top of page

Ep 62. Why more classroom technology is making students learn less

This transcript was created with speech-to-text software.  It was reviewed before posting but may contain errors. Credit to Canadian Podcasting Productions.

 

In this episode, Anna is joined by Dr. Jared Cooney Horvath, a cognitive neuroscientist, educator, and bestselling author of The Digital Delusion.  They examine what decades of research and international data reveal about classroom technology, screens, multitasking, attention, and memory, and why more technology often leads to less learning.

 

Jared explains how offloading knowledge to devices can undermine higher-order thinking, why human teachers’ expertise, and practice, remain central to learning, and when technology may help or hinder students. The conversation also tackles how schools and families can navigate an increasingly tech-saturated education system.

 

This is a thoughtful, evidence-informed episode for educators, parents, and anyone questioning whether digital tools in the classroom are helping students learn or holding them back.

 

This episode is also available in video at www.youtube.com/@chalktalk-stokke 

 

Order The Digital Delusion here: https://www.lmeglobal.net/digital-delusion 

 

TIMESTAMPS

[00:00:22] Introduction

[00:04:50] Cognitive decline among Gen Z

[00:09:14] The decline international test scores and the correlation with technology

[00:11:28] Screen usage in schools

[00:13:03] Relationship between EdTech and countries that invest less in it

[00:16:16] Effect size in education in the context of EdTech

[00:20:49] What forms of EdTech work?

[00:25:17] When EdTech is a better than nothing

[00:32:57] Practise and producers are essential to learning

[00:33:30] What is creativity?

[00:34:20] Why offloading learning to technology harms creativity

[00:38:50] AI: The Tool Nobody Asked For

[00:44:17] Difference between K-12 and university students using EdTech

[00:47:14] EdTech creates multi-tasking

[00:54:27] Advice: Responding to “digital devices are ubiquitous”

[00:55:50] Advice: Responding to “these students learn differently”

[01:00:32] General advice for parents and school leaders

[01:03:46] Laptops vs iPads vs notetaking by hand [01:06:48] Being a Luddite in the 21st century

 

[00:00:00] Anna Stokke: Welcome to Chalk & Talk, a podcast about education and math. I'm Anna Stokke, a math professor and your host. Welcome back to another episode of Chalk & Talk.

 

This episode is available in both audio and video. You'll find a link to my YouTube channel in the show notes and please do give the show a follow-on YouTube. Before we get started, I'd like to acknowledge new support through a grant from La Trobe University.

 

This support helps me to continue to share evidence informed conversations that connect research with practise and advance the goal of improving student learning, particularly in math. I've got an important episode today. My guest is Dr. Jared Cooney Horvath, a cognitive neuroscientist, educator and bestselling author.

 

His new book, The Digital Delusion, examines the impact of classroom technology on learning and raises serious questions about whether screens in schools may be harming learning. In this conversation, we examine what decades of research and international data tell us about screens, multitasking, attention and memory. Jared explains why more technology often leads to less learning, how offloading knowledge to technology undermines higher order thinking, and why human teachers' knowledge and practise remain central to learning at every age.

 

We also discuss when technology may be helpful and when it's not, and what this means for educators and parents navigating a tech saturated education system. This episode raises serious questions about what the widespread use of education technology may be doing to learning and what we can do about it. I found this conversation informative and thought provoking, and I hope you do too.

 

Now, on with the show. I'm delighted to be joined today by Dr. Jared Cooney Horvath. He is a cognitive neuroscientist, an educator and bestselling author of several books.

 

His latest book is called The Digital Delusion, How Classroom Technology Harms Our Kids' Learning and How to Help Them Thrive Again. And it was just released in December. He has a PhD in cognitive neuroscience from the University of Melbourne and has expertise in human learning and memory and brain stimulation.

 

He currently serves as an honorary researcher at the University of Melbourne and St. Vincent's Hospital in Melbourne. Jared works at the crossroads between the lab and the classroom, spending most of his time working directly with teachers and students or helping schools, organisations and companies improve learning and engagement. His research has been featured in popular publications, including The New York Times, Wired, BBC, The Economist and PBS's Nova.

 

And I'm really looking forward to talking to him today about his great new book. Welcome Jared. Welcome to the podcast. It's great to have you here.

 

[00:03:19] Jared Cooney Horvath: It's awesome to be here, Anna. Thank you so much for having me on.

 

[00:03:21] Anna Stokke: I was lucky to get an advanced copy of your book, and I read the whole thing cover to cover. And I want to say it's a great book. It's a fascinating read.

 

It's troubling, but it's fascinating. And I think it's an important read for any parent or educator. So, we're going to talk a lot about that today.

 

Let's set the scene. The screens are everywhere, including in schools. And your book specifically takes a critical look at the promises of edtech.

 

You start by arguing that we do have a problem, that there's evidence that edtech harms learning. You then explain why, drawing on your background in neuroscience, and then you give practical advice for what we can do about it. So, would you say that sort of summarises the roadmap in the book?

 

[00:04:06] Jared Cooney Horvath: That sounds perfectly right. I went with the data mechanisms applications angle to it, because data only says so much. Mechanisms gives you a good grounding, but then everyone says, cool, so what? I hopefully we answered some of those questions too.

 

[00:04:20] Anna Stokke: You open the book with this hard-hitting statement, our children are less cognitively capable than we were at their age. And you argue that Gen Z, that's the cohort between around 1997 and 2012. Is that right?

 

[00:04:37] Jared Cooney Horvath: Yeah, right around 2012 was when we flipped into Gen A.

 

[00:04:40] Anna Stokke: You argue that Gen Z is less healthy, less happy and less knowledgeable than previous generations.

 

And I have two children that are Gen Z, by the way.

 

[00:04:50] Jared Cooney Horvath: I'm not trying to make a bad argument.

 

This is average of all kids across that generation. So, there will be some who are flying and some who are absolutely sinking. But in the middle, by and large, we're seeing declines in basically every base of cognitive measure we would care about.

 

[00:05:05] Anna Stokke: Can you walk us through some of the evidence behind that claim?

 

[00:05:08] Jared Cooney Horvath: I think if you take a look, there was a book recently called The Anxious Generation, which I think most people read. That shows a lot of great statistics about the health and the happiness angle of the next generation. Kids are just far more lonely than we ever were.

 

And this is not their fault. This is the world that we've built around them. The food they eat is not good for them.

 

The sedentary lifestyle has not been good to them. I mean, you've got kids with eye and back problems before they hit teenage years. That ain't great.

 

I mean, granted, when we were watching TV, I'm sure some of us got scoliosis or eye problems sitting too close to the screen. But the sheer volume of that happening now is getting a little alarming. What I really then like to focus on was the cognitive aspect, because I think other researchers have really taken the health, the mental stuff.

 

I wanted to say, well, what about learning? One of the weirder things is if you take a look at cognitive measures across the 20th century, so I'm talking memory span, I'm talking attention span, I'm talking creative abilities, critical thinking, even general IQ, every generation has gained on their parents. Every generation is doing better cognitively than their parents. And that's great.

 

And we attribute a lot of that to school. Each generation spends more time in school. School cuts your cognitive teeth.

 

Congratulations, you're doing better than your parents. That's exactly what we want. Gen Z is the first gen to flip.

 

That was their the first generation in over 100 years to do worse than us in all of these measures. In fact, they're only doing better than us in one measure, really rapid visual processing, which if you've ever seen what kids do on a phone, that shouldn't be surprising. That's what they do.

 

They process visual scenes incredibly fast. Everything else, memory, attention on down, including IQ, they are now lower than us. And Gen A seems to be going lower as well.

 

And there's no reason to assume that it's biological. You can't see that big of a biological flip so quickly. And that generation is going to school more than any other generation was in the past.

 

Somehow in right around 2002, attending school and cognitive development decoupled. And we haven't seen that decoupling in a century. My goal was to figure out why.

 

What the heck caused that split there?

 

[00:07:11] Anna Stokke: Where is the data coming from? Are you going with PISA data or experimental studies or what kind of data are you looking at?

 

[00:07:20] Jared Cooney Horvath: All across the board. We'll use international testing data. So, PISA, TIMSS, PIRLS, all that international stuff you've got here, in the U.S. We have NEAP.

 

You can use that. But there's also since 1962 was the very first meta-analysis published on the use of digital technology in schools. Back then it was called ICT, informational computer technology.

 

We have actual research lab data and on the ground practise data for over six decades now. So, tap into as much of that as you possibly can. And what's good about some of the cognitive stuff, too, is you're also able to tap into pure psych data like IQ testing.

 

They don't really do that much in schools anymore, but we still do that in universities, in lab settings. So, you can also tap into there to kind of see what's going on cognitively with all these different people.

 

[00:08:07] Anna Stokke: I did look fairly closely at one point at the most recent PISA data because they do analyses on this kind of thing.

 

They ask students and they ask teachers how much time is spent on screens or phones and that sort of thing. I actually usually mention this to my students when I start at the beginning of the year. Even having a phone on the desk beside you or even someone that's beside you using a phone seems to be correlated with lower scores, right?

 

[00:08:37] Jared Cooney Horvath: It's the digital, penumbra or like a zone of influence.

 

Or if you're sitting close to somebody who's using digital tech, you will learn less than somebody who's sitting further away from that person. There is something about these tools that seem to, they have a radius and they're just going to do what they can to take away your attention and harm your learning.

 

 

[00:08:56] Anna Stokke: So, about the PISA scores and the TIMSS scores and the PIRL scores, there's actually been declines pretty much everywhere.

 

How sure are we that there's a link to technology? That's one of the contributing factors to the declines.

 

[00:09:14] Jared Cooney Horvath: You can see this brings us into a very interesting question is correlation and causation. There are some scientific fields where you can do randomised controlled trials, and you can absolutely do your best to link the two.

 

With that said, it's very rare that you're going to find pure causation anywhere. But in the social sciences, especially if we're talking about learning classrooms, schools, performance, it is all correlation. We will never be able to run randomised control.

 

It's just ethically, you can't do it. The only way to turn correlation into causation at the social level is basically through mechanism. If you can gather enough correlation over enough time, over enough parts of the world, and they're all showing the same thing, that gives you good stead to say, we think these are probably causative.

 

But then once you can explain, here is the mechanism by which it would be causative, that is about as far as social sciences can ever go. And that's when we put our stamp and say, we think we've got this solved. If we go back to say, like the PISA data back in 2012, that was the last time they did a pen and paper test.

 

That was the first time they also asked kids, how often are you on a computer for learning purposes at school? And that was when they first saw this big, massive decline in every single country. Kids who were not on a computer did best. Kids who were on it for an hour did a little worse.

 

Two hours, a little worse than that. Three hours. And it just was almost straight slope down.

 

2015, same exact data. 2018, same exact data. It all came back looking the same.

 

And now you can start to say, well, cool, it's just correlation again. But no, you remember. So that's over three testing years, over 90 countries, over millions of kids.

 

And the percentages will have changed each year. So, the percentage of kids who used a computer in 2012 for three hours a day versus the percentage of kids who used a computer for three hours a day in 2018 will have grown. But the ratio stayed the same.

 

So that was our good sign that this data is showing us. This is probably the variable that's changing things. The computer variable is the only one that's not changing year on year.

 

We better take a closer look at what's going on. Can we find a mechanism? And if you can, then you've got yourself a good argument.

 

[00:11:17] Anna Stokke: When you talk about screen use, you're not just talking about phones, right? You're talking about laptops, Chromebooks, iPads, any type of screen in school, right?

 

[00:11:28] Jared Cooney Horvath: This is desktop. Because go back the first year that they were asking this in 2012, 2015, we didn't even have tablets then. Those were pure desktops.

 

How often do you use a desktop computer at school? So, this is not just cell phones. This is anyone who's using a screen or digital tools for learning. To be fair, we should say recently that term has now come to mean any student facing internet connected device.

 

You can imagine there might be a digital piece of digital technology that allows surgeons to practise surgery, right? That way you don't have to cut open an actual person. You can practise on this machine. That is not an internet connected device.

 

That's a digital tool that would be in a completely different realm. So today when we say digital tools, we're going to say any screen that also allows you access to other programmes on the internet.

 

[00:12:12] Anna Stokke: Is it the access to the other programmes that's the issue?

 

[00:12:16] Jared Cooney Horvath: That is one of them. That is the intractable issues. That is going to be one of them.

 

[00:12:19] Anna Stokke: There's other things though, like just maybe looking at a screen all day, that sort of thing that's maybe contributing to some issues.

 

[00:12:27] Jared Cooney Horvath: So that's where you get to see. There are usage issues and one of them is going to be the attention thing due to distraction. But then there's also pure biological issues is does the tool itself correspond with how human beings have evolved to think, to learn, to interact? And in a lot of cases, no, the tool was never built to do those things.

 

It's not a usage issue. When people say it's just a tool, it's how you use it.

 

Mate, no, it ain't. Cigarette is just a tool. I'm pretty sure the tool itself can be dangerous too.

 

The tool itself has its own problems that will, we think so long as our biology is our biology, it will never be able to overcome.

 

[00:13:03] Anna Stokke: Back to just sort of this international data. So, are there countries that don't use tech in schools as much as others?

 

[00:13:13] Jared Cooney Horvath: Yes. So, there's interesting data showing that any country that invests less in ed tech improves more on every outcome. And it's a very tight correlation, something like 0.6. It's very good. So, basically the more your country invests in tech, the worse your kids are going to perform on these international tests, which is another correlation here.

 

And somebody asked me the other day, well, how do you explain countries like Singapore or Japan or South Korea? These countries that are always performing better than everyone else. And yet they embrace tech. If you go look at the scores, they are showing the exact same patterns as everyone else.

 

The kids who use tech more in those countries are doing worse. It's just, it turns out they're talking about tech. They're just not using it in school.

 

They're using it out of school more than in school. And I think another big thing to take home here, and this is an interesting one for your listeners. If you go and try and look this data up now, it will look as though our kids have been performing about the same every year.

 

You'll think to yourself, hmm, they may be dropping a point or two a year. What is this dude talking about? There's something very important that we do with these tests that no one talks about. It's called ‘renorming’.

 

Basically, what happens is every time we give the test, we reset the scores to make things look equivalent across years. Let's say a bunch of kids took it this year and they got 90 percent. A bunch of kids next year take the same test, and they only get 80 percent.

 

What we're going to do is we're going to give them a bonus 10 percent. We're going to call their 80, 90 to make it look the same as last year. That way we can compare year on year whether we should be or not.

 

When people look at these tests to say it doesn't look like it's changing all that much, that's because no one accounts for the ‘renorming’. When you go in and you do the data analysis, and some people have thankfully done this, when these tests moved online. So anytime a test moves digital, like the PISA in 2015 became an online test.

 

The TIMSS in 2018 became an online test. Scores drop significantly for everyone. As soon as you bring the testing from pen and paper to online, you can expect a 14 to 20 percent drop in every single person who takes it.

 

But they just eliminate that, so we don't notice it. And the next year they do lower and the next year. So, there are more drops than most people would assume just by looking at the pure data.

 

Once you do your digging, you realise there's a lot. It's only getting worse. The more we keep relying on tech to kind of do this and learn for it and perform it, everyone is just suffering from this one.

 

[00:15:38] Anna Stokke: And this is, we do have access to data over a lot of years, right? Like how long has EdTech been around? Like 60 years or something.

 

[00:15:47] Jared Cooney Horvath: ‘62 was the first meta-analysis.

 

[00:15:49] Anna Stokke: It always feels with technology that it's new. It just always feels like that.

But actually, we do have a lot of data. And you wrote about this meta-analysis of over 21,000 studies. There was a positive effect size, 0.29, right? That's actually not that much.

 

When you're thinking about meaningful impact in education, you want, like 0.4, positive effect size of at least 0.4.

 

[00:16:16] Jared Cooney Horvath: That's how the researchers will get you. Is a lot of people, as soon as you start talking about tech, they're like, well, here's a meta-analysis that says it has a positive effect size.

 

This study says, and for the listeners who don't know, an effect size basically is just if you run research on something. You can give it a number. Zero means your research had no impact.

 

Like, let's say I'm testing a drug. Everyone takes a drug. No one had any impact.

 

That's a zero-effect size. As soon as people start getting healthy from my drug, now I start getting a positive effect size. And the higher the positive effect, the better impact I'm having.

 

But you can also go negative. If I give you a drug and people start getting sicker, I can go into the negatives as well. So, it's just basically a measure of how much impact that I have.

 

So, people will always come out and say, look, you have a positive effect size of digital tech. That means it helps learning. The dirty little secret of educational research that no one ever talks about is everything works in education.

 

Human beings are wired to learn. If you change your socks, people will learn. If you lock a kid in a cage, they will learn.

 

So, John Hattie, he's a statistician. He recently meta-analysed over 350,000 effect sizes in education. Over 95% were positive.

 

And the only ones that were negative were very specific things that you would guess, like prolonged illness. Will harm learning. Divorce in the family will harm learning.

 

Those kind of things. But everything else is positive. In medicine, we can say positive, good, negative, bad, zero, neutral.

 

In education, you can't say that because everything is positive, which means everything works. So, we need a different baseline in education. Now, what baseline you choose, that's going to be different depending on who you talk to.

 

But what you'll see almost universally is it's going to sit between positive 0.4 and positive 0.5. You need to be sitting in that realm there if you want to say we're having a positive impact on education. Otherwise, you're doing worse than just status quo. If you even go with just the easiest one, let's say 0.4 is our new baseline, you see that tech has an effect size of positive 0.29. You're under the level you need to be at to say we're actually having a positive impact.

 

You would be better off doing nothing with tech and you would still see a 0.4 growth if you never touch those tools. So that's where we start to see that they can lie to you by saying we have a positive effect size. But if you know a little bit more about educational research, you can say, yeah, at what level? And now you can push back and say, I'm sorry, that's not good enough.

 

[00:18:35] Anna Stokke: I think that's good for listeners to know because there are a lot of people, of course, selling ed tech. So, we do have to be really careful. It always sounds really good and really shiny and lots of great fads.

 

But you have to be really careful about what people are telling you about effect size.

 

[00:18:53] Jared Cooney Horvath: I think, too, what you're going to find with ed tech vendors, they're going to pool effect sizes from K through 12 and university. They always get pooled together.

 

But we know learning at the university level is different than learning at K through 12, simply because once you hit college, you develop what are called self-regulated learning skills. You can do more on your own than kids before they enter the university system. If you look at there was a new meta-analysis published on ChatGPT in education, and it says we have a positive effect size of 0.68. Everyone use it.

 

Yay, 0.68. That's over my 0.4. That sounds awesome. Only five papers in that entire meta-analysis were done with K through 12 kids. The rest were done with university or adults.

 

When you only take out those five studies with K through 12 students, the impact of ChatGPT on learning was 0.16, way below the 0.4 we need. The next way they're going to try and trick you is they're going to pool data from your kids with university kids, where we know you're always going to get better impact of tech. And they're going to say, look, now use it across the board.

 

And it's like, no, you need to kind of divvy those out if you want to make sense of this.

 

[00:19:58] Anna Stokke: And I would also question like what kind of tech they're talking about. For example, I would think something like, Anki.

 

Have you heard of that one? It's like a flashcard app that does space practise. Basically, you can put your flashcards into it, the things that you want to test yourself on, and you can do retrieval practise, and it will do space practise or interleave practise, and it will adjust based on how well you're doing. And so, for a university student, that is actually a really useful tool.

 

But is it a useful tool for a grade four student? They're not mature enough to use something like that, right? I'd kind of question even, there are a lot of different types of ed tech. Some could be really helpful, and some aren't. Actually on that note, like what forms of ed tech actually do work?

 

[00:20:49] Jared Cooney Horvath: You'd see there are basically three levels where ed tech will work, but they're a lot less powerful than people assume.

 

So, one is exactly what you're talking about is what we're going to call intelligent tutoring. Some programme that gives you questions, immediate feedback, organises questions in a way that allows you basically to deal with recall, feedback, spacing, interleaving, that brings all those together. That's basically what you would call drill and kill, which everyone used to hate in school.

 

And if you're a teacher who does drill and kill, you're evil. But if you're an ed tech product that does drill and kill, we love it. And that actually works.

 

The only thing to remember about that is that type of training works really well at surface based learning. Now, surface based learning is key. You need that.

 

You can't go deep without it, but that's not where you want your learning to end. Surface is where you start your learning. Then you have to start taking it deep.

 

How do you conceptualise how you apply your knowledge? That's where that tool no longer has any impact. So, if you're still at the surface, if you're just learning something brand new time, time to lock those facts down, bam, that's one place where ed tech can be useful. The next is learning disorder remediation.

 

If you've got someone who's struggling with phonemic awareness, you can use tech to just drill them on phonemes. And you see why this works is because it's doing the same thing as intelligent tutoring. It's basically just let's get into drill and kill.

 

We need to lock some skills down so we can do stuff with them later. So again, if you're at the surface with learning disorders, still works. The third aspect that tends to work is what we call procedural training.

 

So basically, you have declarative learning, which is learning of facts, and you have procedural learning, which is learning how to do things. Learning the circulatory system, declarative. Learning how to cut open a heart to perform surgery, procedural.

 

If you have tools aligned directly to procedural training, it tends to work as well. So, like if you're an F1 driver and you can't shut down Monaco to practise your laps, use tech to drive that simulation. If you're a surgeon and you don't want to kill a patient, if you're a pilot learning how to land an aircraft, use a procedural simulation.

 

They work. Now, the trick to recognise about all these three things is one. They do not work as well as doing these things live and in person.

 

If you could land an aeroplane, you would learn more in an actual aeroplane than a simulator. If you could make your own flashcards and understand how to do interleaving and spacing yourself, that would be better than a tool. But a lot of people don't.

 

They're not as good as other methods. They just, you use them because we have them. The other thing to recognise is what's called the transfer issue.

 

The very first intelligent tutor, believe it or not, was created in 1926. That was the first time someone built a machine that basically asked questions, gave feedback, asked questions, gave feedback. It went nowhere.

 

The next guy to do it was Skinner in the forties. His went nowhere. Someone did it in the sixties.

 

Theirs went nowhere. We see versions of it in the eighties through the nineties. It didn't really take off till now.

 

And people keep wondering, why did it never take off in the past? All of those people will tell you, transfer issue. Go back to the very first machine. The guy stopped working the machine.

 

We have a letter from him that he wrote to B.F. Skinner where he says, kids did great when they were using my machine. But as soon as I ask him the same exact question in a different context off of the machine, they can no longer answer the question. So, it was like they couldn't move their skills from my programme off of the programme.

 

And we see that same thing now. If you train, do a lot of training on a programme, you also have to supplement that with offline training. Otherwise, you will be very good.

 

So long as you got your flashcards on your phone, you're doing fine. I ask you that question now on a test in a quiet room. You won't be able to answer it because you've tied it to the wrong context.

 

You've got to be very careful about the transfer issue when you're using these online training tools and practising getting it offline as much as you can.

 

[00:24:32] Anna Stokke: A couple of things there. The first thing is the human teacher is always better than the ed tech is what you're saying.

 

And I 100% believe that. I think maybe you're saying that there might be times that when supplementing might be OK.

 

[00:24:50] Jared Cooney Horvath: Yeah. If you're at home and you're studying and you don't have a teacher there to help you and you go to town.

 

[00:24:55] Anna Stokke: Here's an example. We have online homework at university.

 

Now, is online homework better than written homework that's going to be marked by a marker? Absolutely not. It'll always be better if it's marked by a marker. But if your budget doesn't allow you to do that, I think it's better that you use the online homework instead of no homework feedback at all.

 

[00:25:17] Jared Cooney Horvath: I always say my rule of thumb when it comes to tech is something is going to be better than nothing. If you don't have a choice, if it's tech or nothing, of course you use tech. Let's imagine there's a kid with a learning disability who literally cannot engage with learning unless they have tech.

 

I'm not an idiot. Bring in tech. If let's imagine there's some global pandemic and we shut schools down.

 

I know that'll never happen. Of course, we use tech. Something is going to be better than nothing.

 

But kind of, as you said, if you are lucky enough, if you're fortunate enough to have the choice between two somethings, always pick the something that aligns better with your goal. And if your goal is learning, that's something you pick will almost never be tech. Tech will never be, by other methods we have, if learning is your goal.

 

Because I think you made another good point there is I think what's a good term for it? Like administrative is a concern. It's like, look, it takes time to grade papers. It takes time to edit work.

 

So administrative work, that is an outcome that some people want, in which case, if that's your primary goal, use the tool that makes it easier for you. But again, if your ultimate goal is learning, then sometimes ease of administration might have to take a backseat to these harder tools, these more analogue tools, because it's going to help your goal of learning a little bit more.

 

[00:26:27] Anna Stokke: I'm going to ask one other situation. In math, there can be a wide range of students in the classroom. Making it difficult for the teacher to cater to every student's need. Let's say you have some students who are very advanced and they're always ahead of the class and these students could use extra challenge.

 

And two things kind of sometimes happen in that situation. One thing can be that sometimes, particularly in elementary, the teachers are not like they're generalists. They're not really trained in math.

 

And so, it can be hard for them. To come up with challenging problems for these students. And then the other thing is they'd have to work with those students somehow, and they're trying to help all the other kids.

 

If you have a programme that's set up pretty well and it's giving contest problems and things like that, and students can work through those instead of doing no math, that might be a good situation. Do you think?

 

[00:27:25] Jared Cooney Horvath: Look, there's no right or wrong to any of this is Neil Postman said something really cool back in the 80s. I think that guy was the best thinker.

 

I wish I could have met that guy. He said, I have no doubt that a good teacher can use tech to help kids learn in an incredible way. I also have no doubt that that same good teacher could use a pen and paper to help kids learn.

 

What you're talking about is good teaching versus time poor or less constructive teachers. In which case, yeah, look, you can't work if you're using it. Well, sure.

 

Will it work if you're using it poorly? Sure. Now you get to say, what is my context? Exactly. As you said, as look, if I know my 50 minutes for this class, I need to work with remediation with these three kids.

 

Do I leave these other 27 kids doing nothing? Of course not. Something is better than nothing. Then you might have to bring in tech.

 

That's where it's, it's just going to be a context by context thing. And I wouldn't argue, but I'd say go back to the nineties. If you were lucky enough to be in school in the nineties, congratulations, you caught the golden age of education.

 

That was right pre-tech. That was when we peaked on every measure of everything. So, we're doing, that was fun.

 

We were, had the same problems back then, but we didn't have tech to lean on. So, what were teachers doing back then? You get super creative when you start to realise, wait a second. Textbooks exist for a reason.

 

Those are still completely useful tools. Having outreach programmes, allowing schools to stream during certain classes. There are reasons why these things existed that might feel archaic today, but no man, they worked an absolute treat and they're working better than what we're trying to do now.

 

I'd say if we just look in the past, our ingenuity is more than we give ourselves credit for.

 

I always say, remember that the atom bomb was built without any computers. We built the aqueducts. We built the mechanical clock.

 

The amount of things humans were able to do without the help of computers. We've just basically stopped thinking. We even made it to the moon using the only computer they had for the moon landing.

 

It has less computing power than one of those singing birthday cards today. This is what we can do. And we say, hey, we're humans.

 

We got a problem. Let's solve it. And I think right now we're just, we're so used to saying, hey, we're humans.

 

We have a problem. Let's let the computer solve it. I was like, ah, how much better could we be if we just started relying on ourselves again? That's an interesting side story here.

 

Across the U S over the next year, you're going to see a couple of states are going to start enacting laws that allow opt out for parents. So, if you've got a kid in a K through 12 school and you do not want them on tech, you are allowed legally to opt them out of non-essential tech. If they have to take a test online, fine, whatever, but nothing else.

 

A couple of schools in Australia have already tried that. And what they found was if you have a percentage of kids who opt out, those teachers now basically have to do two things. They have to have analogue versions of everything and digital versions of everything.

 

And almost to a person, all of those teachers after a couple of months will stop doing the digital. And every kid now has to do the analogue because the teachers remember how much better it was when they were actually thinking, what do I need to actually ask you? What answer are you actually giving me? When you force them to do both analogue and digital, even the teachers were like, yeah, it's more work, but gosh, the kids are doing much better with it and they go to analogue. So, I have the same hope that might happen here.

 

We remember what we're capable of.

 

[00:30:37] Anna Stokke: You have to have everybody doing it.

 

[00:30:39] Jared Cooney Horvath: Yeah.

 

[00:30:40] Anna Stokke: And I'll tell you a story. We as parents were kind of the ones who were the last parents to let their kids get a phone. And we didn't want our kids on Instagram and the social media platforms and stuff like that.

 

And they were kind of annoyed at us at the time. And it's all fine. They were good kids.

 

But I asked them not too long ago, like they're grown up now, right? They're in university. Are you glad we didn't let you go on Instagram? And they said, no, absolutely not. I was surprised by that because I thought, you know, this will pay off later.

 

And maybe it did like in ways that they don't see. But their answer to it was, well, all our friends were on Instagram and we missed out on a lot because of it.

 

[00:31:25] Jared Cooney Horvath: I guess when you get up to the forties and fifties, I'd be lucky if I remember any of my friends from high school.

 

I think things change. Hopefully, as you get older and you start to remember it, there's more to high school than what happens in front of those lockers.

 

[00:31:37] Anna Stokke: You know what? They're pretty good at concentrating and focussing. So maybe it helped.

 

[00:31:41] Jared Cooney Horvath: They'll pay dividends in the end when they're around their first work space and they're looking around and other kids or other employees can barely string a sentence together and they're doing all right. I think they'll say, all right, that was a good idea.

 

[00:31:52] Anna Stokke: I want to talk about, so I don't sound like a tech apologist because I said some things positive about tech, because I'm not completely anti-tech like you. You call yourself pro-education and that's the same with me. I want to think about the situation and whether it's good in that situation or not.

 

But I want to talk about offloading because this is something that has been coming up in math forever. I can remember years ago teaching a class and a student was saying to me, yeah, but Conrad Wolfram says that you can just offload all these basic things to technology and then people can concentrate on creative problem solving. OK, I've heard this argument for years and now it's worse because AI can do almost anything.

 

The argument tends to be that in math, that students don't need to learn procedures. They don't need to learn facts like this ends up being a waste of time. And we can use that time by concentrating on creative problem solving.

 

What do you think about that?

 

[00:32:57] Jared Cooney Horvath: That sounds to me like an expert talking about novices in a way they absolutely shouldn't. We call it the expert blind spot. The better you get at any one thing, the harder it becomes for you to teach that thing because you forget the process it took to get there and you try and teach kids where you're at now versus where you were 20 years ago.

 

That's why training and pedagogy when teachers train in pedagogy, they become good teachers. When you make experts teachers, they're horrible teachers because you start to realise teaching is a craft in and of itself. But let's go back.

 

Let me just break down real fast for you. Creativity. What is creativity? Creativity, believe it or not, is nothing more nor less than problem solving.

 

It's you have a set of facts or knowledge in your mind. You have an unknown problem set over here. And creativity is nothing more nor less than the mixing of matching of your current knowledge sets to this unknown set to create what we call a new knowledge structure.

 

Easy way to think about it. I think about it like Tangrams. If you ever played that game, you have all these little shapes over here.

 

Then you have this little grey figure. You need to rearrange your shapes to match this figure. And when you do, you now have a new way of understanding how these shapes work.

 

By the way, I just want to make clear. I say nothing more nor less because there's nothing more. I mean, I've talked to all the creativity researchers in the world.

 

No one has gone beyond that with their definition. But I also say nothing less because I think problem solving is probably peak human. I think that's the best thing we do.

 

I put that at the pinnacle of what we can achieve. But one of the funniest things that most people don't know about this creativity, critical thinking, all these higher order skills, the vast majority of their impact come when you're not paying attention, using subconscious processes when you step away. So, let's say you're working on a creative problem.

 

Let's say you do it for two, three hours. You can't come up with a solution. You will leave.

 

You'll go for a walk. You'll go cook dinner. You'll go take a nap, take a shower, do whatever.

 

When you leave the problem, that's when your biology gets to work. Now what happens is completely devoid of any input from you. Your brain is going to start to say, well, look, you had that problem.

 

You were thinking about these facts. You missed all of these. Remember in 1990, you learn this.

 

Remember in 2012? What if that fits here? And your brain will get to work. An hour of your brain doing its own thing. Every once in a while, your brain's going to go, oh, I just found a match that you missed.

 

And you're going to have this moment we call insight where you're going to be cooking dinner and all of a sudden you're going to go, oh, shoot, I have an idea. Wait, someone give me a piece of paper. That's your biology doing its thing.

 

That's why you get your best, go to bed with a problem, wake up with solution. You do your best thinking when you stop thinking, basically. Now the problem is, is we call that diffuse thinking mode.

 

All that diffuse thinking is subconscious, which means that form of thinking can only use facts. Ideas, knowledge, information that you have embedded within your brain. Your brain cannot access the internet and look things up.

 

Your creativity will always be limited by your knowledge, by what you know, not what you can access, not what you can have a machine do, what you have embedded within your biological system. So, people think that that's pretty crazy.

 

If we give, and we did this, the PISA did this back in 2018, they test 90 plus countries on factual knowledge. What do you know about math? What do you know about English? Reading, writing, what do you know about science? Then they gave all those kids a creativity test and then they correlated the two. The correlation between factual knowledge and creativity was 0.92, almost perfect, which meant the more your kids know, the more creative they will be. So now go back to the issue of offloading.

 

If we use tools to stop learning stuff, to stop learning procedures, to stop learning facts, don't worry about facts, just look it up. You will kill the very thing you were hoping to free up time to do. There will be no creativity.

 

There will be a lot of copying, pasting, but once you enter in diffuse thinking mode, your brain goes, well, you don't know anything, so I got nothing to do. So go enjoy your walk. You're never going to come up with anything new.

 

So, all these higher order thinking skills are dependent, they emerge from lower order knowledge. And the more we use tools to offload that lower order knowledge, this is remember when I said Gen Z is worse than us and everything? Creativity, critical thinking included. Gen A is worse than them.

 

Because they're learning less, they are performing less on these higher order skills. So yeah, number one way to kill creativity is to make sure we use tools to offload learning. The worst way to do it.

 

[00:37:08] Anna Stokke: I need to carry you around on my shoulder.

 

[00:37:11] Jared Cooney Horvath: Anytime that argument comes up, we can say, oh, wait a second.

 

[00:37:14] Anna Stokke: You describe that so eloquently. It's just something I just know. I know that you can't be creative in math without knowing a lot of stuff, but you just describe it so well. So, I really appreciate that.

 

And I'm going to be sharing that clip all over the place.

 

[00:37:29] Jared Cooney Horvath: Here's an easy one that I always use. So, I'm doing a documentary right now on genius.

Where does it come from? Is it born? Is it made? I love it. But one of the stats that I just love, I keep coming across is you will never find anyone who reaches the level of eminence or genius who hasn't done a bucket load of practise drilling learning. So basically, the average is about 12 years.

 

We say from beginning of learning to win a master, a genius will produce their first work. And then another 12 years after that, until they start to produce their best work. So, anyone who's like, oh, some people are just more creative than others.

 

No, you go to anyone who's done any good work and say, how long did you spend training? They will say longer than you. You, no one has ever skipped that phase. And the people who actually embrace that phase of learning, it's hard to learn math.

 

There's a lot of drilling. There's a lot of practise, but the people who do it are the ones who are the most creative in mathematics. As they get older, you'll never find a genius mathematician, a Nobel prize winner who hasn't spent decades learning the basics first.

 

You cannot skip that step.

 

[00:38:33] Anna Stokke: Never. Absolutely. Thank you. So, a little more about AI. You have a great chapter on this.

 

It's titled AI, the tool nobody asked for, solving problems nobody had. What can you tell us about AI? Give us all your thoughts on AI.

 

[00:38:50] Jared Cooney Horvath: This tool has no reason to exist. I've had the incredible displeasure of speaking with a lot of people in the AI development field. No, not a single one of them can tell me what this tool was built for. There used to be a time, when if I created a tool, it had to solve a problem.

 

And not only did it have to solve a problem, it had to solve it well, better than anything else. And I had to prove it to you. I had to say, look, my new soap will clean that stain off your shirt.

 

My new fridge will keep your lettuce crisp. Trust me. And if it didn't do it, no one is going to buy it.

 

I failed. AI, especially large language models. These things come out.

 

Nobody knows what they're for. If you ask the developers, they said, we just wanted to see if we could do it. It was an experiment.

 

They released it to the public with the express understanding that you tell us what it's good for. We don't know what you're going to do with it. Show us what you're going to do with it.

 

Since when did we become product testers? And then the worst was they shoved it into schools immediately and said, not use this in this way for your kids to learn, but use it and tell us how kids are learning. If I invented a drug and I just said, “Hey, go give this to kids and let's see what happens.” No one would ever do that.

 

Yet. That's exactly what we did with ChatGPT and all these LLMs. And it is harming kids learning reparably in some cases, I worry.

 

 

It is not a learning tool. The most important thing to recognise about LLMs, AI, and this isn't even true to an extent, A, they don't exist for any reason, but B, if it can be said that they exist for any reason, it is for expert production. They are production tools.

 

If you are an expert and you already know how to do something, you just don't have the time to do it. You use AI to get that done for you. And the only reason it works is because as an expert, you can vet the output immediately.

 

So, you know, if it's bullshitting you or not. So, like I'm really good at stats, but it might take me two hours to do stats when I could use this machine to do stats in two minutes, do it. So, I do, and I can only do that effectively because as soon as it gives me an answer, I can say that number is totally wrong.

 

What? Or, Oh no, that looks correct. Let's go with that. This is not a learning tool.

 

When we mistake production for experts for learning with novices, we make a huge error. So, T this tool to kids, they will use it to offload thinking, offload ability, and they cannot vet the output. So, all they can do is copy and paste it and send it back to us.

 

And now at the end of the day, they will have learned nothing. So, this tool, if you can say it's used for anything, should only be used by experts who know enough to vet the output when they're just trying to do things they don't care about pure offloading. This is not a learning tool.

 

No kid ever need touch it. If you want kids to learn how to read, to learn how to write, to learn how to comprehend, to learn how to make arguments, this is not a tool designed for any of that. It is designed to make sure you don't have to do any of that.

 

No kid should ever touch it. We can have a debate whether or not teachers should use it, but I never heard a valid argument why a kid would ever need to touch AI. For those who are now thinking to themselves, well, how will my kids ever be prepared for a job in the future? Two things to remember about that.

 

One, we have data for decades that shows the more kids are trained on tech, the worse their technological literacy becomes. When you train, we make a mistake of thinking using a tool is the same as understanding how to use a tool. They're two very different things.

 

And when we train kids on tools, they can use them just fine. They use them more than anyone else, but they don't understand them any better. You'd be much rather off teaching, learning, thinking patterns, what we do in a traditional K through 12 education, and then bring the tool in and say, hey, by the way, you know, you can use this to help you with that.

 

Oh, there you go. That's why if you look, man, Gen Z has the same basic digital literacy as Gen X. So basically, some of our parents have the same digital literacy as our kids. Why? Because you don't need to be trained on these tools.

 

You need to train in how to think and then bring the tool in. So that's number one. Number two, I would always say is this is since the iPhone, digital technology has been literally developed to be used by kids.

 

Steve Jobs didn't even lie. He said, I wanted babies to be able to use this phone. That's how easy digital tools are now.

 

Are you telling me we really need to devote K through 12 education time to training kids on a tool like AI, which is nothing more than a box you type words into? No, it is the easiest thing in the world to use. I learned it overnight. Simple.

 

We don't need to teach them how to use tools. We need to teach them how to think. And then we introduce the tools later.

 

That is what we did in the past. That's always worked. I don't know why that wouldn't work now.

 

[00:43:19] Anna Stokke: I think the main point you made there that this is not something for novices. Like don't give this tool to people who are learning things because they won't learn. I can see it too, like in my students and I can tell when they've used AI.

 

The problem is it's not the sort of thing that you can prove. Whereas it used to be the kind of cheating that you would see if you gave an assignment, a student might cheat and you might find that exact same solution somewhere else or another student did it, but now you can't tell. And this is kind of why I think AI might end up doing us a bit of a favour because now I'm even hearing like my colleagues in English are saying they are not giving any take-home assessments anymore.

 

Like everything will be done in class because you can't tell if the student has used AI or not. It's not plagiarism.

 

[00:44:17] Jared Cooney Horvath: I always say you guys are in the hardest realm, right? So, at the university sector, we see our kids eight times a year, 10 times a year for lucky when they hand us a final paper.

 

I don't know student A from student B. So, if you're using AI, mate, my hands are clean. I don't know. But if you're in K-12, that's where we get to go back to deep pedagogy and you can start to say, well, wait a second.

 

I have a process. I'm not just looking for what you've written. At university, I trust you can write.

 

I just need you to tell me some stuff. K-12, I don't trust you. So, if we're doing an essay, say in sixth grade, we're going to start by just write a paragraph, right? Write your thesis statement.

 

Now let's edit that. Next week, let's work on our first paragraph. Now let's edit that together.

 

By the time you get to the final draft as a K-12 teacher, you should already have the full trajectory of that kid's thinking and development. So that if now magically they come in with this perfect essay at the end, you can comfortably say, that is AI, dude. That is not you.

 

I know you. I know your work. That's nonsense.

 

At the university, we just don't have that. So, we have to come up with other ways to say, well, forget it. We're doing in-class essays now.

 

That's how it's going to be.

 

[00:42:23] Anna Stokke: Are they using ChatGPT in schools?

 

[00:45:27] Jared Cooney Horvath: Oh my gosh, across the board. They are district after district.

And it's usually the people who aren't teachers are basically forcing teachers to use it. They're saying, how do you incorporate this? We don't want to be left behind. AI is everywhere.

 

Somebody said this to the other day, and it's totally true is you could use AI to help learning. Like what if kids started typing in quotes like, don't give me an answer, give me a question and a prompt, and then I will think about it. And you vet that.

 

My counter to that is always, I could lose my car to increase my fitness. If I put my car into neutral and decided to push it to work today, I could get stronger, faster, better.

 

Of course, I'm never going to do that. No one is ever going to do that. So yes, ChatGPT could be used by students to increase their learning, but I don't know what world anyone lives in where they think a teenager is going to go home on a Friday night and be like, don't give me the answers, ChatGPT, give me questions so I can think.

 

It makes no sense. When you actually track how kids are using it, it's basically just a cheating tool. Like I've seen no data that says anything other than kids use it to avoid the process of thinking.

 

[00:46:29] Anna Stokke: It's not even always intentional. I think that students get in a position where they have a lot of things to do. And there's one thing that they could maybe finish with ChatGPT and they'll learn it later, but then they never do.

 

[00:46:42] Jared Cooney Horvath: It's not a learning tool. We do not embed it into our pedagogy. We don't say, oh, instead of reading a book, why don't you ask ChatGPT for a summary? Scrap it.

 

If you're encouraging your kids to use AI in any way that's not 100% guided by you and I haven't heard a good way to use it, man, you better come with some really deep justification because it makes no sense to me.

 

[00:47:02] Anna Stokke: Let's talk about multitasking. You describe attention as the gateway to learning.

 

How do phones and multitasking or screens and multitasking interfere with that?

 

[00:47:14] Jared Cooney Horvath: We've had some data. Now we can start to look at mechanisms and this is one of them. Attention is everything when it comes to learning and memory.

 

You have a part of your brain called the hippocampus, which is basically your gateway to memory. Anything you've ever learned declarative will go through your hippocampus. Attention, think of it like a filter that determines what's going to make it into your hippocampus and what's going to get blocked out.

 

So, there is way too much stuff bombarding your brain at any one moment. You got sound, smells, lights, everything coming in. We have to set up our attention filter that says this is relevant, let it in.

 

This is all irrelevant, block it all out. The problem with human beings is we only have one attention filter slot, the little nugget here behind our left eye. So that means if we ever want to try to do two things simultaneously, we can't.

 

The best we can do is start one, switch out our filter, jump to this one, switch out our filter, jump to this one. Now, moving your filter between tasks takes a big toll. It makes you slower, kills your accuracy, kills your memory.

 

It's just probably I always say multitasking is the worst thing we can do for learning and memory. Cut to technology. Per year, and I'm talking K through 12 now, teens and tweens, kids will use digital technology for learning purposes about 450 hours per year.

 

That includes schoolwork time and homework time that they're going to use computers for learning. I know it's a tonne. Those same kids will use the exact same machine to passively consume rapidly switching media content for over 2,500 hours every year.

 

So, this means this one screen in front of me, basically for 2,500 hours every year, kids are going to train themselves to do one thing on it. Multitask. They're going to jump from this tab to this tab, to this thing, to this thing.

 

That's how it works. It's called the attention economy. These things are literally designed now that I didn't even hide it to continuously flip to shift your attention again and again and again and again to make you multitask.

 

Now when I say, cool, you're going to spend 450 hours per year learning on this after you spent 2,500 hours training yourself to do the one thing I can't have you doing if you're going to learn. Is it any wonder that when we use tech for learning, kids make it on average only six minutes before they start multitasking. And if you use a computer in class with a teacher present 38 minutes of every hour, they will be off task multitasking.

 

They are not learning. Here's where you start to see it's an intractable mechanism. If computers only ever existed in the classroom and were only ever used for learning, they may have been useful.

 

I have other reasons to assume they never would have been. We can talk about that later, but they may have been useful. But the second the computer entered the home and more importantly, the pocket, it stopped being a tool for learning.

 

It started being a tool for rapid entertainment. Now, when we try and pretend that we try and shoehorn this function in, hey, kids, I need you to do this. Now, the vast majority of them will never be able to overcome that hurdle and actually lock in their attention enough to do any work on it.

 

That's why when you move tests online, scores go down. When you move reading online, comprehension goes down. When you move anything online, it's called the mode effect.

 

Performance will go down. And multitasking is the biggest driver of that. We'll just immediately start looking for things we can do on there.

 

[00:50:15] Anna Stokke: That's kind of the main issue. What if you can't access the Internet?

 

[00:50:22] Jared Cooney Horvath: On the device if you have a locked system, you have a much better shot of aligning attention. Go back to like that haptic feedback machine or a flight simulator.

 

There's nothing else to do. I can keep my attention much more focused. At that stage, you're going to move into some other biological problems, though, that are still going to make digital worse than live and in person.

 

One of them is going to be the empathy issue. Empathy is one of the biggest drivers of learning. We know it has an effect size of like point six three.

 

It's good. So, you can then ask, what the heck is empathy? Most people think empathy is an emotion. It's a feeling.

 

It's something I have. I'm empathetic. You do not have empathy.

 

Nobody has empathy. Empathy is not a personal thing. It's not an emotion.

 

It's not an individual thing. Empathy, by its very definition, is transpersonal. Empathy occurs when two sets of biology start to do what we call synchronisation.

 

Basically, if you're empathising with another human being, your heartbeat start to go at the same time. You start to breathe at the same pace. If I could image your brain, your brain start to mirror each other.

 

So, empathy cannot happen by a person. It needs two people. No one's controlling it.

 

You just synchronise with another person. Congratulations. Why would empathy be so important for learning? When a student synchronises with a teacher, another human being, in that moment, they start to think like the teacher.

 

All of a sudden, understanding becomes so much easier. You start to get the arguments. You start to understand the nuance.

 

You start to predict where they're going next because you're on their exact same wavelength or flip it. When a teacher and a student are empathising, if a student starts to struggle, that empathy allows the teacher to understand exactly what the kid needs to move forward. That's why it's one of the biggest drivers of learning.

 

It drives understanding and helps the teacher get to the information they need to move the kid through struggle. Cut to technology. Empathy, by definition, requires two sets of biology.

 

What does a computer not have? You cannot empathise with a computer. One of the biggest drivers of learning we have requires two sets of biologies, which is why we say human beings have evolved to learn from human beings. We have inborn mechanisms that say you will learn from others like you in this way.

 

And it's in there whether you want it or not. Computers do not allow that mechanism of learning to come through. So even if you have a computer that's not connected to the internet, that's a locked system, you will still learn less than if you were being taught that by another human being.

 

And we have the research, as stupid as this sounds, they do research where have a bunch of kids in a lecture hall listening to a lecture. You're not allowed to ask questions. You just got to be there, film that lecture, have another group of kids watching it live on a screen.

 

The kids in the room will learn more than the kids watching it on the screen, even though it's the same thing. No one can interact just by being there. You could start to empathise with the person.

 

You can start to resonate. You start to understand things better. So that's what I'd say is the next intractable mechanism is even if you can combat the multitasking issue, you're now going to be stuck with the empathy issue, which go back.

 

Remember when we were saying one aspect within which EdTech is useful is procedural simulation. That's because you don't need to resonate with someone for procedure. If I'm landing a plane, that's very different than listening to somebody describe history to me or describe a book to me.

 

So, you can see why you can get away with a little bit more when it comes to the procedural stuff. But again, man, if I could watch a human being actually open a heart, I'll learn more than a computer. If I could do it myself, like I will still empathise with somebody who's flying a plane if I could watch that versus a computer screen.

 

But so that's that becomes your next intractable issue. Until we learn to build human biology into computers, it'll never overcome that hurdle.

 

[00:53:53] Anna Stokke: You actually go through eight common arguments from tech evangelists and you give people advice. People have to get the book because this might really help you if you're trying to make changes and if you know that it's better to have less technology in your school. And so, you have eight common arguments.

 

I'll mention a couple that I think I've heard quite a lot. So one would be, well, digital devices are ubiquitous. Like everybody's doing it.

 

We don't want to be the school that doesn't have these things. What do you say to that?

 

[00:54:27] Jared Cooney Horvath: There’s a long-storied list of ubiquitous things that we do not put in schools. Drugs, driving, divorce, litigation, yoga.

 

We understand that we're not the only people teaching kids. They learn from each other. They learn from the parents.

 

They learn from the world they live in. As a K through 12 education, we have a very strict function. We don't teach everything.

 

We teach specific and useful things just because a tool exists doesn't mean it needs to exist in our classroom. So, once we go back to our values, why are we here? We're here to help kids take agency over their own thinking and learning. Then we can only bring in those tools that align best with that purpose, not the tools that exist, the tools that best serve our goal.

 

So, once you realign with values and you start to realise, wait a second, there's so much ubiquitous stuff we do not have in schools. This can be just another one of those things that we decide, look, it's not helping our cause. And they will learn it in those twenty-five hundred hours every year.

 

That's 90 plus days of their life. Every single year outside of the classroom, they'll be on these tools.

 

Trust me, they will learn them just fine. We do not need them to impede on what we're trying to do here.

 

[00:55:32] Anna Stokke: And then on the flip side of that, yeah, they're on devices all the time. These students learn differently. This is part of their world. That's another one we hear.

 

Right. This is part of their world. We have to bring in the technology because that's just what it is like for these students.

 

They learn differently than we did.

 

[00:55:50] Jared Cooney Horvath: When you try and adapt learning to a tool, you will always make a mistake. So, you've got to remember, learning is biological.

 

The most controversial statement some people hear today. But trust me, it makes sense once you hear it. All human beings learn in the exact same way.

 

Some people get tense, but bear with me here. All human beings digest in exactly the same way. No one ever has any problems with that.

 

Why? Because it's self-evident. No one has three stomachs and chews cud. Nobody spits acid.

 

It's obvious. Yeah. Now, by all means, some people would need support.

 

What we can put through that system might change. Some people can have gluten, some can't. But the process by which we digest, extract nutrients, excrete waste, exactly the same for everyone.

 

All human beings respire. We breathe in the exact same way. Of course, self-evident.

 

Again, some people might need support. What we choose to breathe might change, but the process stays the same. All people learn the same way.

 

Learning is biology. If you're a human, then you will learn in the exact same way as everyone else. You will go through surface learning.

 

You will go through deep learning. You will start to apply that through transfer. It's the same for all of us.

 

Again, the inputs might change. If you're blind, I can't show you a picture. Some people might need supports.

 

Again, if you're blind, maybe I have to build some Braille books for you. But the process by which we learn does not change. The fact that kids use technology doesn't mean they learn any different than us.

 

Doesn't mean their memory is now any different. Their attention is now any different. They're playing the same game we are.

 

So, we got to ask what is best for learning. And if we knew that kids off of tech in the nineties were way outperforming kids with tech now, is the right argument to say, let's try and adjust pedagogy more for the tech? Or is the smarter argument to say, let's go with what we know works best for human learning. And it ain't this.

 

And that's okay, man. You already spent your million dollars on it. Sunk costs fallacy.

 

Don't worry about it. Rather waste that million dollars and next year, spend a million dollars on teacher prep, then waste two million dollars because you're just trying to chase a losing bet. Kids do not learn any differently than anyone else over the last hundred fifty thousand years.

 

These tools, no software upgrade has changed biology. I'm sorry.

 

[00:57:55] Anna Stokke (57:55 – 58:02)

And the larger cost actually is the human cost. The students that actually don't learn as well as they could

 

[00:58:02] Jared Cooney Horvath (58:02 – 59:50)

It's worth noting preferences, too. All human beings learn the same way. Another interesting fact. Well, people then go, well, but the input changes.

 

Absolutely. But what we've always found is when you match pedagogy to the content, every student learns better, whether they want to or not. Now, the trick is we've been told, especially with tech, that we're supposed to match our pedagogy to the preferences of the student.

 

I like to do this. Therefore, Texas, I'll give you that. I like to learn in this way.

 

Cool. Human beings are very bad at knowing what's good for learning. We know it feels good. It's called the illusion of fluency. When something feels good to us, we assume we understand it better and we're learning more from it. And it's the complete opposite.

 

When something is incredibly difficult for you, that is almost always when you're learning more. You just don't even notice it. So, when we come back and say kids have tech and they love using tech, I don't care.

 

My kid loves eating cake. That does that mean I go, well, if they like cake, I better figure out how to make cake in every meal. No, we go, OK, what's best for your health? What's best for your well-being and fitness? It ain't cake.

 

Let's go with what's best, whether you like it or not. It's the same thing with learning. When we adapt to our kids' preferences, they will learn less than when we go back to say what works best for learning.

 

And it's not surprising. Everyone learns the tango better if they get up and do it. I don't care if you're a visual learner, an auditory learner, you like this, get up and dance.

 

You're going to learn it better. Everyone learns a book better when they read it than if they have somebody describe it to them. There are pedagogies that align with human learning, but they align best with the content.

 

And remember that you won't get suckered into that idea that because my kid likes TikTok, I better start teaching in two minute chunks. Worst thing you can do is try to appease their preferences. We're not good at knowing what really works for learning.

 

That's the teacher's job. That's what we do.

 

[00:59:50] Anna Stokke: And it's the hard stuff that you don't want to do.

 

It's a teacher's job to make you do those things, to make you practise those things, to get you to practise those things that you're not good at, because that's how you get better at them.

 

[01:00:01] Jared Cooney Horvath: When I was in Australia, they were talking about, the big movement was strength-based education. Let's do your education of what kids are good at.

 

And I understand the sentiment, but I always, my counter was always, no, education is a weaknesses-based institute. If you're already good at soccer, we ain't playing soccer. I'm here to give you breadth.

 

That's why it's a general K-12 education. And we're going to focus on what you don't like so that you have more abilities than just locking you in as a kid to, oh, you'll always just be soccer guy or whatever it's going to be.

 

[01:00:32] Anna Stokke: What sort of advice would you give to say a school leader and maybe also parents?

 

[01:00:38] Jared Cooney Horvath: I would say if we go start at kind of the school leader kind of level, and you're considering this stuff, step one is you put a moratorium on all new tech spending. For the next year, you do not buy any new tech and you spend this next year just doing a tech audit. See what people are using, see what people are not using.

 

And you're going to find, man, you have a lot of dusty tech in your place. That's just costing you money. Once you've kind of knocked lockdown, narrowed down what tech is being used, now you can start running usage on it.

 

And you can start to say, okay, compared to five years ago, before we had this tool to now, how are people using it? Are, is learning better? Are kids doing better? Ask surveys from your teachers. Do you think your kids are performing better? Ask the kids, go back to the grades, whatever it's going to be. And that's how you're going to weed out what tech is worth keeping and what's not.

 

And these little audits, they take about a year to get done. But my goodness, if you're like most schools, you're going to find about 95% of all your tech can leave tomorrow because it's either wasting space or it's harming learning compared to other tools that you already have present. And then you can start redistributing funds.

 

We have a whole chapter just for principals on that kind of stuff. The only way to make memories deeper is to recall them. That is the only mechanism by which we have to make deeper memories is we replay those memories again and again and again in our head.

 

Tech, by definition, asks you never to recall anything. It asks you to offload it and then look it up. A simple example.

 

What was your best friend's phone number growing up? 7407857. That's Dan O'Swanson. I haven't dialled that number in 35 years, but I know it all the same.

 

What is your best friend's phone number today? My best friend is Sergio. He lives over, I don't know his number. I know what buttons to press to get it, but I don't know it.

 

This is the difference between recall and retrieval. When you offload knowledge to a device, which is what most devices do, you will form a memory for how to find it. But you'll never form memory for what the thing actually is.

 

At the teacher level, how do we bring recall back into everything we do? We stop using tech to say, don't worry about that. I'll just look it up here and start going back to, yep, we're using this. Sorry, five minutes at the start of every class.

 

We're taking our pre-quiz five minutes at the end of every class. We're doing our exit pass notes. Instead of taking notes on the fly, we're going to take notes from memory at the end of class.

 

How do we build these things back in to say, yep, we're going to have to start using the brain to recall, recall, recall once again. And there's some good ideas about that and homework in the book as well. That's a whole chapter.

 

The number one bit of advice I give to students and parents, buy a printer. If you do nothing else, if you change nothing else about your kids schooling, about your home life, you change nothing but print every assignment out. So, homework, books, readings are everything on paper.

 

Your kids will learn more, comprehend better, and their performance will go up just by changing the medium they use. So now there are more things we can do to boost all of that stuff. But I say the easiest, the lowest hanging fruit is if you did nothing different with your life, but print everything out, you'll learn better.

 

I get it. My college freshmen are like, what's the number one study technique? I say, buy a printer, go plant a tree, print everything out. If you did nothing else different with your life, you're going to do better just if you have paper.

 

[01:03:41] Anna Stokke: What about an iPad? Like taking notes or writing on an iPad, still a problem?

 

[01:03:46] Jared Cooney Horvath: If you have a stylus on an iPad, you can get away with it. The reason why note-taking builds memory. So, a lot of people think note-taking is something you do while you learn.

 

Note-taking is the learning. It's the writing down of material, forces you to think about, organise how it all fits together. That is where your learning is happening.

 

When you type notes, you don't have to organise anything. You just type what you hear. So that's why if people take typed notes, they'll get copious amounts of notes.

 

They always get more than everyone else. But if you ask them, what did we learn today? They'll go, I don't know. I’ve got to go read my notes.

 

They don't learn anything. They learn the sound of the words. But when you're taking notes by hand, you know, you can't get every word.

 

You're already processing the meaning. That's why handwritten notes, by the way, are so gross. They got lines and arrows.

 

It shows that you're processing understanding. Technically on an iPad, if you had a stylus, you could do that, but you're still going to get a double hit there. One is most programmes on iPads will translate into text.

 

They will take your crappy handwriting, and they'll turn it into print for you. Which case you're basically done now is now it's not your notes anymore. It now we're back to kind of surface looking notes.

 

The other thing is to recognise is that human beings, a big aspect of our memory is spatial organisation and patterning. Anytime you form a memory for something new, that information will become stamped with a three-dimensional location. Where in space did this happen? So that's why, if you're an avid reader, you'll recognise this.

 

You read a book. You won't always be able to remember verbatim everything you read, but you will remember where in the book things happen. Like if you want to flip to that, oh, there was an awesome action scene.

 

Ah, hold on. Give me one second. I'll take you two seconds to get there.

 

Cause you know, where in space it happened. When you take notes by hand, you now have a clear three-dimensional location. That piece of paper, that notebook, that information ain't going nowhere.

 

That is a huge boon to your memory. You take notes with a stylus on an iPad, or now you're flipping between pages. There is no clear three-dimensional location anymore.

 

You lose depth. You lose an X, Y axis. You lose a part of memory.

 

So, you can still learn back. If you had to kind of say it, handwriting best, stylus second best, typing third. It sits in between.

 

It's better than typing, but it's still not going to be as good because of that spatial angle as writing things down. You'll have some students who recognise this too. In high school, in college, I'd take notes in my notebook and then I would do a little doodle on the top right corner of every page.

 

And each doodle was different. Later, all I'd have to do is think of the doodle and that entire page would come back to me. And that's the spatial.

 

Once you get one anchor in space, the space around it opens up in your memory, you're good to go. But on a screen, you just don't have space. Everything moves around and is never stuck in one location.

 

You lose that kind of game you can play there.

 

[01:06:21] Anna Stokke: Generally taking notes by hand is better. Reading a physical book is better.

 

And doing your assignments on paper is better. Get a printer.

 

[01:06:33] Jared Cooney Horvath: You're never going to go wrong if you do it analogue.

 

[01:06:35] Anna Stokke: We've had just a great conversation, and this is really interesting. We should think about closing off. So, you call yourself a Luddite.

 

I mean, it's kind of a pejorative. What's the story behind that?

 

[01:06:48] Jared Cooney Horvath: It's become a pejorative, but if you think about it, for those who don't know, Luddite, that's the term you use for anyone who hates technology. Because back in the day, during the industrial revolution in England, when a bunch of textile mills started to get machinery that could do power looms, all that stuff, all the old workers came in and started burning it.

 

They burned it under the name of fictitious leader named Ned Ludd. So, they were the Luddites. And now Luddite just means, oh, you hate technology.

 

You hate evolution. No, that completely misses the whole point of history. The Luddites weren't fighting against machines.

 

They used machines for a century. They were fighting for a way of life. They valued community and structure in the way it stood.

 

And they said, look, if you bring in tools, you're not just making textile processing faster, you will destroy all culture and way of life around it. They realised there's no such thing as a neutral tool. Every tool has a worldview embedded with it.

 

And if you don't reckon with what that's going to do to everything, you shouldn't be using that tool yet. So, a good kind of rule of thumb is this is another Neil Postmanism. He always says tools bring what's called ecological change.

 

If you have a garden and you bring a new beetle into that garden, you don't just have your garden plus a new beetle, you will have a whole new garden. The entire food chain has to change. The nutrients now have to change.

 

It's an ecological shift. Everything has to adapt. Tools are the same way.

 

When you introduce a tool to a culture, it's not just culture plus tool now. The whole culture is going to shift with it. And they realised, look, we don't like the change it's going to make.

 

We are pro community now. So, they were never fighting against tools. They were fighting for a way of life.

 

That's why I call myself a Luddite. I'm not fighting against tech. I don't give a rat's patootie about tech.

 

I'm fighting for learning. I believe school K through 12 education is a deeply human endeavour and we need to maintain that humanness of it. Otherwise, we're just going to bring in a tool without reckoning with its ultimate cost.

 

I would not be surprised if edtech has the power to completely kill K through 12 education as we know it. Some people cheer that they champion that they say that's great. No, every attempt we've had to use tech to change K through 12 education has failed.

 

Those kids learn less. Those kids come out less prepared for life. They are not better, stronger, faster.

 

They are worser, weaker, better, lot better, whatever the other analogue is there. But so that's all I'd say. Look, I'm a Luddite because I'm not anti-tech.

 

I'm pro-learning and that's what I'm willing to fight for. If you want to say I'm anti-evolution or anti-progress, sweet. Your progress is impeding on what I know works biologically.

 

I'm going to fight that every day of the week.

 

[01:09:24] Anna Stokke: You've issued a call to action at the end of your book to stand up to edtech and schools and let's do it.

 

[01:09:33] Jared Cooney Horvath: Tech companies aren't going to suffer, man.

 

If they don't have their fingers in our kids' schools. They're still going to be just fine.

 

[01:09:38] Anna Stokke: They don't need the kids.

 

[01:09:39] Jared Cooney Horvath:  No, they got nothing to worry about.

 

[01:09:41] Anna Stokke: They can take everything else, leave the kids alone.

 

[01:09:43] Jared Cooney Horvath: Bingo.

 

[01:09:44] Anna Stokke: Is there anything else we didn't cover today that you want to add?

 

[01:09:47] Jared Cooney Horvath: No, just thank you so much for being an advanced reader. You help us in your feedback and how it kind of sits with you. Makes me feel like we can actually make a difference with this.

 

So, I'm just hoping to get the book out far and wide.

 

[01:09:58] Anna Stokke: You bet. And I was in good company, by the way.

 

You had Hugh Grant, the actor. I actually have a colleague named Hugh Grant, so I have to say the actor. But Hugh Grant, the actor, actually endorsed your book. You had Dan Willingham, Paul Kirshner. I've had them both on the show. And Jonathan Haidt.

 

I mean, definitely people should get the book. It's a great book. And it's been an absolute honour to talk to you today.

 

Thank you so much.

 

[01:10:24] Jared Cooney Horvath: Thank you. And keep doing what you're doing. You're doing awesome work.

 

[01:10:26] Anna Stokke: Thank you. Thank you so much for listening. If you enjoy this podcast, please consider showing your support by leaving a five-star rating on Spotify or Apple Podcasts. Don't forget to subscribe on your favourite podcast app or on YouTube so you never miss an episode. You can stay connected with me on Instagram, Facebook, TikTok, X, Blue Sky or LinkedIn.

Anna Stokke

a.stokke@uwinnipeg.ca

Department of Mathematics & Statistics

The University of Winnipeg

515 Portage Avenue, Winnipeg, Manitoba

Canada R3B 2E9

204-786-9059

  • Twitter
  • YouTube
bottom of page