Expertise – a matter of practice? Part B

Deliberate practice: Is that all it takes to become an expert? 

David Z. Hambrick ,  Frederick L. Oswald , Erik M. Altmann , Elizabeth J. Meinz, Fernand Gobet , Guillermo Campitelli .  Intelligence 45 (2014) 34–45

https://www.msu.edu/~ema/HambrickEtAl.pdf

In the previous post we've seen that practice explains about 34% of the variance in chess performance.  A significant amount,  but there certainly are other factors that explain the variance.  What are they?  And what about music?

The musicians who took part in the music studies discussed in this paper had different degrees of expertise.  They stated their formal music education   and estimated the number of hours they practiced each week throughout the years.

The performance criterion was expert ratings of the player's performance, the number of errors made by pianists while accompanying, pianist's ability to sight read the score and other similar measures.

As in chess, music practice explained 30% of the variance in music performance.  In one of the chess studies, a certain chess player needed 26 years of practice to attain master level, while another player attained master level in less than 2 years.  Likewise in music: at each skill level, some musicians needed a fifth of the mean practice time for that level, while others needed four times the mean practice time.

If practice explains only 30-34% of the variance in performance in chess and music – what else explains it?

An early start – chess players who began at an early age became more successful, regardless of the amount of practice they had.  Simonton discovered, that the most successful and acclaimed classical composers began studying and composing music at an earlier age than less successful classical composers.  There may be a critical period for acquiring certain skills.

Intelligence – performance level in music and chess is significantly correlated with IQ. Of the cognitive abilities, working memory capacity accounts for the most significant amount of variance in performance.  The study tells of a six year old boy who was a piano prodigy, played in concerts and almost never practiced.  His working memory score in the Stanford Binet test was 158.

Genetics also predicts performance level.

Personality – especially grit – predicts high level performance in many areas.  Grit leads to an ability to invest in hours of hard practice, to overcome difficulties etc.  I guess that other personality factors like interpersonal ability, self consciousness and openness to experience are also relevant, as well as other factors that I haven't mentioned. 

One can argue, that since this study is based on player's estimations of the amount of practice they had had years ago, the data is not so reliable.  But the authors claim that retrospective reports have reasonably good correlations with diary based reports. 

Another reason for my tendency to believe these claims is the fact that IQ scores are correlated 0.4-0.7 with school grades.  This is a kind of a "mirror image" of the study we are discussing.  0.4-0.7 correlations are quite significant, but obviously other factors also influence success in school.  There's no doubt that student's motivation and grit matter, and also teaching quality, school climate, the home environment and other factors.

Expertise – a matter of practice?

One of the founding fathers of the study of intelligence, Francis Galton, was a very interesting and colorful figure.

Galton, who was Charles Darwin's cousin, wanted to implement Darwin's theories on humans, and to improve humankind by mating between people with the best mental, moral and physical skills.  Nowadays such an idea sounds shocking, but in the nineteenth century it might have sounded reasonable. Galton looked into families of people who were considered wise and successful, and saw that the closer the  blood relation to a wise person is,   the better are the chances of this person's relative to be also wise.  Galton deduced that "wisdom" (the term "intelligence" didn't exist then) is, at least partly, hereditary.  Galton studied children who were adopted by the Pope, to see what is more influential: genetics (which was "weak" in these children's case) or environment (an environment rich with stimuli at the Pope's surrounding).  He found that there is an interaction between genetics and environment, but genetics is stronger than environment.

Apropos Galton, he probably was the only psychologist in history who was paid by his experiment participants for this privilege!  Galton studied other things, like women's attractiveness in different parts of England, and the relationship between prayers for the wellbeing of kings and their life expectancy (he found there was no relationship…).

While Galton worked in the 19^th century, John Watson, one of the founding fathers of Behaviorism, wrote in 1930: " Give me a dozen healthy infants, well-formed, and my own specified world to bring them up in and I'll guarantee to take any one at random and train him to become any type of specialist I might select—doctor, lawyer, artist, merchant-chief and, yes, even beggar-man and thief, regardless of his talents…" 

Anders Ericsson sided with Watson.  In the 1990's he interviewed violinists and pianists and found that the violinists and pianists who were considered to be the best of their kind accumulated more than 10000 hours of practice by age 20.  Amateur violinists accumulated about 5000 hours of practice and amateur pianists – about 2000 hours of practice by age 20.  Ericsson argued that  genes may contribute to individual differences in people's willingness to engage in deliberate practice over a long period of time, and thus may indirectly contribute to individual differences in performance, but  he explicitly rejected the view that innate ability can account for why some people become experts and others fail to do so.  He argued that expert's performance reflects a long period of practice more than innate ability or talent.

It's easy to see why Ericsson's 1993 paper on this subject became a hit and was cited more than a thousand times.

Clearly, many people who want to become experts – as violinists, physicists, athletes and so on – don't make it despite investing hours of practice and effort.

In this paper:

Deliberate practice: Is that all it takes to become an expert?

David Z. Hambrick ,  Frederick L. Oswald , Erik M. Altmann , Elizabeth J. Meinz, Fernand Gobet , Guillermo Campitelli .  Intelligence 45 (2014) 34–45

https://www.msu.edu/~ema/HambrickEtAl.pdf

the researchers asked: is practice enough to become an expert?  Can  individual differences in performance be explained mostly by individual differences in practice hours?

They studied these questions by analyzing research done in two areas of expertise:  chess and music.

The chess studies included players who were rated in the international rating, the Elo, were members of prestigious chess clubs, participated  in national chess programs and so on.  These players gave information about the age they started playing and estimated the number of practice hours they had each day during their training years.  The authors found that practice explained 34% of the variance in chess performance (the player's Elo rating).  34% is a significant amount, but practice did not explain everything!  They also found, that some people needed much less practice to reach an excellent level of performance in chess.  Some expert chess players needed 3 times more hours of practice than other expert chess players.  Some expert chess players needed less practice hours than intermediate chess players.  Some intermediate chess players practiced more than expert chess players.

The paper describes the story of the Polgár sisters who, beginning at a young age, received several hours of chess instruction every day from chess grandmasters and their father, a chess teacher and author of several chess books.  The sisters differed both in the highest rating they achieved and in the amount of practice they accumulated to reach that rating. One sister's peak rating was 2735 in an estimated 59,904 h of practice, whereas another sister's was 2577  in an estimated 79,248 h of practice.  The two sisters who became grandmasters had accumulated a great deal more practice by the time they reached their peak rating than had the eight grandmasters who reached top-ten in the world with a mean of about 14000 hours of practice. 

 

Deliberate practice is clearly not sufficient to account for individual differences in chess performance.   What else explains chess performance? What about excellence in music?    These topics will be discussed in the next post.