by David Robson
The latest neuroscience, however, shows that we learn best when we are confused; deliberately limiting your performance today actually means you will perform better tomorrow. And a failure to recognise this fact is another primary reason that many people – including those with high IQs – often fail to learn well.
One of the earliest studies to show this phenomenon was commissioned, strangely enough, by the British Post Office, which in the late 1970s asked the psychologist Alan Baddeley to determine the best methods to train its staff.*
* Baddeley, incidentally, is also the reason that British postcodes contain around six or seven characters – which is the maximum span of human working memory. He also advised on the specific positioning of the numbers and the letters to make them as memorable as possible.
The Post Office had just invested heavily in machines that could sort letters according to postcodes, but to use these machines, their 10,000 postal workers had to learn how to type and use a keyboard, and Baddeley’s job was to discover the most efficient schedule for training.
The assumption of many psychologists at the time had been that intensive training would be better – the postmen should be allowed to devote a few hours a day to mastering the skill. And this was indeed the way that the workers themselves preferred: they were able to see real progress during that time; at the end of their sessions, their typing felt much more fluent than at the beginning, and they assumed this carried over into long-term memory.
For comparison, however, Baddeley also created a few groups that learnt in shorter stretches, over a longer period of time: just one hour a day, compared to four. The workers in this group didn’t seem to like the approach; they lacked the sense of mastery at the end of their session and didn’t feel like they were progressing as quickly as those taking the longer sessions.
But they were wrong. Although the sessions themselves felt unsatisfying compared to those of the people who developed more proficiency in a single day, these subjects ended up learning and remembering much more in relation to the amount of time they were putting into it. On average, a person with the ‘spaced’ approach mastered the basics within thirty-five hours, compared to fifty hours for the intensive learners – a 30 per cent difference. Individually, even the slowest person in the one-hour-a-day group had mastered the skill in less time than the quickest learner among those devoting four hours a day to the test. When the researchers followed up their studies a few months later, the spaced learners were still quicker and more accurate than those learning in blocks.3
Today, the spacing effect is well known to psychological scientists and many teachers, and it is often represented as demonstrating the benefits of rest and the dangers of cramming. But the true mechanism is more counter-intuitive, and hinges on the very frustration that had annoyed the postmen.
By splitting our studies into smaller chunks, we create periods in which we can forget what we’ve learnt, meaning that at the start of the next session, we need to work harder to remember what to do. That process – of forgetting, and then forcing ourselves to relearn the material – strengthens the memory trace, leading us to remember more in the long term. People who learn in longer blocks miss out on those crucial steps – the intermediate forgetting and relearning – that would promote long-term recall precisely because it is harder.
In this way, Baddeley’s study gave some of the first hints that our memory can be aided by ‘desirable difficulties’ – additional learning challenges that initially impair performance, but which actually promote long-term gains.
The neuroscientists Robert and Elizabeth Bjork at the University of California, Los Angeles, have pioneered much of this work, showing that desirable difficulties can be powerful in many different circumstances – from maths to foreign languages, art history, musical performance and sport.
Consider a physics class revising for an exam. In the West, it’s common for teachers to present the principles and then to get students to repeat endless series of similar questions until they reach nearly 100 per cent accuracy. The Bjorks have shown that, in reality, learning is more effective if the student solves just enough problems to refresh their mind, before moving on to a new (perhaps related) subject, and only later should they return to the initial topic.
Like the spacing effect, this process of switching between tasks – known as interleaving – can lead the student to feel confused and overburdened, compared to lessons in which they are allowed to focus solely on one subject. But when they are tested later, they have learnt much more.4
Other desirable difficulties include ‘pre-testing’ or ‘productive failure’ – in which students are quizzed on facts they haven’t yet learnt, or given complex problems they don’t yet know how to solve.
You can see it for yourself. Without looking at the answers below, try to match the following Italian words to the English terms.
• I pantaloni
• Tie
• L’orologio
• Trousers
• La farfalla
• Boot
• La cravatta
• Bowtie
• Lo stivale
• Watch
Now look at the footnote below to see the answers.*
* I pantaloni – trousers
L’orologio – watch
La farfalla – bowtie
La cravatta – tie
Lo stivale – boot
You might be able to guess a couple, perhaps from their similarity to English or French terms, or from previous exposure to the language. But the surprising thing about pre-testing is that it doesn’t matter if your initial answers are right or if you really don’t have a clue; it’s the act of thinking that will boost your learning. Like the fruitful forgetting that comes with spaced study, the frustration we feel at not understanding these terms leads us to encode the information more deeply, and even if you are not normally very good at learning languages, you should find that this vocabulary ultimately sticks in your mind.5
(This was, incidentally, the reason that I asked you to guess the truth of those statements at the start of this chapter. By causing you to question your existing knowledge, it should enable you to remember the subsequent information far more clearly.)
Productive failure seems to be particularly fruitful for disciplines like maths, in which teachers may ask students to solve problems before they’ve been explicitly taught the correct methods. Studies suggest they’ll learn more and understand the underlying concepts better in the long run, and they will also be better able to translate their learning to new, unfamiliar problems.6
The introduction of desirable difficulties could also improve our reading materials. Textbooks that condense concepts and present them in the most coherent and fluent way possible, with slick diagrams and bullet point lists, actually reduce long-term recall. Many students – particularly those who are more able – learn better if the writing is more idiosyncratic and nuanced, with a greater discussion of the potential complications and contradictions within the evidence. People reading the complex prose of Oliver Sacks remembered more about the visual perception than people looking at a slick, bullet-pointed textbook, for instance.7
In each case, we can see how the elements of confusion that our whole education system is geared to avoid would lead to more profound thinking and learning, if only we let students feel a bit frustrated.
‘Current performance is a measure of accessibility [of information] right now, but learning is about more fundamental changes that will be reflected after a delay, or a transfer of this learning to somewhere else,’ Robert told me when I met him and his wife in the faculty centre of UCLA. ‘So if you interpret current performance as a measure of learning, you will get lots of things wrong.’
Scientifically speaking, these results are no longer controversial. The evidence is now unarguable: introducing desirable difficulties into the classroom – through strategies such as spacing, interleaving and productive failure – would ensure that
everyone learns more effectively.
Unfortunately, it is hard to persuade people to appreciate these effects; like Baddeley’s postal workers, students, parents and even many teachers still assume that the easier you find it to learn something today, the better you will perform tomorrow, even though these assumptions are deeply flawed. ‘We have all these results showing that people prefer the poorer ways of learning,’ Robert added.8 ‘So you’re not going to make your students happier right away.’
Elizabeth agreed. ‘They interpret [confusion] negatively, as opposed to thinking that this is an opportunity to learn something or understand it in a better way.’
It’s as if we went to the gym hoping to build our muscles, but then only ever decided to lift the lightest weights. The Bjorks have found that these ‘metacognitive illusions’ are surprisingly resilient even after people have seen the evidence or experienced the benefits for themselves. The result is that disappointingly few schools try to make use of desirable difficulties, and millions of students are suffering as a result, when they could be learning much more effectively, if only they knew how to embrace confusion.
At least, that’s the view we get from the USA and the UK.
But as we have seen previously, we should be wary of assuming that the biases documented among ‘WEIRD’ countries represent human universals, when studies of East Asian cultures often show very different attitudes.
Rather than us feeling that learning should come easily, various surveys – including Stigler’s own research – show that students in countries such as Japan appreciate that struggle is necessary in education. If anything, the students in these cultures are concerned if the work isn’t hard enough.
These notions can be found in the parents’ and teachers’ attitudes, in popular sayings such as ‘doryoki ni masaru, tensai nashi’ (‘even genius cannot transcend effort’) and in the country’s folk stories. Most Japanese schoolchildren will have heard the story of nineteenth-century scholar Ninomiya Sontoku, for instance. As a poor young boy, he was said to use every opportunity for study, even when collecting firewood in the forest, and many schoolyards still contain statues of Sontoku with a book in front of his nose and firewood on his back. From earliest infancy, children in Japan are immersed in a culture that tolerates struggle and challenge.
Crucially, this acceptance of struggle extends to their mindsets about their own talents: Japanese students are more likely to see their abilities as a work in progress, leading to the growth mindset. ‘It’s not that the Japanese don’t believe in individual differences,’ Stigler told me. ‘It’s just that they don’t view them as limitations to the extent that we do.’ As a result, an error or mistake is not seen as a sign of some kind of permanent, inevitable failure – it is, according to Stigler, ‘an index of what still needs to be learned’.
These beliefs help to explain why East Asian students, in general, are willing to work longer hours, and even if they aren’t naturally gifted, they are more likely to try to make up for that by studying harder. Just as importantly, however, these beliefs also have consequences for the ways the teachers handle their lessons, allowing them to introduce more desirable difficulties into the curriculum. The result is that their pedagogical methods regularly embrace confusion to enhance learning and comprehension.
When tackling a new topic in maths or science, for instance, it’s quite common for Japanese teachers to begin their lessons by asking students to solve a problem before they’ve been told the exact method to apply – the use of ‘productive failure’ that we discussed a few pages ago. The next couple of hours are then devoted to working their way through those challenges – and although the teacher offers some guidance, the bulk of the work is expected to come from the students.
American or British students baulk at the confusion this produces – and when they begin to struggle, the teacher would be tempted to give in and tell them the answer.9 But Stigler found that Japanese students relished the challenge. As a consequence, they think more deeply about the underlying characteristics of the problem, increasing their ultimate understanding and their long-term recall.
And when working through problems, they are also encouraged to consider alternative solutions, besides the most obvious one, and fully to explore their mistakes (together with those of their classmates) to understand how one approach works and another doesn’t. From a very young age, they are encouraged to take a more holistic view of problems, and to see the underlying connections between different ideas. As one elementary maths teacher in Stigler’s studies put it: ‘We face many problems every day in the real world. We have to remember that there is not only one way we can solve each problem.’
British or American schools, in contrast, often discourage that exploration, for fear that it might cause extra confusion; for each type of maths or science problem, say, we are only taught one potential strategy to find a solution. But the scientific research shows that comparing and contrasting different approaches gives a better understanding of the underlying principles – even if it does result in more confusion at the beginning.
‘They’ve created a classroom culture that tries to support the prolonged extension of confusion,’ Stigler said. ‘The Japanese feel that if you can sustain that in the classroom, students will learn more. Whereas we’re very focused on [simply] getting the answer – and if you want the students to get the right answer, you then make it as easy as possible.’
In this light, our reactions to Stigler’s story of the Japanese boy struggling to draw 3D cubes at the blackboard make more sense. Whereas Americans, or Brits, would see the boy’s initial troubles as a sign of weakness or stupidity, his Japanese classmates saw his perseverance. ‘His errors were not a matter of great concern; what would be worrisome would be the child’s failure to expend the effort necessary to correct them.’ The boy didn’t cry, as Stigler had initially expected, because in that cultural context there simply wasn’t any reason to feel the same level of personal shame that we would expect.
On the contrary, Stigler says that as the lesson went on, he felt a wave of ‘kindness from his classmates and the teacher’. ‘No one was going to cut him a break to say that was good enough until he’d accomplished the job. But on the other hand, you had the sense that they were all there to help him,’ Stigler told me. They all knew that the struggle was the only way the boy was going to learn and catch up with his peers.
It is true that some lessons do involve elements of rote learning, to ensure that the basic facts are memorised and easily recalled, but this research shows that Japanese classrooms provide far more room for independent thinking than many Western commentators have assumed. And the benefits can be seen not only in the PISA scores, but also in tests of creative problem solving and flexible thinking, in which Japanese students also outperform the UK and USA, demonstrating that these students are also better equipped to transfer their knowledge to new and unexpected tasks.10
Although some of the strongest evidence comes from Japan, the value of struggle does seem to be common to other Asian teaching cultures too, including Mainland China, Hong Kong, and Taiwan. In Mandarin, for instance, the concept of chiku or ‘eating bitterness’ describes the toil and hardship that leads to success. And Stigler has since expanded his focus to explore other countries, such as the Netherlands, which has also outperformed the USA and UK. Although they may vary on many factors, such as the class size, or the specific methods that the teachers use to deliver material, the top-performing schools all encourage students to go through those periods of confusion.
Stigler has now been researching these ideas for decades, and he suggests his findings can be distilled to three stages of good teaching:11
Productive struggle: Long periods of confusion as students wrestle with complex concepts beyond their current understanding.
Making connections: When undergoing that intellectual struggle, students are encouraged to use comparisons and analogies, helping them to see underlying patterns between
different concepts. This ensures that the confusion leads to a useful lesson – rather than simply ending in frustration.
Deliberate practice: Once the initial concepts have been taught, teachers should ensure that students practise those skills in the most productive way possible. Crucially, this doesn’t involve simply repeating near identical problems ad nauseam, as you might find in the Western maths classroom, but means adding additional variety and challenges – and yet more productive struggle.
These are profound discoveries that offer some of the most robust ways of improving education and academic achievement, and in a few pages we’ll see how anyone can make use of desirable difficulties to master new skills.
But these findings are not only interesting for what they tell us about human memory; I believe that they also reveal some profound insights into the cultural origins of the intelligence trap.
If you consider classrooms in the UK and USA, for instance, our mental worth is often judged by who can put their hand up quickest – giving us the subtle signal that it’s better to go with an immediate intuitive response without reflecting on the finer details. And you are not going to be rewarded for admitting that you don’t know the answer; intellectual humility is actively discouraged.
Worse still, the lessons are often simplified so that we can digest the material as quickly as possible – leading us to prefer ‘fluent’ information over material that might require deeper consideration. Particularly in earlier education, this also involves glossing over potential nuances, such as the alternative interpretations of evidence in history or the evolution of ideas in science, for instance – with facts presented as absolute certainties to be learnt and memorised.12 The assumption had been that introducing these complexities would be too confusing for younger students – and although the teaching methods do allow more flexibility at high school and university, many students have already absorbed a more rigid style of thinking.
