How Bad Ideas About Learning Spread | 5 Examples 

Thanks - me too! : )

If I could give you more hearts, I would. Agree with everything you say. And these problems that you identify are intertwined in various ways For instance, people get emotionally attached to "systems" because they provide a whole philosophy of education, even though you could, in principle, detach the system, which is a product that someone is making money off of, from the philosophy - and in many cases the stated philosophy doesn't necessarily support the system in the first place! The "big idea" problem is not limited to education. It's common to many reform projects. It's not that big system overhauls are always bad things, but they often do not take into account what people are doing now, which is critical when you start talking about change and improvement. If you made a series of small changes over many years, always moving in the right direction, you would get pretty far. Re: emotional attachment. Education is a bit "special" in that everyone has had an educational experience. Everyone has been a student and many people have been teachers in some capacity. This narrows people thinking to just their own experience. I think it's hard for people to understand that there are basic principles that are important, but these principles can be enacted and adapted in various ways depending on the context. IMO, the transactional nature of grades ultimately encourages cheating and other low-effort strategies to obtain the metric without getting the actual reward, which is the learning. Anyhow. I'm sure I could go on and on about these things. Really appreciate you sharing your thoughts!

@@benjaminkeep I think the issue with the "big changes" is that there often really isn't that much there, there. A lot of the really popular teaching books are absolutely filled with "filler": sidebars and anecdotes and checklists and basically the same 3 ideas rearranged enough times to be worth $24.99. And then that new program is supposed to solve everything--throw out everything we did last year on collaborative learning, this year is all about cold calling! I really think it's driven by this need to have a marketable package that needs to justify it's own price tag. In terms of grades, please don't think I'm defending the system. It's an awful, terrible, horrific system and the reasons you cite really only scratch the surface. But it is the system we are in, and it shapes every aspect of classroom practice. It's like if someone were wondering why American physicians weren't integrating research-backed strategies into clinical practice, but the issue of cost in our kluge of a system wasn't part of that research or discussion, but it's front and center in every examination room conversation in this country. It's just really striking to me how central grades are in the classroom and how lightly they are considered in the literature.

​​​@@amandaashmead5770 If I may invoke the wisdom of Foucaultian thinking and Nickolas Rose, grading makes the administration and governance of individuals a calculated and rational act. It is based on norms - either in comparison to some threshold, or in comparison to the population. With threshold norms I mean you need a minimum level of questions correct in order to pass, and population norms refer to an assessment of your standing as compared to everyone else. And while this sounds dystopian, it really depends on how these tests are used. For example, if you do really poorly on your in med school, I'd be glad a test score would determine you're not allowed to get your diploma yet! And social comparison is the only way personality tests would make sense Assessments and tests (which pertain to the branch of 'psychometrics') also draw on the appearance of objectivity. Rather than determining if someone's a good and successful or a bad and unsuccessful student on the basis of rule of thumb, tradition, prejudice, or something else, the test allows judgment on the basis of 'truth' Numbers don't lie, after all. ...then again, if you think about it, they also don't tell the truth. While tests tend to provide information, calculated via predetermined criteria, it is really the people who interpret the numbers. It is people who try to tell a story with those numbers To nuance the cynicism towards testing I might have evoked, a lot of smart people with the best of intentions have thought about issues. Whether the outcome of a test does or does not correspond with reality is a topic of critical importance in psychometrics, and in psychology as a whole. We call that 'construct validity': whether a test actually measures what it purports to measure In any case, tests, assessments, and grading are thought of as indispensable to the ideal of organizing and regulating society in a fair, rational, and effective way. Tl;dr: This perceived indispensability of grading is why it is so difficult to get rid of the practice Source: Rose, N. (1996). Expertise and the techne of psychology. In Inventing our Selves: Psychology, Power, and Persoonhood (pp. 81-100). Cambridge: Cambridge University Press.

Thanks for the kind words. It's taken me a little bit of time to figure out the medium, but looking forward to putting out good stuff in the future.

I will add that book to my reading list! Personally, I tend to agree that people way underestimate the value of rest. A while back a did a video on the effect of sleep on learning and it was just crazy reading all the studies. I remember people bragging about pulling all-nighters or studying for 8 hours straight. I just think that's counter-productive and more for show than someone really studying for all those hours. It does not good to study when you're exhausted.

Glad to hear!

Many thanks for your kind words!

Yeah, having been a flashcarder myself, I know what you mean about the feeling of doing the flashcards and being done. No decisions to be made. There is an advantage to that. And maybe part of developing more effective free recall/retrieval practice techniques means solving the scheduling problem. I wish I could say I had a super-effective study system that anyone could easily apply, but I don't. Although I was in school for *checks notes* 24 years, I am not a full-time student and didn't start researching learning in earnest until I was in the PhD program (so didn't have a lot of chances to test out ideas, especially in a classroom structure). Probably know more about what doesn't work than what does. Because I'm not a full-time student (or a student at all), I have the luxury of moving more slowly. For instance, I read some non-fiction books pretty slowly and will do a free recall session about once a week on them to capture details of what I've been reading. And sometimes in the same session I'll do two or three different topics. I don't think it would be bad at all to do something like Wed morning: free recall of material from Monday and Tuesday; Friday morning: free recall of material from Wednesday and Thursday; Sunday: free recall of material from Friday. Monday: free recall of past week's material; Tuesday: random week from the past two months; Thursday: random week from 2+ months ago. It's not perfect, but it approximates a kind of interleaved schedule where you don't have to think that much (if you try this kind of thing out, please let me know how it goes!).

100%. I do think the desire for short-term realizable gains on non-transfer tasks is a huge component. It is a bit of a leap of faith. That's also why it's hard to separate research methods from best practice in learning - if you don't know how to measure learning gains effectively (ideally on delayed transfer tasks), then there's no way to meaningfully improve your practice or training because you're constantly chasing after the wrong thing.

I haven't read Neuroteach, so I'll check it out. I don't know Rick Beato either, but your story raises a point I wanted to get into in this video but didn't, which is how often personal experiences lead to misattributions of causes. If you go to the Nuryl website, they have the "based on science" section, which is a lot of based-on-science sections for edtech products. You say things that are generally true (e.g., there's a lot of brain development going on prenatally and in the early years of life), but that doesn't really address whether the program works. You could substitute the same copy for any "brain-based" program (and what isn't brain-based?). It also seems to follow the "have new idea, develop product, explain how it works later" path of development. None of which is to say that people shouldn't have fun experimenting with things. And, at first glance, I see little downside to using Nuryl - it explicitly encourages social interaction; Baby Einstein seemed to be used mostly as a baby sitter, rather than a shared experience (my reading of the literature is that screentime that involves shared interaction is perfectly fine and beneficial for children). But my guess is a lot of the intelligence you see from these kids just comes from having very interactive parents who are doing a bunch of things to help their kids learn (like word flash cards, math problems, etc.). I did see a clip of his kid doing some basic 1-by-3 and 4 digit multiplication. I doubt that this has much to do with listening to music in the womb.

You got it - won't be out right away, but I'll try my best.

Hmm... somehow a different reply got attached here. So I'm re-writing a reply. My two cents on the NGSS: it's the right direction. Although whether the standards themselves will create sustained change in that direction... I'm more skeptical. There are other forces that prevent teachers from effectively doing what NGSS asks (e.g., routine standardized testing; a desire to see quick results; the teaching crisis - high turnover, low pay, low benefits, terrible working conditions, bad PD; the standards in "report" form rather than as an actionable toolkit).

[FYI for anyone else who's reading this: Direct Instruction is a very specific form of teaching, where the teacher memorizes a script and gives all students the same exercises and scripts. These scripts have been developed by researchers in specific domains. It is not "lecture-only" (there is a lot of practice in DI, which is a good thing). It also quite distinct from explicit instruction, which means to tell students what you're learning (the purpose of the lesson, what you're going to achieve, etc.) which is quite a good thing. Direct Instruction (capital letters) is also not the same as direct instruction (lower case), which many people might use casually to refer to tell-and-practice, lectures, or other kinds of teaching where the teacher plays the primary role in telling the student something. ] Thanks for the great question - I am somewhat familiar with it (I have Englemann's DI guide to teaching your kid to read, for example). I have not gone deep into the literature, but here are my initial thoughts. 1) I think the empirical support for the method is probably over-stated. I say this because no one outside of the small group of researchers connected to DI agrees with them. And from glancing at the meta-analysis that's frequently cited by them, and critiques of that meta-analysis, I don't think it's very well done (granted, it was done in the 70s; probably pretty decent for that time). Most of the problems seem to lie in clearly defining the measures and accurately estimated the effect sizes. They interpret their colleagues' disagreement as a sign that their colleagues are ideologically beholden to Dewey, etc. Which I think is partially true. But people - especially researchers - are not impervious to evidence. The same basic studies have been cited over and over across the decades as "establishing" DI's superiority. But I would rather see improved studies run by people other than proponents of the idea, using multiple different measures and methods, that all reach the same conclusion. 2) I think DI can be useful in some domains. Standardization and sequence testing (refining the learning sequence over time) are advantages. One of the hazards of many kinds of teaching reforms comes down to implementation, which is usually quite poor and/or variable. DI essentially eliminates implementation problems because all teachers follow the same script. As a generally applicable approach to teaching, however, I think it's not a good idea. The problem is that it's not adaptive and responsive. Same script regardless of prior knowledge? Terrible idea - how do you know that the student will interpret the script properly? No diagnostic testing? Same feedback regardless of student? Again, how do you know that the student is interpreting feedback correctly? If you look at the countries with some of the best education systems in the world (Singapore, Japan, Finland... etc.), no one uses anything remotely close to Direct Instruction (capital letters). One of the major issues is that DI erases teacher expertise, and the places with (seemingly) the best education systems almost invariably rely on teachers as professionals to make teaching decisions all the time. Imagine if you gave your doctor a script to follow instead of tools to use. That's kind of what DI does. 3) DI, IMO, follows a common pattern in education reform. Someone has a new idea. They refine that idea and have some research support. Then they start expanding the scope of this idea beyond it's original research scope. And to do that, they have to make it seem like the method is totally different (and opposed to) what already exists. For instance, in the John Stuart Mill book you mention, the abstract talks about how the school system is beholden to Dewey and other progressive education philosophies. But if we exhumed Dewey from his grave, he would have a heart attack if the U.S. system was described as "Deweyian". There's very little practical learning going on in most schools (e.g., home economics, shop, gardening, etc., nevermind science labs that actually mimic scientific inquiry). Of course, I guarantee you there are pro-Dewey/progressive books out there who say the exact opposite - that our education system is beset with DI practices. While there is a lot of "tell-and-practice," it's not DI as Englemann refined it. Actually, what we have is a mish-mash of bad and good ideas not at all working in a coordinated way. Personally, I think we could benefit from more refined learning sequences, like Englemann and his colleagues developed. And I would rather have a standard baseline level that teachers are performing at than the huge variation we see in teaching levels. Part of what makes the U.S. slow to change is that there everything is done on the state or local level (as far as curriculum, learning sequences, etc. are concerned) - so everyone is changing all the time but not in the same direction. But I also think that prior knowledge, feedback, learning communities, motivation, retrieval practice, etc. all have important roles to play. All of my opinions above are revisable, given the right evidence.

@@benjaminkeep Thank you for the through response! You make a good point about the evidence probably being over-stated. I find it hard to know what to make out of the research because as you say it, most of it is done by a small, secluded group of people. A lot of it is also pretty messy, and there’s also this whole ideological gung ho that doesn’t help the situation very much. You mention how things can go astray if students’ prior knowledge is not taken into account, which is definitely true, but going from their website, I believe that they have placement/diagnostic tests that they use to group students and prevent that sort of thing. Though I guess that still may give rise to some implementation obstacles. On the question of how to know whether students are interpreting feedback properly, I think that’s one of the main motivations behind the scripts. Having carefully curated scripts helps make the explanations consistent and prevent ambiguity. They also do a lot of choral responses (where they quiz students and have them all answer at the same time), which helps the teacher identify whether everyone understands the content or not. Anyways, I do agree that it’s weird how DI strips teachers out of a lot of their usual duties. I think that the way that they argue against that is that in DI the teacher assumes something akin to the role of an actor, so that even though they may have a script, they still need to put in effort in order to deliver the content in a way that keeps the students interested. It’s still pretty odd though. Don’t quote me on this, but If I remember correctly, DI originally didn’t use scripts, the creators just found it really hard to train teachers without them so they ended up switching over. Also yeah absolutely, I think that education reform pattern is definitely there. DI feels to me like a really good approach when what you want to teach is very well defined or specific, like teaching kids how to do addition and subtraction, but I have no idea how you would work out a teaching sequence for more open ended subjects or more creative tasks. Definitely not the end all be all that is going to solve all problems of education. Regardless, I still recommend the Stuart Mill book! It explains a lot of the thought process behind the teaching sequences which to me are really fascinating. Simple stuff like: if you are quizzing someone on a group of items, make sure to quiz them in a random order to make sure they don’t memorize the order instead of the items themselves… To stuff like: that if you want to teach someone what a fraction is (especially a kid), starting out with the set 1/2,1/3,1/4 can be a bad idea because it may lead them to think that a fraction needs to have 1 on top or that fractions only describe values smaller than 1. It’s very neat.

Okay, I will check it out! Thanks for writing such a thoughtful reply. Your last paragraph particularly strikes home, because both of those examples are key examples that really have nothing to do with Direct Instruction (the method) and are both things that researchers of many different schools of thought could agree on. It's strange though, because we often don't leverage "the easy stuff," which is stuff like that.

The research I've read is positive on it, but I haven't read deeply about it. Like a lot of education reforms, I suspect that what "standards-based grading" means in practice varies somewhat. But from what I understand of it, I like it. It dovetails with formative assessment and focuses on the development of specific knowledge and skills over time.

I've used Anki before, too. There's nothing inherently wrong with flashcard apps. I think they can be helpful in some cases. The problem is thinking that they are some kind of complete solution to language learning or memory more generally, which is how Supermemo pitches its product. Cued recall (which is what flashcards are) is generally considered to be less effective than free recall (recall performed without a cue). Flashcards lack relevant context, are too fixed in meaning, and do not really involve the key skills (of say, reading, listening, speaking, or writing a language) You might check out this one for some of the limitations of the spaced repetition approach. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-iW7Fp9Mtq1Y.html And this older video, which covers some of the same ground, but talks a little bit more about flashcards ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-ZIGrHI353no.html Also, Justin Sung's videos, which I respond to in the first link go into good detail as well.

People can - and do - learn throughout their entire lives. : )

I don't have any plans to do that, but would be open to chatting with him.

@@benjaminkeep Definitely!

Soon! Next week will start another round of weekly uploads. Thanks for waiting!

Hey - sorry - I think I missed replying to your comment earlier. This is a really good question. It's an aspect of deliberate practice that is often brushed off, IMO. Part of the answer is that students aren't supposed to set well-defined specific goals. I don't mean that it's not a good idea to try to do so, but when you don't understand much about what you're trying to learn, it's really hard to break down the subject into useful parts. That's what a coach or teacher is for. Making a goal "specific" is important, but it's also about pursing the "right" goal. For instance, suppose you are learning to play soccer. You might have a specific goal, like "improve my passing accuracy". That's taking a sub-part of the many skills involved in soccer. Probably a good idea - there are exercises that focus just on passing. But what kind of passing are you practicing? If you only practice passing from a standing position, then you're not really practicing the key skill that makes players really good (passing while moving and getting around defenders). One way of approaching setting a specific goal from the student's perspective is to think about what the expert is doing that is making them so good. If I'm learning to play chess, of course the ultimate goal is to "play chess better". But that's not a helpful goal because it doesn't distinguish between more useful learning activities and less useful activities. A more helpful goal would be "build a repertoire of opening patterns". That's specific enough that you could say, "okay, I want to focus on the most used openings; I need somewhere to learn what those opening patterns are; I'll probably be trying to recreate those patterns on an open chess board and from both black and white positions; I'll probably be using those patterns in real games..." etc. Not sure if the above helps, but this is going to be an ongoing topic that I'll broach in more videos, too.

@@benjaminkeep thanks your comment was very helpful. But in Anders Ericsson's book peak he says that they're are short term goals and Long term goals. So for example a short term specific goal for remembering digit's according to the book"I will remember more digit's than i did in the previous session" if I remembered 5 then my next goal is 6 and Also according to the book again he says there was no long range goal because none of them knew how many digits one could possibly memorize " which kinda left me confused. And I am not sure if you read the book or not so this might be new information to you

My understanding of Ericsson's position is that limits on human performance are arbitrarily large. That is, if you look at performance over long periods of time in any sport or task, people keep getting better (sprinting and marathon and swimming record times are all much shorter than they used to be 100 years ago, for instance). And there are also lab studies where people just keep getting better at memorizing numbers for a very, very long time. So putting an arbitrary limit as a long-term goal seems presumptuous. That may be what he's talking about there. Having a short-term goal "like memorize 6 now that I can memorize 5," I view more as a metric that helps you keep track of your progress. It's more motivating when you know you're making progress. I bet Ericsson would say it also helps focus your effort to the next goal.

I copy n pasted this from the book so that you have a better understanding also what do you mean by limits? He said that there is no limit to performance. Anyways please read this. PURPOSEFUL PRACTICE Purposeful practice has several characteristics that set it apart from what we might call “naive practice,” which is essentially just doing something repeatedly, and expecting that the repetition alone will improve one’s performance. Steve Oare, a specialist in music education at Wichita State University, once offered the following imaginary conversation between a music instructor and a young music student. It’s the sort of conversation about practice that music instructors have all the time. In this case a teacher is trying to figure out why a young student has not been improving: TEACHER: Your practice sheet says that you practice an hour a day, but your playing test was only a C. Can you explain why? STUDENT: I don’t know what happened! I could play the test last night! TEACHER: How many times did you play it? STUDENT: Ten or twenty. TEACHER: How many times did you play it correctly? STUDENT: Umm, I dunno … Once or twice … TEACHER: Hmm … How did you practice it? STUDENT: I dunno. I just played it. This is naive practice in a nutshell: I just played it. I just swung the bat and tried to hit the ball. I just listened to the numbers and tried to remember them. I just read the math problems and tried to solve them. Purposeful practice is, as the term implies, much more purposeful, thoughtful, and focused than this sort of naive practice. In particular, it involves the following characteristics: Purposeful practice has well-defined, specific goals. Our hypothetical music student would have been much more successful with a practice goal something like this: “Play the piece all the way through at the proper speed without a mistake three times in a row.” Without such a goal, there was no way to judge whether the practice session had been a success. In Steve’s case there was no long-range goal because none of us knew how many digits one could possibly memorize, but he had a very specific short-term goal: to remember more digits than he had the previous session. As a distance runner, Steve was very competitive, even he was only competing with himself, and he brought that attitude to experiment. From the very beginning Steve was pushing each day to increase the number of digits he could remember. And again this is his words

I don't disagree with Ericsson here. Here, he's just saying you can't just rely on metrics like "amount of time" or "number of run throughs" and think that just mere time doing the thing automatically leads to large improvements. Having a short-term goal is helpful for structuring a specific practice session. Even if you don't reach the goal, it structures your attention. This is distinct, to me at least, from pursuing the right skill, like I was talking about in the first reply. Note: he's hedging a bit between the (research-derived) idea of deliberate practice and his term "purposeful practice," which describes a kind of effective practice that students can do without, necessarily, the benefit of a teacher.