I am very doubtful of the suggestions that those arguing against testing and spaced repetition are just not trying to help people grow.
My 6 month old is currently learning a whole host of things, and we are doing repetition and watching his growth -- so no disagreement on the premise that these things are good.
But we are also making great pains to ensure that he is playing, is always happy doing new things, and doesn't have to continue if he gets overwhelmed and/or bored. We do this because the mentality of someone when learning is a huge factor in what they get out of the activity.
I had the impression that the current push against testing and repetition is because it ended up with children hyper-focussed on "passing the test" skills but not spending the time to question things too deeply and gain that understanding that actually helps the workforce (like Feynman noticed in Brazil).
There is a push, which has been around for decades, to measure how effective teaching is for a school system taking in taxpayer dollars. This is measured via standardized tests. The tests are generally passed via testing and repetition. Often this becomes the metric for measuring teacher, principal, superintendent performance and school funding. So these authorities begin having the students memorize, do repetition and testing - and testing not to guide a student's progress, but to reward or punish them. The school authorities have the students due this to pass the standardized tests, at the behest of those who have authority over the school authorities and want measurement.
So then the question is, is this the best way to learn in general? You can read studies of education including John Dewey's from over a century ago to see that it is not. The purpose of the education system is not to educate, but to do this thing described in the first paragraph. Nowadays the public schools are contending with charter schools, vouchers and the like, so there's a more injection of profit, religious fundamentalism and the like in the educational taxes people pay than there was a few decades ago.
On the other hand we've butchered standardized testing so hard that people report high school graduates as useless. While testing may result in test-focused learning, not testing seems to result in... not learning at all. In the end it's hard to teach someone who's not interested and we have to decide whether we want to force them.
Ah, I met many Ivy League kids, and I got the impression that while they were smart, they were moreso neurotic. (That is, unless their parents were rich!) Why should we let our society be run by a bunch of neurotic control freaks dead set on hitting targets? A place no different from Hell!
Failure means the ride is coming to an end. Even a small failure could completely derail your dreams of being a great scientist, economist, or the next Zuckerberg. Unless your parents are rich, of course. And if they're not rich, you may be carrying the honor of your entire family back in the Old Country on your shoulders, which honor will be permanently tarnished by your failure.
So yeah, if you're in the Ivies and not from old money, you've got one shot and you'd better get it right.
> But we are also making great pains to ensure that he is playing, is always happy doing new things, and doesn't have to continue if he gets overwhelmed and/or bored.
Obviously I can't read into the nuances here, but I guess, there is this risk of overdoing it.
I get what this piece is trying to say, but it's ignoring the fact that schools are trying to maximize learning with pupils who often don't want or care about learning (unlike with athletes or musicians who are generally learning their craft by choice).
A significant part of teaching disinterested students (not just in a grade school but in general) is about making the subject interesting enough that students will want to spend time on learning and continue to delve further in their free time.
If you're trying to teach someone web development, would you have them churn through a stack of predetermined bootcamp-style projects, or would let them try to build something they have personal interest in? I bet the latter method would turn out much better for the student in the long run.
As the gymbros say, motivation doesn't get results. Discipline does. Few children are going to be interested in learning all that they need to learn. Esp. when it comes to math. So you need to instill the discipline in them to do it -- good work and study habits, drilled into them until they... actually become habit.
That's the problem with, for example, A Mathematician's Lament. Lockhart is looking at the problem from the perspective of a seasoned mathematician, not a primary schooler without the requisite skills. He only got where he was in the field by memorizing his times tables and practicing elementary proofs until he could do them in his sleep. Only then, after having done the boring stuff, could he even begin to perceive the beauty and art in mathematics.
Gymbros are wrong. The hypefocus on effectivity is what stands in the way of healthy lifelong exercising. If your whole exercise regime is based on discipline, it will fail entirely the moment you have other stressors in the life. Because then it becomes obstacle to what you need rather then something that helps you be happy.
Gymbros are gymbros because gym is their priority 1, over everything else. They are already motivated. You know who used to actually end up exercising regularly? Guys who would meet for soccer game regularly so that they meet friends. And people who actually like the sport they do.
I think it depends on the type of motivation, when I think back to my time at high school (Australia in the 90's) there was a contrast between how English was taught and how Math was taught.
In my English class the teacher would assign the class a book, or a poem etc. Take this home and read to the end of chapter X before next class. At the start of next class the teacher would pick half a dozen random students and ask them questions in front of the class about what we had been assigned to read. These weren't the kind of questions you could bluff an answer to.
Believe me you were motivated to do the readings because no one wanted to get called up to the front of the classroom and look like an idiot by not being able to answer the question. You were motivated by fear.
In Math on the other hand we were given a textbook, told to go home and do exercises from the book to practice what we'd been taught. It entirely ran on the honor system no one checked to make sure we did the exercises, as a result I know a large portion of the class didn't bother. I wonder what would have happened if the math teacher were to have called up random students to front of the classroom and made them solve a problem on blackboard at start of each lesson.
This article is a bit terse, but at least it hits one thing: math is a different subject than most, math is a skill - just like a sport or learning an instrument (or learning a foreign language, for that matter) and mastering a skill is different than acquiring knowledge about a subject.
The problem is we don't point out that difference to students and inform them how that difference should change their approach. For example, I told my kids that math is a skill, and like all skills, practice is required for mastery. It's not math homework, it's math practice. That's not a cutesy statement, either. They understood that even if they had no homework assigned, they should still spend a few moments practicing their math. That approach paid-off in spades!
> math is a different subject than most, math is a skill
Interesting framing. It prompted me to wonder, what aspects of science education are taught as a skill, are what aren't but could/should be?
Scientific argumentation/dialog (part of science literacy) seems occasionally taught as a skill. Perhaps roleplay, reflective practice, and drill, are then underutilized?
An order-of-magnitude sense of size is regrettably very not taught as a skill. So despite being taught primary through undergraduate, and being part of most every science and engineering curricula, it's very little used by other content, and even graduate student grasp is often weak. Teaching a sense of size as a skill, as a framework for organizing one's knowledge of the physical world, might be an interesting opportunity.
Much of intro physics seems taught as plug-and-chug skills. So being a "skill" seems at least somewhat orthogonal to transferability, utility, etc. Is applying physics to your everyday life ever taught as a skill? Or an experimentalist mindset?
> Is applying physics to your everyday life ever taught as a skill? Or an experimentalist mindset?
I was a physics prof for a while. The answer is Yes and No. Good teachers and good programs try to do this.
The No emerges from the fact that education is a social process between humans. "mindset" is how you think. You can't force someone to change the way they think unless they want to. You can tell them how to think, you can provide conditions in the class or lab to encourage this, but at the end of the day, for adults at least there are limits to how much you can push them [1].
Some/most but not all students will agree to change their mindset, but usually only if they like or respect the teacher or subject enough. If a student is taking 5 courses in a semester, they likely won't have the mental bandwidth to change how they think as pushed by each of those 5 courses/teachers.
As a prof, you have to respect these human limitations. You can't just design a course where you expect it is the only thing the students are doing, and that they already love/respect you enough.
PhD students, on the other hand, are rightfully pushed quite hard by their supervisor to think correctly. The social dynamics are quite different there.
[1] School children, especially young ones, for better or worse, are pushed much more by their school teachers because their minds are more malleable.
My kids weren't willing to practice either musical instruments or math. Some people just aren't intrinsically motivated that way, regardless of how you inform them or explain the benefits.
I thought this was going to be about the no-child-left-behind mindset, which results in teachers spending most of their time focusing on the slowest learners and very little time on students who are already at/above grade level. That mindset is certainly common in our school, and learning for the latter students is closer to flatlined than maximized.
Maximizing learning via SR is interesting, but it seems like it's the furthest thing from our kids' teachers' minds.
That's a fair point about what you expected to read. That said, while I agree that "teach to the slowest learners" is common and results in flatlined learning for at/above-grade-level students, I would argue it's somewhat orthogonal to the claim "disagree with the premise of maximizing learning."
If you teach to the slowest learners, then it's still possible you might agree with the premise of maximizing learning (for the slowest learners). So, I can't use "teach to the slowest learners" to argue that many people disagree with that premise in an absolute sense (for all learners).
I considered mentioning in my first post that the school's behavior does not maximize overall learning because such a large portion of the students are at/above grade level. It is possible that their method of teaching the below-grade level students is consistent with maximizing learning for those students. I would think that for students who are behind, you would want to increase engagement/fun in order to increase the likelihood of maximizing learning in the long run. If you just wanted to get their scores up that year, you might focus more on drills instead.
East Asian education culture (i.e. cram schools / high stakes testing) has shown you can always maximize learning at the expense of happiness. And work hours. And it works well to develop STEM talent. People with aptitude are naturally going to excel while those who aren't can still drill acceptable levels of competence, even if they hate it. If your life/future depends on knowing math, a surprising amount of people with no love for math will end up being decent at math.
Anecdotally - I was extremely miserable at university, so miserable that I could not bring myself to study. Failed computer science papers. Hated programming more than anything I had ever come across. Mental health problems hit me hard. I could not learn.
Fast forward to me being out of uni and learning it in my own time, I absorb learnings like a sponge at an accelerated rate. It all makes sense. I want to study and learn almost every second of the day. I’m 100x more efficient in my learning, faster, and retaining so much information.
The difference? Happiness
Passing tests, exams, does not make you smart. It makes you “another brick in the wall”. University does not make people smart. School doesn’t make people smart.
In fact, I think we school too much, and it makes people regress. It’s one-size fits all. Play, and self-trials are what have led to my most major positive outcomes in education.
Exams were unnecessary stress.
When I have kids, I won’t be sending them to school full time. I will encourage play, self exploration, and discovery. And teach them curiosity. The world needs less academic rigour, and more curiosity
Education is both for personal growth and systemic human capita / talent development. System can be extremely diminishing returns on individual level, but still effective on collective/state level. Not every country in position to build workforce via brain drain of talent from elsewhere. Policy wise what's left is to brute force talent development, same way sports leagues have their own talent development system by throwing meat at problem and filtering for different tiers of talents to fill different segments of demand. A mixure of predisposed and forcefully cultivated. The goal isn't to get self actualized workforce who do what they love, but to get as much talent relevant for national priorities, i.e. east asia education bias STEM vs humanities. If you beat 6/10 kids into being competent at math when 3/6 those kids would rather be poets, but eventually get funnelled into STEM because extreme measures made them good at science, in lieu of being good at writing, then returns is function of how aligned system is with national priorities (i.e. need 5/10 workforce doing math). What I will say is culturally, preoccupation with credentialism is bad at recapturing talent like you who fall through cracks but found another way. I'm sure the system can be optimized to account for individual differences and even somewhat balance/account for individual desires, but currently even rich well administered countries rarely have resources to make bespoke curricula to ensure most flourish. That said, what's happening in east asia (especially cram schools) is IMO way past diminishing returns, but it does seem to work.
I think it's important to recognize that collective of miserable peoples can still be pushed to do great things. East asia work culture is miserable people working 200% harder for 10% competitive advantage, and now they dominate in a lot of high tech sectors after relatively brief period of development. Of course they're paying for it with low birth rates. But it's crazy how much of the world does not run on happiness. And I dare suggest human progress is built more on a few people doing what they love supported by unending amount of miserable people just along for the ride.
I would point out that "repetition" is not all equal. Education has a tendency to front load lots of repetition which creates boredom and disengagement. The spacing of repetition is often poorly done and rarely individualized.
Part of what makes "spaced repetition" so effective is that the spacing is customized based on performance and you aren't forced to keep frequently practicing the things you know well and can focus that effort on the things that aren't quite clicking.
As testing becomes more standardized, it does a worse job of assisting with the education process and serves only as a measurement tool. Tying funding to standardized test results turns them into a gamifies metric that actively makes education worse.
Edit: I believe an important goal of education should be to instill joy in and desire for learning. These are important for the kinds of self directed practice that are key to making progress because we have a some natural sense for what challenges us and what doesn't. This should be supplemented by experts explaining how skills link together and build on top of each so we know how to learn a skill that is currently too hard to practice directly.
I agree that spaced repetition is often done poorly and needs to be customized based on performance.
Just putting this here because you might find it interesting: I spent several years building a fully-individualized spaced repetition system that not only
1) works in a hierarchical knowledge domain (where repetitions on advanced topics "trickle down" to update the repetition schedules of simpler topics that are implicitly practiced), but also
2) minimizes the number of reviews by choosing reviews whose implicit repetitions "knock out" other due reviews (like dominos), and
3) calibrates the spaced repetition process to each individual student on each individual topic (student performance and topic difficulty are competing factors).
There are likely many, many definitions of what "maximizing learning" means. And that drives different approaches to education. Not "whether education should seek to maximize students’ learning."
That could be taken as a fairly naive statement -- there are countless successful examples of non test-driven education out there.
What are these “parameters” by which performance is measured?
You’re all programmers, you all set parameters all the time that are completely arbitrary, you design the systems through which actions are performed. What does it mean to be a great musician, to be able to write music as good as Mozart? But Mozart never wrote anything as good as Mozart! To be great means breaking the boundaries of all values, to create something that goes beyond being merely “good enough”; good enough according to some standard set by somebody else.
Hit your targets, maybe you’ll be good. We should not be raising kids to be “good enough”…infact, let them be terrible! I’d rather they be awful at everything in every concievable way then hit the same targets over and over again.
> To be great means breaking the boundaries of all values,
No? A great runner doesn’t go beyond the values of running faster over longer distances.
Proving great new theorems does not require going beyond traditional notions of proof.
Not all well-defined scoring functions take the form “f(distance to some subset of the space of possible behaviors)” for some decreasing function f. You can have a well-defined function which is unbounded.
If the space of possible behaviors/results isn’t compact, then following a scoring function may lead to going further and further out,
not just getting close to one thing.
Measurable performance doesn’t put into question the measure itself. Recursion, which is kind of bullshit imo, did require reinventing the concept of a “proof,” and now, as a side effect, we have programming.
Of course we are wrapped up in certain histories, but those histories can only command us, in a certain sense, if we follow their measures—we would be going “further out” into the same. “Running” might be the oldest useful activity of humanity, but competetive running probably doesn’t antedate the industrial revolution—there was no concept of a “run” for hunter-gatherer societies. Frankly, nobody attempted to engineer humans through education, training, to be good at hitting specific, measured targets before the industrial revolution. Without consciousness of this, one is no different from the machines which test them. Maybe one person wins a gold medal one year, another another, but year in and year out they are ruled over by a number alone. There is no choice or freedom here, in the end a great runner is no better than a great pair of shoes, perhaps the ones they’re wearing.
The point of this discussion is education. I would like to think, though I can never prove it, that we are more than things that can be engineered to hit specific targets. And if we raise our kids to do that, then they will never be more than machines. There is no measure for a great thinker, a great artist: they create their own. So I’d rather live in that world, where everyone freely determines their actions, than in one where the breadth of all their actions are predetermined for them.
I just don’t think that a particular measure of value, is necessary something for which “hitting the target” is a good analogy.
Going beyond what has been done before, even in qualitatively novel ways, can be favored by/within a fixed notion of what is valuable.
One can explore wild new ideas and actions whole still in pursuit of one fixed way of evaluating what is valuable.
Now, if your claim is something like, “We can never (or at least not in the near future) adequately describe what we really value, and focusing too tightly on achieving those approximations to it which we know how to express, leads to worse outcomes.”, then yeah, I suppose I agree.
>One can explore wild new ideas and actions whole still in pursuit of one fixed way of evaluating what is valuable.
It was this way with Leibniz and Spinoza, that everything fell under a One, an infinitely analyzable substance that, at the very limit, at the very edge, which only the intellect of God could understand, was the perfect synthesis and the completion of the entire universe. Its for that reason that I don't call myself a Hegelian, but at least I believe in something that Hegel saw in Kant: that the power of the infinite is not for some God in heaven infinitely distant from mankind, and we are just the poor image, his murky reflection. The end of the Phenomenology of Spirit is called "Absolute Knowing," and what is the truth of absolute knowledge? There it is, God became Man, and from "Out of the chalice of this realm of spirits / Foams forth to him his infinity." God is dead, and from that sovereign truth reigns the absolute knowledge that, in every action determined with full conscious intent, the infinite is grasped, the infinite power to determine world history!
Values? The horizon, the constellation of value, is like a net that ensnares that seed of power. One must be strong enough to break through value itself, to achieve the revaluation of all values. To take flight into the other side of infinity.
Understanding is better than memorizing, learning to learn by yourself instead of learning a small and uneven set of things, increasing the amount of things that you know that you don't know as focusing in a small set of increasingly obsolete knowledge.
We are not at an age of information scarcity, quite the opposite. There are things that you must learn, like basics of reasoning, critical thinking and learning. But there are a lot of things, specially new ones in the current acceleration of events, that you should be able to learn by yourself.
Yes, spaced repetition and regular testing are effective in a purely mechanical sense. They’re good ways to acquire knowledge and skills. But, it’s really worth noting second order considerations. Does the median student enjoy the testing and repetition enough to be motivated to do them? Do they understand the utility or applications of the skill? Are you even teaching the right set or sequence of skills? I think these considerations have gotten more traction in recent years, possibly at the cost of acquiring the skills themselves. That said, I don’t think the cynicism in the article about teachers just wanting fun easy activities is warranted. They have to consider other things than the most mechanically effective methods of learning.
Spaced repetition should be automated by inserting it into the most popular games that education authorities can license in bulk. So students get to play their favorite games for free but they have to do spaced repetition learning to get to the next level in the game. Doesn't require school buildings or teachers.
Teachers should be incentivized based on the grades gained by their students during their tutelage. So grades gained = grade after - grade before summed over all students.
Waldorf schools work.
There is no testing, there are no grades until high School.
Lots of the silicon valley elite send their kids to Waldorf schools.
Are they all completely unaware of the effectiveness of deliberate practice?
I rather doubt it.
I think they are aware that motivation is what really matters.
My three kids all went to Waldorf School and never had grades until they entered high school at a public school.
None of them ever made anything less than an a+ in any class in a public school.
This was transitioning cold straight out of Waldorf with no testing to a public school.
They all got into very good colleges and are performing extraordinarily well at college. (My oldest daughter graduated from Stanford with a CS degree and straight A's)
Mathematics is mostly not that useful. Unfortunately most useful skills are not explicitly taught by our education system.
Negotiation. Cooperation. Compromise. Communication. Soft-skills are poorly taught and hard to learn. We mostly learn these skills subconciously. Even when trying to learn them conciously it is difficult.
Mathematics hasn't been of much use to me over decades of working in software. 90% of my degree in electronic engineering was mathematics (I didn't know better). Some skills needed for my software practice: (a) focusing on the right problem (b) talking to users (c) concentrating on usability (d) design skills (e) negotiation, (f) obstinacy in the face of problems outside of my control (g) choosing who to work with.
University level mathematics is clearly required in some software domains.
The book systemantics claims that systems exist to perpetuate themselvesand not to perform the function for which they were created.
This is working as expected for the school systems which were created to make drones for the factories and to appease the Communist revolutionaries in the 1800s.
Completely misses the point. Rote learning teaches you to do circus performances. So you can do math now, so what? You don't know what it is, how it works, or how or where to use it.
> And there are plenty of teachers who are incentivized to use easy, fun, low-accountability, hard-to-measure practice techniques that keep students, parents, and administrators off their back.
I've been visiting schools in London and I've noticed that teachers/headmasters "sell" their schools with claims that their schools are "inclusive" or use "play based learning" or "great activities" etc. Everything and anything, but being serious about learning. I kind fo get the impression that the approaching to schooling is to get kids off parent's backs. Like "we'll take them, they'll have fun, don't worry about it". Extremely frustrating.
If anyone has here experience with schools in London, shoot me an email and lets talk? (email in profile)
There's a lot of that too in the USA. Unfortunately, the strategy works because the learning debt doesn't come due until the students graduate, and at that point, it's somebody else's problem.
Parents get sold an idea that their kids are learning, but this learning isn't actually verified along the way, so nobody is held accountable for making it happen.
(Sure, the kid might say they're learning and the teacher might say they're doing a great job, but can you really trust that when both parties are incentivized to say that learning is happening and penalized for saying that it's not?)
By the time the kids graduate, the learning debt is so massive that it's easier to just change the narrative. Johnny got smacked in the face by math class at college and no longer wants to be an engineer? It hurts too much to accept the truth that life could have been different if he had seriously learned math in high school. So instead it's "ah well, engineering is hard and Johnny was never really cut out for it." Or even "Johnny's engineering class is unfair."
Years ago I was watching so-called "best practices" videos on teaching estimation in early primary. I was astonished. Often the skill being coached seemed to be participative performance in a collaborative pretense of understanding. What to do, and not do, so we can both pretend learning has occurred and move on. With the teachers themselves largely unclear on the concepts.
Ongoing formative assessment is usually closely associated with instruction, with attendant conflicts of interest. I wonder if say LLMs, might transformatively permit fine-grain and adaptive assessment. So a "what have you been learning?" dialog for objective identification, and then assessment across the objective neighborhood. With potential for tutoring of course. But here emphasizing a possibility for rapid verification of learning.
> Often the skill being coached seemed to be participative performance in a collaborative pretense of understanding. What to do, and not do, so we can both pretend learning has occurred and move on.
And the more you do that, the more you're shutting off any possibility of any learning occurring in the future (because the student will not have acquired the prerequisites for whatever they're supposed to be learning).
Sorry, no. Creating a search-based PIM over old backups seems a doable thing. But for now, hmm, youtube has a "before:2015" syntax..., but still, "teaching estimation" or "estimating jellybeans" or blocks, or etc. And some education sites used to self host videos. :/
It wasn't anything singular. Telegraphing/hinting at the desired response. Questions series amenable to surface pattern matching, rather than scaffolding or probing understanding. Coaching for response rather than for understanding. Lack of clarity on why estimate, and how.
Except for that last... I imagine having a film crew hovering was not low stress, especially when things weren't going well. And at the time, when early-primary estimation was still a newish thing, having even a crufty idea what it could look like might still be worthwhile training-wise. Hopefully similar work would now be better?
Getting to top schools is more competitive then ever. For all the complaining about kids not learning, they are actually learning more then we used to.
My 6 month old is currently learning a whole host of things, and we are doing repetition and watching his growth -- so no disagreement on the premise that these things are good. But we are also making great pains to ensure that he is playing, is always happy doing new things, and doesn't have to continue if he gets overwhelmed and/or bored. We do this because the mentality of someone when learning is a huge factor in what they get out of the activity.
I had the impression that the current push against testing and repetition is because it ended up with children hyper-focussed on "passing the test" skills but not spending the time to question things too deeply and gain that understanding that actually helps the workforce (like Feynman noticed in Brazil).