

Establishing a similar point, D. Z. Hambrick, E. J. Meinz show in "Limits on the Predictive Power of Domain-Specific Experience and Knowledge in Skilled Performance." Current Directions in Psychological Science, 2011; 20 (5): 275 DOI: 10.1177/0963721411422061, that general IQ can predict a large share of performance in tasks like piano sightreading, even controlling for hours of practice.

In one experiment Hambrick and Meinz tested 57 pianists with a wide range of deliberate practice under their belts, from 260 to more than 31,000 hours, to see how well they did on sight-reading -- playing a piece from a score they'd never seen before. Those who had practiced more did better. In fact, practice -- even specific sight-reading practice -- predicted nearly half of the differences in performance across the subjects. But working memory capacity still had a statistically significant impact on performance. In other words, regardless of amount of deliberate practice, working memory capacity still mattered for success in the task. . . . Challenging another "experts-are-made" contention -- that beyond a certain threshold, intelligence makes less and less of a difference in accomplishment -- the authors cite a study by Vanderbilt University researchers that looked at the math SAT scores of people with PhDs in science, technology, engineering, or math. Those who scored in the 99.9th percentile at age 13 were 18 times more likely to go on to earn a PhD than those who scored better than only 99.1 percent of their teenage peers. "Even at the highest end, the higher the intellectual ability -- and by extension, the higher the working memory capacity -- the better," says Hambrick.

The result is not surprising (to me), but is important.

No comments: