The thrust of the book and its many examples is that specialization in a narrow field is demonstrably harmful to overall progress and performance in many contexts.
Comparison of Tiger Woods and Roger Federer as 2 completely different approaches to success /elite level success. Tiger was groomed to play golf more or less to the exclusion of all else by his Father from the age of 2. Roger participated in many sports and pursuits and only really got serious about tennis in his late teens.
In general early specialization is a poor predictor or pre-condition for top level success early sampling and diversity are in fact key.
Specialization and learning work in a simple environment where the variables are limited and constrained – many sports are like this. Bounce by Matthew Syed and Malcom Gladwell’s Outliers book have both help to create and perpetuate an overly simplistic view of what goes on with deliberate practice and specialization.
To accumulate lasting knowledge learning is best done slowly even at the detriment of short term test scores which in fact often mask an incomplete grasp of material. In fact the best learning often looks inefficient.
‘The challenge we all face is how to maintain the benefits of breadth, diverse experience, interdisciplinary thinking, and delayed concentration in a world that increasingly incentivizes, even demands hyper-specialization.’
Lazlo Polger bought up his daughters with the specific intention they would be chess prodigies. Planning this even before he was married!
Psychologist Gary Klein is the pioneer of the naturalistic decision making model. In essence his observations show that expert performers have exceptional pattern matching skills – Klein studied fire-fighters and observed how often key decisions were made within seconds – honed by years of experience.
Chess and the Polger sisters exemplify how this works in a relatively narrowly constrained arena. Expert chess players learn and see patterns.
Daniel Khanemen in his research and work sees almost the opposite in decision making. In many contexts we are predisposed by bias and the use of heuristics (mental short cuts in our decision-making) to making poorly calibrated decisions that often discount or ignore swathes of relevant information.
Expertise in a decision making area often bred confidence but not skill! Chess gold and fire-fighting are exceptions to the general decision making arena not the rule. These can be described as kind learning environments because patterns repeat over and over. In complex or ‘wicked’ domains the rules of the game are unclear or incomplete feedback is often delayed, inaccurate or both.
Human intelligence we are learning from our search to build AI and game playing software is our ability to integrate broadly. In a closed world the machines win, when things are open ended humans do. AI are like savants (genius in a very narrow field) they need stable structures and narrowly defined parameters with clear rules.
In many domains in the real world, we have to find a way to learn beyond practice. The most creative achievers in many fields tend to have broad interests and experiences (Connect to work on Social physics by Sandy Pentland and decision making / forecasting by Tetlock)
Whether experience leads to expertize depends entirely on the domain. In any complex and ambiguous environment (most scenarios!) expertise and experience correlate poorly.
Creative achievers tend to have broad interests. Their skill is in avoiding the same old patterns.
Flynn Effect named after James Flynn a professor of political studies the effect is the fact that mean IQ scores have increased in each generation of the 20th
Today’s children are far better at solving problems on the spot without a previous learned method and are far better with abstract words such as law, or citizen.
This appears to be a product of the development of the modern world. Pre-modern villagers (eg experiments carried out in 1930’s Soviet Union) were perfectly capable of learning from experience but failed at learning without experience.
Have to be taught how to think before we can figure out what to think.
Case study of the Ospedale della Pietà was a convent orphanage music school in Venice. Virtosuo performers known as the Figlie – 18th Century Italian musicians who had renowned virtuosity and played multiple instruments. Many of them were foundlings bought up in the care of the state – as music was a method of paying the bills.
Their development as musicians was marked by a very wide ranging sampling period where they were encouraged to try and experiment with as many instruments as possible with little formal instruction.
Musicians who are best at improvisation have not tended to come through a rigorous deliberate practice process but something more osmotic and playful. Dive in, imitate learn the rules later – interesting to note this is exactly how babies and small children learn.
Creativity may be difficult to nurture but it is easy to thwart. In households with extremely creative children there appears to be 1 rule where in an average household there are 6.
For learning that is durable and flexible and can therefore be applied broadly learning fast and easy is a problem. That can of learning needs ‘desirable difficulty’ obstacles that make learning more challenging, slower and frustrating in the short term.
Excessive hinting (typical teaching process) bolsters immediate performance but undermines long term progress.
Desirable difficulties (Term coined by Psychologist Robert Bjork in ’94) include, having to generate answers of your own (even if its wrong, in fact some evidence suggest the more wildly wrong the better in terms of subsequent learning of the correct answer). Spacing. Short term rehearsal gives short term gains, testing after longer gaps. ‘Learning is most efficient in the long run when it is really inefficient n the short run’. Frustration is not a sign you are not learning but ease is. Testing.
Don’t interpret current performance as learning.
‘increasingly jobs that pay well require employees to be able to solve unexpected problems often while working in groups’. Greg Duncan Education economist. Necessity to be able to think outside of experience.
Interleaving improves inductive reasoning – different examples mixed together students learn to create general abstractions that allow them to apply learning to material they have not encountered before. This approach more efficacious that learning in blocks.
Johan Kepler example – his observations and mathematics redefined how we understand the universe.
He used analogies to help progress his theoretical thinking – what is the nature of the invisible force the sun appears to exhibit across the solar system. Thought of other invincible but observable phenomena such as magnetism and smell. He eventually figured out that all bodies not just the sun exert a force – this explains why some planets have an elliptical orbit as their path is impacted both by the gravity of the sun and another planet. He correctly believed that the moon was responsible for tides on earth.
Analogical thinking is the practice of recognizing conceptual similarities in multiple domains that at face value seem to share little in common. It is a powerful tool for solving wicked problems. With wicked problems single domain experience is too limited and can be disastrous.
Example of this the riddle of the cancer curing ray which if powerful enough to destroy the tumour will also destroy intervening flesh. If weak enough not to destroy healthy tissue also not powerful enough to destroy the tumour. Paired with analogous stories of how a general divided his troops a solution becomes more apparent.
2001 Boston consulting group created an intranet site to give consultants collections of material to facilitate wide-ranging analogical thinking. Sorted by discipline, anthropology, psychology, history etc and then by concept, change logistics, productivity etc and strategic theme competition, merger, innovation etc gave lots of thought provoking analogous ideas, how did William the Conqueror merge Normandy with England etc.
Successful problem solvers are better able to discern the deep structure of a problem before they try and match a strategy to it. Less successful people see only the surface level superficial characteristics.
A problem well put is half solved.’ John Dewey logic, the theory of enquiry.
Investigation of the worlds best and most successful research labs shows certain characteristics. When they get observations that are unexpected they don’t assume the measurement is wrong (and therefore the theory is correct) but collaboratively try to figure out what might be going on – often by generating a lot of analogies. Breakthroughs very often came with input or ideas from out of the problem lane.
Looks into Angela Duckworth research on grit – passion and perseverance with the counter-point of view that you can have too much grit.
1 problem with this research is that the study groups were already selected, west point cadets are not a normal population. Take a sports analogy if you looked at success in basketball from a pool of NBA players it might not show that height is an important predictor of success (because NBA teams already selected for height) so in this restricted population it appears less important than other possible factors.
Duckworth herself acknowledges this by pointing out that by studying highly pre-selected groups we have necessarily limited the external validity of our investigations.
Perseverance in the absence of common sense and good judgment is not a desirable quality. Too often we are hampered by the sunk cost fallacy and throw good money or time after bad. Have to know when to quit and change direction. Far better idea to flirt and experiment with your possible selves to find good match fit but too often obliged to specialize early and then get trapped by the investment into sticking – probably miserably!
Defence Sec Ash Carter visiting West Point in 2016 was flooded by cadet concerns that the career path was far to rigid – match quality is therefore often poor, what you expect you will enjoy and be good at and what you actually enjoy and are good at are often different. Experimentation and test and adjust very important. 50% of West Point Grads leave at the earliest point post return on service at 5 years.
Personality changes more than we expect with time, experience and context. We learn who we are only by living – the more we sample activities, social groups, jobs and reflect and adjust our personal narratives the better we find our fit. Reflection is not enough there must be action 1st since we are all a host of possibilities which we discover by doing. Test and learn rather than plan and implement.
The outsider advantage.
Alph Bingham VP or R&D Strategy at Eli Lilly (US Pharma multi-national) he collected 21 problems that had stumped researchers and put them online to solicit input from other domains. It worked so well the idea was eventually spun out as a separate company named InnoCentive – this facilitates seekers paying to post challenges and rewards for outside solvers.
NASA had a problem they had not solved in 30-years, how to predict solar particle storms. Within 6 months of posting it was solved by a retired engineer who suggested using radio waves picked up by telescopes.
1989 Exxon Valdez oil spill. One of the most intractable problems with spill remediation was pumping oil out of recovery barges after its skimmed off the water. The solution was started with the analogous thought of how to drink a slushy - have to keep stirring the straw around. Then remembered working on a construction site and seeing a concrete vibrator which keeps concrete fluid so it wont set till in place by vibrating the mixture so the elements don’t coalesce. Its cheap and works in the oil barges perfectly.
Specialists get too narrowly focused – in fact they are like Russian dolls as they divide further and further into sub-specialists. Knowledge is a double edged sword is allows you to do some things but makes you blind to other things you could do.
Gunpei Yokoi employed by Nintendo – he had long been an enthusiastic hobbyists, piano, dancing, diving, model trains. He made Nintendo’s 1st toy – basically because the Company President found him playing around and turned it into Ultra hand which sold over a million units.
Yoki built many toys and then games and developed a philosophy of new uses for old ideas – he talked about withered technology, old so it was extremely well understood, cheap and simple – many toys and games later Nintendo bought out the game boy which has sold over 118 million units worldwide and is by far the best selling console of the 20th ‘Lateral thinking with withered technology’.
Freeman Dyson eminent physicist and mathematician, ‘we need both focused frogs and visionary birds’. Birds have a broad view over the horizon frogs are down in the weeds.
T shaped people have depth in 1 area but also a lot of breadth. I shaped people just narrow depth. Well defined and well understood problems good for specialists. The more ambiguity and uncertainty there is (normal in any systems type problem) breadth is increasingly important.
3M’s systematic investigation of the ingredients of invention, innovative teams and individuals found it is often to do with the ‘adjacent stuff’
When the National Transportation Safety Board analysed its database of flight accidents if found that 73% happened on the 1st day of a flight crew working together!
Fooled by experience. Looks at the work of psychologist and political scientist Philip Tetlock (For detail see his book Superforecasting here )
His research shows most ‘experts’ make terrible predictions – no better than random in statistical terms. Uses analogy of hedgehogs and foxes for experts and generalists.
A hallmark of the interaction of the best teams is ‘active open mindedness’ Psychologist Jonathan Baron. They are better forecasters and have better judgment because they test out rather than seek to prove a hypothesis.
Balancing the risk of mindless conformity and reckless deviation. Best thinkers are able to tolerate and work within high levels of ambiguity.
Examination of NASA culture especially around the 2 fatal accidents with Challenger and Columbia. Compares original culture under Wernher Von Braun which balanced NASA’s rigid process with an informal individualistic culture that encouraged dissent and cross boundary communication. After his tenure this was lost and the culture became command and control rules process etc. This pushed conflict into corners and underground and engineers who did not have the data to prove something kept quiet – (Challenger!)
There is a difference between chain of command and chain of communication which should be a healthy cross pressure.
When Bill Gore left Du Pont to set up Gore Tex he built it with the observation that most companies do their most impactful and creative work in a crisits because the disciplinary boundaries go out the window. He made sure that ‘dabble time’ was a cultural staple.