Range: Why Generalists Triumph in a Specialized World 2/5

David Epstein

Part 1

3/ When Less of the Same is More

• Many parents want certain outcomes for their kids while completely ignoring the means to get to those outcomes. This includes not just the endless hours of practice, but (less known) experimentation in the areas where kids will not succeed, but which have to be explored nonetheless. [MK: and when the kids turn 16, they run away from home to avoid the abuse.]

• The breadth of training predicts the breadth of transfer, i.e. the more abstract models the person gets, the more tools they have to rely on in cases of uncertainty.

4/ Learning, Fast and Slow

• In cases of uncertainty people are seeking rules, not concepts, to take shortcuts (see above) – this is painfully true for kids’ development [MK: as a father of two as an ex-tutor to 3 more I can confidently say this is true]. But when it comes to learning concepts, experience can backfire. It becomes an impediment to grasping the true concepts. [MK: you may recall some higher-up saying: this is simple, we’ve always done it this way; or, don’t overcomplicate things, the right solution is …]

• Problems can be split into “making connections” (building on applying concepts) and “using procedures” (applying routine algorithms). The schooling system is leaning towards the “toolbox” approach, i.e. teaching kids the set of procedures to apply. Systems thinking requires the opposite: the speed of solving problems is less important than the number of ways to approach the problem to solve, no matter the time.

• Doing well in class most of the time means solving problems fast – and this is a trap, since the speed is the result of the effectiveness, which relies on algorithms, which rely on standardized tasks, which are artificial and don’t exist in real life.

• Generating answers and strategies (even the wrong ones) helps learning and understanding; tolerating big mistakes presents the best learning opportunities. Helping students with answers doesn’t help them in the long run.

• Going topic after topic after topic (i.e. block practice) sounds very logical, but wrong. By the time the year ends the successful % of recall drops dramatically unless there’s a forced recall (i.e., a test), say, 1 month after the presentation of the material. (It’s called the “spacing effect”). The alternative is a “mixed practice”, i.e. the topic is approached from different angles at different times.

• The best learning is slow and almost invisible. Unfortunately, the teachers who are popular tend to be useless in the long run, and the teachers who challenge students and (reasonably) give them poor grades account for much of the learning success in subsequent subjects. [MK: one can think of the boot camps for a very different reason but the same kind of outcome.]

• The current schooling system has just realized that the real-world problems are loosely defined, and that university graduates no longer have routine-based well-paid middle-class jobs ahead of them. They keyword here is “routine-based”, as such jobs are increasingly getting automated or outsourced to gig economy workers.

• [MK: Practice doesn’t mean perfect; it means functionally stupid.]

• Repetitive environments are very good for block practice, but getting better is a result of mixed practice, even though students only understand it (maybe, if) in hindsight.

• “Desired difficulties” make knowledge stick as they forcibly make mental connections outside the routine / procedural thinking.

• Early children education programs [MK: if you have kids, you know all too well to listen to people boasting the achievements of their dear babies who speak foreign languages at 3, play golf at 4 and become chess masters at 5] are based on “closed” skills that are practiced via routine. It makes no difference if the child starts reading at 3 [MK: that’s me] or 6, starts talking at 1.5 or 2.5 or starts drawing somehow at 2 or 5. By the age of 7-8 most kids’ abilities are synchronized (which you can’t say about their parents’ ambitions, though). [MK: It’s very easy to get yourself a child prodigy; doing so sets the child to the dangerous path of peaking early and going downhill lacking the understanding of what’s really required for proper development.]

• What’s really important is the development of complex skills taking much longer to reliably surface. It’s the job of a good teacher or parent to ensure the child walks the entire path without jumping off it.

• Proper knowledge takes a lot of time to develop. It’s frustrating, but it’s OK. That’s the painful path to being a smart generalist.

Part 3.