learning – The Science Talent Project

Talent, desire, training, or tricks – what makes a top performer?

At the end of one of my recent posts I mentioned that aspiration to do better may be the characteristic distinguishing top performers and medium performers in many fields. However, while most people would likely agree that such differences in motivation may make a difference in regular jobs, there are also fields like professional sports, soloists in music, and academia, where competition is fierce and one has to be better than one’s peers to retain one’s job. Wouldn’t all those professionals be equally fiercely motivated? If that is the case, motivation would be a bad explanation for performance differences. Should we again try explain things by ‘talent’ or, if we look further, are there other factors we should take into account?

Most people, by definition, are not top performers in their jobs or careers. However, they generally do not have to be; as long as they are perceived by their employer or customer to deliver more value than they are paid, or customers or employers cannot very well judge the quality and quantity of their work, they can generally remain in their jobs and earn money for themselves and perhaps their family. So most people have no great need to improve in their job.

However, life is quite different if one is a professional sporter, artist, or untenured scientist. If you don’t perform very well compared to your competition (which tends to be considerable) you are soon out of clients, sponsors, funds, or a job. You really cannot slouch that much, every professional in these fields would have at least some motivation to perform well and pay attention to their performance. So it would be natural to think that if all performers are motivated if not very motivated, differences in performance in those fields cannot come from motivation, but must come from other factors, like talent.

Of course, talent may exist. In some fields, like the 100 m sprint, the dominance of black sprinters suggests some genetic talent that can give a small but crucial advantage over people who did not originate from the west of Africa. Sometimes genes play a role. The genetic advantage in this case may be small (0.15 s, or less than 2%), but in sports, very small differences count, as differences between best and worst professional sprinters may only be like 10%, which is noticeable, but is obviously much less than the reported factors of 2 to 10 differences observed for scientists, programmers and teachers.

So while talent may give explain some performance differences, one should also look at other factors. Quality and quantity of training, for example, and -indeed- differences in motivation.

The easiest factor to assess are differences in training. In general, top performance started training earlier in their field than less prominent performers. Mozart who started music lessons at age 2, and Tiger Woods, who even before that age was placed by his father in a children’s chair to watch dad Woods hit golf balls again and again. Michael Jackson and Andre Agassi similarly started (or rather were started) their training early, driven by dominant dads. It is hard to separate the effects here: brain plasticity is likely highest at young age, social support/pressure is enormous as the trainer is someone incredibly important to the trainee, the amount of time available for training is very large, and social recognition for the ‘prodigy’ will likely also enhance motivation. How early training starts, how much training one has received, and how much attention one pays to the training that is received simply because the trainer happens to be someone very important to you will all influence how much is learned. Definitely, differences in performance will at least for a part reflect differences in the amount of effective training, one can definitely explain differences even among professionals for a large part due to differences in training instead of immediately claiming differences in ‘talent’.

Another reason to believe in the importance of training, also in fields like science which seem to have much less fanatical dads drilling their offspring, is the phenomenon reported by the sociologist Harriet Zuckerman in her book “Scientific Elite“: over half of US Nobel prize winners had been either PhD student or postdoc in the laboratory of another Nobel prize winner. Most of those who hadn’t, had worked as PhD student or postdoc of a scientist who was nominated for a Nobel prize (but did not get it), or had a supervisor who worked closely together with another (future) Nobel prize winner. In short, most extremely successful scientists worked closely in their ‘formative years’ with scientists who also were extremely good. And they indeed reported that their experience with those top scientists had been very important to them, not so much as for learning subject matter knowledge (like physics formulas) but ways of thinking and doing research.

Now, one could argue that learning might not have played a role at all here; the ‘mentors’ could have gotten their Nobel prize together with their mentee (like happened to physicist Gerard ‘t Hooft who shared his Nobel prize with his mentor Martinus Veldman), or the mentors used their social clout (Nobel prize winners can nominate candidates) to bring their favorite pupils to Stockholm too. Or only the most promising young scientists were selected by the mentors to join their labs; the young scientists would very likely received a Nobel prize on their own, their talents were simply recognized early.

The above factors may indeed have an influence, however, a closer look at the data suggests that education plays an important part too. The persons who supervised most future Nobel Laureates, Arnold Sommerfeld and Enrico Fermi, were neither the most famous scientists of their age (so they could not necessarily expect the best and brightest to flock to them), nor did they have the strongest networks they could use to ‘Nobelize’ their mentees; Sommerfeld never won a Nobel prize so could not nominate people himself, Fermi died relatively young. However, the striking similarity between the two mentors is that they cared a lot about mentoring their students. In Cropper’s book “Great Physicists“, Fermi was the only one described as wanting to become a professor because he loved teaching, and while he carefully selected his students, he was willing to spend infinite amounts of time and care on mentoring them. Sommerfeld was a supervisor who would have 1-to-1-discussions about their projects and developments in science in general for about 1 hour every other day with each of his PhD students, which is an enormous amount compared to the more typical half an hour a month that normal supervisors are willing to spare. Of course, Fermi and Sommerfeld were themselves also very good researchers. However, among the many great researchers in the 20th century they stood out not for being the best researchers but as those professors most dedicated to give promising young scientists a great education. And the record numbers of Nobel prizes for their pupils (7 for Sommerfeld, 6 for Fermi) seems to indicate that they were quite effective in that.

So even in science, training is important, and it is likely that top educators are in shorter supply than top potential talent is.

And the original premise of this blogpost, that in highly competitive fields all performers should be equally motivated? That would also seem to be untrue, in two aspects even: not only is the amount of motivation different between people (and likely in individuals over time), but people actually have multiple possible motivations, influencing their effectiveness in reaching goals.

For example, Daniel Pink (in his book ‘Drive’) cites research claiming that motivation to master one’s field is more correlated to the number of patents obtained than the desire for money is, even when controlled for effort. The article “Making It in Academic Psychology: Demographic and Personality Correlates of Attainment” indicates that both the quantity of motivation differs among professional academic psychologists (so motivation is unequal even among practitioners in a competitive field), and the types of motivation. Quantity of motivation is important: more motivated psychologists publish more. And kind of motivation is too: if psychologists were mainly motivated by the desire to outdo others, they published below-average work, if psychologists desired to master their field or simply liked to work hard, their work was above-average.

Talent may exist, but based on the above, training and motivation can move one up and down in any competitive hierarchy, which can at least make the difference between keeping one’s job and needing to find another career.

The above may sound nice and even somewhat convincing, but what can a practical scientist/artist/sportsperson do with that knowledge? Three conclusions stand out.

It pays to find a good mentor/coach. Definitely, some weeks of effort, of networking, of asking more experienced people who they think are (beside themselves) make good mentors or coaches can pay off enormously. At least much more than those same four or such weeks simply spent ‘working hard’
Try to keep your motivation high, for example by using the general strategies described in my previous posts on ‘passion growing‘ and motivation.
You may also want to stress your personal growth and mastery instead of ‘beating others’. How to do that, I’m not sure yet, perhaps by praising and rewarding yourself when you have learned or improved something, instead of only congratulating yourself when you have ‘beaten’ another person. That being said, not all fields may be like academic psychology, and quite some top sporters became quite successful despite attaching great value to winning (like John McEnroe’s temper tantrums seemed to indicate).

In conclusion: while professionals in competitive fields may differ in talent, they also definitely differ in training and motivation; and while you may not reach the absolute top by only carefully handling your motivation and finding the best mentors you can possibly get, you can definitely achieve more (if not much more) that way.

So what is next for this blog? I’ve discussed the basics of training and of motivation, and while subjects like creativity, networking and marketing are also important for scientists, I’ll probably first check my archive to see which subjects may be most relevant to update and repost. And I’ll also add some book reviews, so you can know where to look or study further.

Till next time!

Why experience does not necessarily lead to excellence

Summary: As discussed in my previous post, a huge amount of effort and experience is needed to become a top performer. Even if there were something like ‘talent’, by itself ‘talent’ is by far not enough; even Einstein worked hard at maths and physics for more than 15 years before his ‘miracle year’ breakthrough. Of course, this leads to the counter-question: if experience is necessary, and perhaps even sufficient (as the Ericsson investigation seemed to imply), why are then not all professional performers equally good? The difference has to be caused by differences in talent, right? The answer is however ‘not necessarily’, as experience is a tricky teacher, which can make you great, but much more commonly teaches you until you reach an ‘acceptable’ level. This post is about how that works and how you can avoid or escape ‘experienced mediocrity’ and reach excellence (or at least pretty impressive performance) instead.

Practice makes p_____. What is the second word?

“Permanent”

-Joke among educational scientists

A while ago I was discussing with some acquaintances the rule that ten years of practice are necessary to achieve world-class performance. One of them said: then we must be world-class, for we have worked in our field for over ten years!

My acquaintances may have been world-class in their field (I wouldn’t know the ranking order in their profession). Generally observing people around me, however, I must unfortunately conclude that 10 years (or, more clearly, 10000 hours) of practicing anything does not necessarily guarantee top performance, or even performance that is anything close to top performance. This is not just in science, where US Nobel laureates publish two times as many papers as scientists still worthy enough to make it into ‘American Men and Women of Science’ (Dean Keith Simonton, “Origins of Genius“). There are experienced teachers who fall far below the level of equally experienced but ‘expert’ teachers (see, for example, John Hattie’s “Teachers Make a Difference, What is the research evidence?“), and in computer programming the difference between the best and the worst programmers even reaches a factor of about 10x ! (Steve McConnell, “Origins of 10X – How Valid is the Underlying Research?“) And also in the programmer case, experience seems not to play any role (though that may partially be because all programmers studied were quite experienced).

So while experience is necessary to reach top performance (as seen by the near-universal validity of a 10-or-something-year rule in many fields), there are huge differences in performance between equally experienced individuals. One could of course invoke ‘IQ’ or ‘talent’ as an explanation, but research suggests a different cause.

Take for example an investigation undertaken by the US army into the performance of its personnel, ranging from cooks to attilerymen (quoted by Earl Hunt, ‘Expertise, Talent and Social Encouragement’ in ‘The Cambridge Handbook of Expertise and Expert Performance‘). This research showed that people starting a new job indeed became better in it with time- but that they only improved the first 1 to 3 years. Then their skill plateaud. Hattie found something similar for teachers; teachers got better with experience, but most of the gain occurred in the first few years. After that, gains were minimal.

Since we know that people need ten years or more to become top performers, but most people stop improving in an activity after two or three years, the hard-to-avoid conclusion is that top-performers are not necessarily better than ‘normal’ people because they have more experience, but because for whatever reason they continue improving and learning.

Now, you may wonder: why do (future) top performers improve? Do they have special capabilities that allow them to keep improving beyond the 3-year mark?

In fact, no special capability is needed to learn longer than three years; children take over a decade to master the vocabulary and grammar of their mother tongue to a decent level. And it is also telling that in the army investigation, smarter people plateaud earlier (sometimes even within one year), even if their plateau was slightly higher that that which employees with lower IQs reached in their three years. This strongly suggests that there is no such thing as a ‘three year block’, but that there is an entirely different reason why people stop improving: they (or at least their brains) consider themselves ‘good enough’.

Consider the hypothetical case of someone living in Berlin who wants to visit Amsterdam. Not seeing road signs leading to Amsterdam, but remembering the saying that all roads lead to Rome, he decides to travel to Rome first, and from Rome get to Amsterdam. And he gets there. So it worked! His brain is satisfied.

While he is traveling again and again from Berlin to Amsterdam via Rome and back, he slowly forms a habit, meaning that neuronal connections in his brain are strengthened (go south at Bern, for example), and in the end he does not have to think about the road anymore (in psychological terms, this is called automaticity). This is nice, because he can think about other things instead, like his job, or his girlfriend.

Basically, you can say that the brain is lazy: if some course of action gets the desired result, the brain will do exactly the same thing the next time. With time and repetition, it becomes an automatism, becoming faster and faster and taking less and less effort, which is good. However, while this automation means that there will be some improvement in speed given sufficient practice, it only makes inefficient methods slightly less inefficient, it will not make one’s habits or skills great. For that, you have be taught the better method, or invent it yourself.

The true ‘bottleneck’ is that in many cases we don’t think it worth the time to investigate or experiment with alternative methods (“if it isn’t broken, don’t fix it”), and bosses or colleagues may not have the time, courage, expertise or drive to teach us better ways.

The neat results of Ericsson of hours practiced versus skill achieved are likely neat exactly because music lessons and practice are very different from a regular job: teachers are always telling students to do things better, and giving more and more challenging assignments. This is definitely not what happens in most jobs, where feedback is sparse, late or absent, and the difficulty level of tasks is not neatly ramped-up.

Why then do top-performers seem to flout the ‘3-year rule’, even if they may not have formal teachers? The answer seems to be two-fold:

a) they try to interact with top performers in their field and/or their works, whether that means having a postdoc at the group of an eminent scientist or reading/studying the works of the great people in their field, like Einstein, as a student, tried to read everything of Maxwell and Helmholz and other 19th century top physicists.

b) they spend time thinking how they can become better, or how they can overcome obstacles. And they experiment! For example, Linus Pauling (Nobel laureate in chemistry and peace), would spend 15 minutes before going to bed thinking about an important problem, hoping the answer would come to him in his sleep. Definitely this is not ‘natural’ behavior or behavior he learned at school; he very likely invented it himself to push his performance even further.

The great challenge therefore does not seem to be that the average person is physically or mentally unable to learn for longer than a few years, it is more that people are generally not motivated enough to spend the time and mental effort and take the social risks to improve beyond an ‘adequate’ level. Of course, it also does not help that for most jobs and activities there is not a clear course or program on how to continue improving, and often even motivated people can fail (witness the high failure rate of people wanting to lose weight or stop smoking). Still, lack of motivation/aspiration seems to be the root cause of premature plateauing of skill. So that may be the most suitable subject for my next post.

Is ‘talent’ just a classy name for working extremely hard?

Summary: Some people seem so incredibly good at their field, that one can understand why the Romans attributed outstanding performance to a ‘genius’, a spirit sent by the gods to inspire the fortunate individual. However, when studying the lives of prodigies and eminent persons, it becomes clear that eminence in any field takes lots of practice for even the most ‘talented’ – at least 10 years.

‘What nonsense is that?’ Tom Poes exclaimed. ‘You gave Sir Oliver the recipe for making gold, which involved molten lead, that needed to be stirred…’

‘With a stone’, the other added. ‘Stirring and stirring, around and around and around. 123,456,789 times; for it is not the formula that is hard, it is the work! After stirring 123,456,789 times the stone has become the philosopher’s stone, and only then the lead it touches turns into gold.’

Tom Poes and Roerik Omenom, ‘The lead reformer’, Marten Toonder

When watching or listening to prodigies or world class experts, it can sometimes be hard to believe that those people are mere humans. How can those people create the most wonderful music, hit the impossible ball, or develop such simple yet marvelous theories? It is no wonder the Romans explained extremely high ability with the concept ‘genius’, denoting that there must have been a special spirit, the genius, instilled by the gods at birth into the fortunate individual.

The existence of ‘prodigies’, children who are already famous performers in their field, like Mozart, Tiger Woods, or Bobby Fischer in chess seems to lend extra credence to the ‘genius’ idea. However, one should remember that prodigies are called prodigies because they perform much better in a certain field than other children. For example, Tiger Woods won his first golf competition at age 2 (the Under Age 10 section of the Drive, Pitch, and Putt competition at the Navy Golf Course in Cypress, California). However, he was not able to beat his father, who was merely a good amateur golfer, until he was 11. So while the golfing skill of Woods was extraordinary for a child his age, in absolute terms even an extremely talented toddler like Tiger was by far not yet good enough to take a serious shot at the world championships: he won his first major trophy in the adult competition at age 21, at which time he had been golfing for more than 19 years. Similarly, Mozart was indeed a child prodigy, but his first composition that is still regarded as a masterwork (instead of a pretty good work – for a 10 year old) was the Piano Concerto no. 9, written when Mozart was 21 years old. By that time, he had had composition lessons for at least 10 years. Bobby Fischer, who was not yet quite an adult when he became a chess grandmaster at age 15, had nevertheless been practicing chess for 9 years already.

In conclusion, even prodigies apparently need to put in time to reach an (adult) world-class level in their chosen field.

Researchers who looked at prodigies as well as people who were not considered prodigies but nevertheless reached great eminence in their field (for example Einstein, Darwin, Marie Curie, but also chess grandmasters, poets, writers, and composers such as Bach or Beethoven), discovered that all of them had spent at least 10 years (or about 10,000 hours) mastering their field before they produced their first masterpiece. This finding has been so consistent, that it has been dubbed the ‘10 year rule’: over a decade of practice is needed to excel in a field.

Now, ’10 years’, while memorable, is actually an oversimplification; the exact amount of time seems to depend on the starting age, the skills needed in the field (strength, aerobic capacity, intelligence, creativity, and/or motor skills), and the competitiveness of the field. In cases where there are few or no full-time practitioners, like in memory competitions, one year may already be enough to win a national championship (as described in Joshua Foer’s book ‘Moonwalking with Einstein‘). But in older, better-known fields the minimum seems to vary from about six years (for painters [Handbook of Creativity, Robert J. Sternberg] and very tall basketball players), to about 15 to 20 years for violinists. However, while the ’10 year rule’ is therefore not very accurate in a literal way, it is still very true and potent in its essence: it takes an enormous amount of effort to reach the top in a field, much more than most people are willing or able to expend. If you merely think “wouldn’t it be nice to be a Nobel-prize winning physicist/wealthy rock musician/world famous tennis player” you are very unlikely to have the perseverance or be willing to make the sacrifices needed to attain that level – because the ‘reward’ is a very, very long way off.

Of course, while these data strongly suggest that a large amount of work is necessary to reach a high level in any activity, it does not say that it is sufficient. Perhaps it is necessary to put in all that time to reach world-class levels, even if you are talented or even a genius. But that does not automatically mean that anyone who spends 10 years doing or practicing something will become incredibly good at it. You may need both talent and time: without talent, time spent practicing may not have much effect, or at least probably won’t make you rise above mediocrity.

Fortunately, such a hypothesis can be investigated scientifically; one could in principle (if one can clearly quantify skill) try to compare the time needed by different people to reach a certain level of skill. Such a kind of research has for example been performed by John Sloboda and colleagues for musical achievement (Sloboda, J.A.; Davidson, J.W.; Howe, M.J.A.; and Moore, D.G. “The Role of Practice in the Development of Performing Musicians”, British Journal of Psychology 87 (1996) p287-309.) This investigation (as did investigations before) found no evidence for ‘giftedness’ in music; children, whether considered ‘talented’ or ‘untalented’ took approximately equal amounts of time to reach a certain level in musical skill. The ‘talented’ just practiced more hours and therefore seemed to progress faster. An earlier study by K. Anders Ericsson and colleagues at a conservatory (in Ericsson, K.A., Krampe, R.T., and Tesch-Römer, C. “The role of deliberate practice in expert performance“. Psychological Review, 100 (1993) p363-406) found that the best violinists had had about 7,500 hours of solitary practice at age 18, while the ‘merely’ good violinists had on average 5,300 hours, and the future music teachers who studied at the same institution 3,400 hours. There were no clear cases of ‘drudges’ who had studied a lot but with little to show for it, or ‘geniuses’ who had reached the highest rungs with relatively little effort.

So next time you feel a small pang of envy when hearing about a successful artist, musician or scientist, you may at least find some comfort in the thought that that person probably had to work extremely hard to get to that point. However, while Ericsson may not have found drudges in his study of music students, in daily life there do seem to be people who work hard, but never get very good at what they do. How can we explain that?

But that would be an excellent subject for my next post.