Breaking predictions, building memories

Marketing speaks with Jared Cooney Horvath, Harvard neuroscientist and lecturer at the University of Melbourne, about the brain and how to make marketing communications stick.

This article originally appeared in The Experience Issue, our February/March 2018 print edition of Marketing magazine.

In his early years of teaching, Jared Cooney Horvath was surprised at the amount of hype around the brain. It was, unfortunately, only hype.

“The brain was getting really sexy,” he recalls. “Everyone was coming into our school saying ‘you need to know about the brain and you’ve got to buy this product and you’ve got to teach your kids this’.” But none of them knew what they were talking about. It was a buzzword. So, he went to school to learn it himself, completing a masters in Mind, Brain and Education at Harvard and a PhD in Cognitive Neuroscience at the University of Melbourne.

While education still makes up the majority of his focus – he’s now a lecturer at the University of Melbourne – he is interested in how marketing and trying to understand how the brain functions can be applied to help brands understand consumers. He caught up with Marketing before speaking at Initiative’s ‘The Science of Stickiness’ event to explain how the brain really works and how branding must break predictions if it’s going to work at all.

Marketing: What are some key patterns or things marketers and educators should be thinking about the brain?

MK0218 200 cover experienceJared Cooney Horvath: The brain works in the complete opposite way that most people think.

The assumption is: information gets in, the brain triggers off systems and that’s your experience of the world – it’s organising that information as it’s coming in. In reality it is the complete and utter opposite. So we have this system – we call it ‘bottom up’ versus ‘top down’. ‘Bottom up’ is the information coming in. ‘Top down’ means at any one time your brain sends messages to say, ‘Here’s how I think the world should be, so please interpret it that way.’

If you’re looking at something but your brain thinks it should be something else, it will send signals back, change the way your neurons fire so that you see what you think you should see. We now know there’s a huge disconnection between the world out there and us, and what’s really driving the brain. Thinking, memory, learning, is essentially you at all times.

It is this predictive brain saying, ‘I know this much about the world, so I’m going to set a prediction about how the world should be,’ and that’s what you experience. So the vast majority of the time, you’re living two seconds in the future, because you’ve already predicted what’s about to come.

You’re living in that prediction.

So it’s only when it’s triggered or challenged that your brain switches on?

Learning is essentially, ‘I know you won’t have a prediction for what I’m about to teach you.’

Learning about the brain – you won’t have a prediction unless you’ve gone to school to study it. Trying to instil in you enough knowledge so you can start to form your own predictions is almost the entire game of education. When it comes to marketing, the idea is, ‘OK, you probably already know half, if not 99% of what anyone is going to say.’ It’s about breaking those predictions. If you know someone’s living in the future, how do you do something that causes their predictive brain to switch off and bring them back to the present because their prediction failed? In education, they’re called ‘misconceptions’.

If I asked you the question: ‘what resonance does an MRI have to be set at in order for you to take a picture of blood flow?’ would you want to take a guess?


Exactly, it’s stupid. The answer could be 10, it could be 30 hertz. You wouldn’t know and it wouldn’t matter. Now let’s get into prediction mode. I’ll ask you an easy question. How many senses does a human being have?


Sight, hearing, taste, touch, smell? In reality, human beings have 21 senses.

In this moment, you’ve made a prediction, your brain thinks one thing, and that prediction fails. I can sit here and teach you, use a misconception, and say, ‘Nope, that’s totally wrong. The thing you thought was locked and loaded is incorrect.’ We can find your misconceptions and trigger them. That activates a secondary network in your brain, which forces you to start rewriting code. Your prediction is off, you’re now in the present, rewriting your prediction for the future.

In education that’s easy: I can just quiz you, talk to you. I’ll know where your misconceptions are, so I’ll know where to angle my lesson. In marketing it gets tricky. How do you find people’s misconceptions, so you can break their prediction and force them to pay attention? If you can do that right, they will have to rewrite their prediction to incorporate what you’ve just told them. That’s the entire way the brain works. It says, ‘I was wrong, I must figure out what’s right and put that back in as the new prediction.’

A lot of the focus today is about seamless brand experience, making everything streamlined and easy. Would you say  brands are doing themselves a disservice focusing on this?

Seamlessness is perfect for the few companies that have cornered their market. If you’re Amazon, seamlessness is brilliant, because the more people run their prediction and the less they think, the more you’re their automatic prediction. Amazon has made it so seamless that it is the only one to turn to. It feeds into the prediction.

If you go back, Amazon broke everyone’s prediction. Google was considered disruptive. It was doing things nobody knew. Now, it is the prediction. If you’re a newcomer not already established in the prediction world, you must break that. A lot of people don’t like this concept. They want it to be easy, want people to feel comfortable. That’s great if they don’t want people to know who they are or give a shit about what they do. But if you have something to get out there and you want people to think about you, you need friction.

No friction, no memory.

Today it’s probably even harder to get started because brands like Amazon and Google are so deeply buried in predictive minds.

If you find some market where there is no prediction, that Google and Amazon somehow haven’t touched yet, you can build a prediction from scratch. Otherwise you have figure out what people are thinking, then do something that makes it wrong, so they have to start rewriting. It’s not about side-stepping it. You have to take it full-on. A big shift in thinking, then, is moving from ‘how do we get people’s attention?’ into ‘how do we break people’s predictions?’ There’s a lot of research now. You need your brand to be big, or they’ll pay less attention. You need quick cuts. You need something flashy.

The last few years have seen a lot of brands try to keep messaging subtle, but it seems that is shifting again.

If you think about how to get people’s attention, you think about these specifics: how big does it need to be? What colours should I use? How flashy should it be? A couple of years ago, subtlety was king.

So everyone was thinking ‘how do I be subtle?’ But once subtle becomes the prediction, it stops working, because people predict subtlety. It no longer grabs their attention. In a world of subtlety, the one loud person gets everyone’s attention. If the whole world gets loud, in two years the joke will be that loud won’t work anymore.

We have an attention threshold. You will only be able to pay attention to things that are powerful enough to cross that threshold. If a mouse sneezed, that would be too quiet. It wouldn’t break the threshold. If a gunshot went off, that breaks the threshold. You pay attention. The problem is thresholds always rise. As your predictions get deep, your threshold goes up.

For example: you listen to music when you’re on the train on the way to work. Do you ever notice that the longer you listen the more you start to turn it up? Click by click. You may start at 10, but by the time you get to work you’re at 14. You come back to the train at the end of the day, put in your headphones and say, ‘That’s too loud!’ This is because, as you were listening to it, your prediction was ‘the volume is here, I get it’, so you stop paying attention to it. Your threshold goes up, which means you have to crank more power to pay attention to it.

It’s the same with advertising. If you research what grabs people’s attention, you’re always going to be chasing a phantom, because as soon as you know what grabs their attention, their threshold will adjust to that, they’ll make a new prediction and it won’t work. You’re going to say, ‘Today it’s big, tomorrow it’s super big, the next day it’s tiny, then it’s funny, then it’s sad.’ If your research says your brand has to be big, I’d say, ‘Yeah, today it does, if you want attention. Tomorrow it won’t work, sorry about that.’ So shift to ‘I’m not here to get your attention, I’m here to break your prediction’, then specifics like that don’t really matter. Today could be loud, tomorrow could be small, it could be yellow versus red – who knows what it’s going to be? – but it doesn’t really matter because that’s not what you’re shooting for.

You’re shooting for an understanding of how people understand commercials now. How do people understand brands today? What is their prediction of how brands should work? Cool. How do I do something that leads to that prediction and then breaks it?

But not something completely different because that won’t grab their attention anyway?

You have to lead them down a path. You have to say, ‘How many senses does a human have? Five? Really they have 21!’ Had I just told you that before, you’d go, ‘Oh, humans have a lot more senses’, but because you made the prediction first, it broke. 

Everybody asks: what makes a good commercial? Google did research on this totally accidentally. I wish I could have jumped through the screen and said, ‘No, you missed the point.’ The best commercial in the world is blackness. You’re watching TV, in the groove, in the middle of a commercial break and it’s cool, and all of a sudden your screen goes black. I guarantee you 100 percent of the time you’ll focus on the TV. Because now you’re going, ‘Is my TV broken? Or is it the network? Is this real?’ Your prediction is loud, flashy TV. All of a sudden, that prediction snaps, because now your TV doesn’t do anything. Google was trying to figure out the best commercial. Is it big, is it vertical, is it small? To get a baseline they just put a black screen on for 30 seconds. More people watched that black screen than any of the commercials. They couldn’t figure out why.

The hard part then becomes how to work a brand message or call to action into that.

How do you link it?

What about the Super Bowl commercial with the potato? The slot may have cost millions. There was no  message, just a potato. It went viral, and Cards Against Humanity came out later announcing it was behind the ads. It worked.

So step one is ‘if I can break your prediction, I can have you start to rewrite your code’. Which is what that did. Step two, then, is memories.

Everyone assumes that ‘the more it comes in [to the brain] the bigger that memory’s going to be’. They think if you see a commercial once, you’ll have a small memory, but if you see it 10 times, you’ll have a bigger memory. If that were true, I should totally remember the periodic table. I studied that for three months. I haven’t repeated anything more in my life and I still couldn’t tell you anything about it.

The only way to strengthen memories is to bring the information back out. This is yet another example of the brain working totally opposite to what we think. You can put things in all day, it won’t matter. But as soon as you start bringing it out? Every time you access a memory, it strengthens and makes that memory deeper.

Put a beautiful Clydesdale on pulling a Budweiser truck. Everyone says ‘that was a lovely commercial’ and never talks about it again. They’re not bringing it up, there’s no memory. Put a potato on, people start Tweeting about it, people start talking about it, even to make fun of it. Every time they say something about it? Congratulations, it sticks.

How do you do something that forces people to do the hard work for you? To bring it out themselves. That could be on social media, but it could just be a jingle. Jingles get stuck not because you hear them a million times, but because you sing them a hundred times. Every time you sing it, it gets a little deeper. So how do you get into that without just making an annoying jingle? How do you do something that forces people to recall, to actively bring your thing out?

The third thing you mention after breaking predictions and making memories is personalising. How does this fit in?

In the classroom, this is the hardest thing.

It’s when you take an idea or a concept and you use that to form a cornerstone of your identity. That is as deep as learning is ever going to get.

It’s never going to happen across the board. Say you study maths, science and English. Chances are you’re going to love and personalise one of those. You’re going to say, ‘I am a mathematician.’ Every new equation, ‘I actually take that personally; if I mess it up it’s because something is wrong with me.’ But then you don’t personalise English. You can learn it and pass it. But you may not care or take it personally.

Leading learners down that path is long and arduous and difficult, but at the end of the day, if you get them there, that’s it. Once they tack an idea to their identity, it takes a lot to get it away. Once something gets personalised, we become rabid. It becomes us. It’s no longer ‘that’s a product’; it’s ‘that’s a me’.

So how do you do that?

The only thing I could conceivably believe is – and you’re starting to see it now – linking to causes. My wife will drive 20 miles outside the city to buy a very specific butter because it does something fancy. I don’t get it. To her, her identity is: these health foods, this type of production. She will go out of her way to shop for that stuff.

Trying to break into that is so difficult. ‘All right, you want to be the health food guy? Here’s an all- natural, fat-free organic Coke.’ People will see through that so fast. Everyone’s looking for the silver bullet.

The question I get from teachers is, ‘What do I do?’ If you’re asking, ‘What do I do?’, you’re looking for that immediate answer, and because human beings are an open circuit, it’s just meaningless. It’s never going to work, because as soon as you do it, it’s going to stop working. I tell teachers to stop asking, ‘What do I do?’ and ask, ‘What’s my objective here? What do I want? What is it that I really hope my students get from this?’ Once you have your prime objective, then how you [approach] that changes depending on where they’re at that day.

This is where I try to link the thinking in. Rather than asking, ‘How do I get people to buy my product today?’ just ask, ‘How do people think?’ Once you can solve that, then you just adapt it daily. Go ‘today it’s flash, so I’m going to break that prediction; tomorrow it’s soft, so I’m going to break that prediction’.

* * * * *

To purchase a copy of the latest issue or a subscription to the mag, visit the online shop »

* * * * *

Ben Ice
BY Ben Ice ON 5 March 2018