Getting2Alpha
Getting2Alpha
Julie Dirksen: Design for Behavior Change
Julie Dirksen is an author and learning strategy consultant, renowned for her expertise in creating impactful learning solutions. She wrote the bestselling book Design For How People Learn, and her latest release, Talk to the Elephant: Design Learning for Behavior Change, is gaining significant attention.
Julie’s diverse clientele includes Fortune 500 companies, international NGOs, tech startups, and research initiatives.
Julie blends behavioral science with innovative learning design to craft transformative educational experiences, and her insights are reshaping learning strategies across various sectors.
Intro: [00:00:00] From Silicon Valley, the heart of startup land, it's Getting2Alpha. The show about creating innovative, compelling experiences that people love. And now, here's your host, game designer, entrepreneur, and startup coach, Amy Jo Kim.
Amy: Julie Dirksen is an author and learning strategy consultant. Her best selling book, Design for How People Learn, is a definitive guide to designing great learning experiences for adults that's rooted in behavioral science.
In her latest book, Talk to the Elephant, Julie tackles the issue of learner motivation. It's not enough to teach someone some ideas, you also have to motivate them to put those ideas into practice.
Julie: I kept finding situations where people knew what to do, but they still weren't doing it. That drew me into behavioral science where we look at what else is going to be causing people not to take behaviors that they know are ideally the right thing to do. How do we help people with some of the more [00:01:00] challenging behavior changes?
Amy: Join me as we dive into Julie's thoughts for designing for behavior change and learn why it's so important to talk to the elephant.
Welcome to our session with the amazing Julie Dirksen. I'm so glad you're here.
Julie: I'm so happy to be here too. Yeah.
Amy: Yeah, I've been looking forward to this for months. You just spent time gallivanting around Europe, Julie.
Julie: Yes. Yeah.
Amy: What were you up to there?
Julie: I was in Italy. I was in Turin, Italy at the United Nations Staff College.
The United Nations has added 5 big competency areas to their UN 2.0 initiative, and one of them is behavioral science. And so I was working with some of the internal learning and development folks. They were doing a learning leaders forum because they wanted to look at how to incorporate more behavioral science into a lot of the training initiatives that are happening within the organization.
And that happens to be what I'm talking about these days. So it was a really good fit.
Amy: Yeah, I'm a huge fan [00:02:00] of your work. So you first came on my radar because of your first book, Design for How People Learn. What's the exact title?
Julie: Design for How People Learn. That's exactly right. Thank you.
Amy: Yeah. So that really caught my attention. And then your second book, Talk to the Elephant is exploding. And some of my clients came to me and said, Oh my God, you have to read this book. It's amazing.
And then he kindly sent me a copy and we're going to go deep into that today. And so for everyone listening, Julie is a master at clearly explaining useful frameworks and showing them how to apply them to your work. So that's what we're going to get into today. So you were telling me earlier how you got to know me and my stuff.
Julie: Yeah. Yeah. Well, I think the initial exposure, maybe it was a talk that you did. But in the kind of mid to late 2000S, we had that sort of explosion of conversations around the gamification piece. And there were not that many cogent voices [00:03:00] in that space, if you know what I mean.
There was a lot of stuff that I think hasn't held up well and, some of those kinds of things, but you and like Sebastian Deterding and a few other people. The people were talking about really interesting and smart things about using game mechanics for other purposes besides pure game design and things like that.
And I also found your work on community building. And that's still the book that I recommend anytime the your book on community building for the web. And I know the examples are old in there now and things like that, but I think all the principles are still absolutely rock solid.
And it's absolutely the book I've recommended because I think in learning design, which is where, I spend my time community and social learning and all of those kinds of things are really important, but they don't really have much out there in terms of methods for building, building good learning communities and helping people .
And your material on that is just invaluable. And so those were the things and then I think we bumped into each other a lot on the Twitter and, when it was still Twitter and in some online spaces.
Amy: Totally. I'm curious. Are you [00:04:00] still active on Twitter or not so much?
Julie: Not so much. It seems like most of the professional conversations have moved to LinkedIn for better or worse.
I have a Facebook community with a 7 or 8,000, 7 or 8,000 instructional design people which has been a little bit quiet. In fact, I was thinking actually about Maybe doing something over the summer where we, we do like some book club with your book and look at community building practices in, using that as a sandbox environment to try some things out.
We'll see if I can get that done. I've got a few other things going on, if not, maybe in the fall.
Amy: That's so interesting because once you get a community set up and running, it doesn't necessarily stay the same. It's such a dynamic beast. And so, two years ago, a year ago, you and I were both probably really active on Twitter, right?
Yeah. Not so much Facebook communities that were once super alive. A lot of people aren't spending so much time there.
So, it's tricky to [00:05:00] follow, but I'd love to come back to that a little later in our talks. I think there's some very interesting trends happening online with social and communities.
We have a upcoming guest, Gina Powell. Who runs a community on Facebook of 27,000 women of the what women. And she's seeing just huge shifts in community dynamics. So, let's come back to that when we talk about like what we're both paying attention to heading into the future, but let's take a look into the past for a moment.
I'm so curious how you turned into the person you are. You're an L&D guru. You've written these incredibly well-respected books. You're working with the United Nations. You have a lot of depth and gravitas to what you do. You're also able to so clearly explain things. Where did you get started? How did you develop the skills?
How did your interests lead you to this place?
Julie: Yeah, my own origin story is the origin story of a lot of people who wind up doing some [00:06:00] kind of learning design is you're good at a thing and somebody asks you to teach it to somebody else. And so way, way back in the day I was, I did a certificate in teaching English as a foreign language, which was my first experience with any kind of formal education around educational psychology or learning design or anything like that.
But then I wound up my, my part time college job was for a finance company, and I was doing like loan processing paperwork, and they said, Hey, do you want to train other people how to do this? And I said, okay, it was the early nineties. Job market was a little thin, so I was like, sure, and then found that the pieces of that I really liked and really thought were interesting was creating the learning materials and also working with the technology.
We were just getting started with. We're still calling it web based training and not learning quite yet. And I also got interested, I wound up project managing a front-end system for their customer service group, which really got me interested in this sort of dynamic tension [00:07:00] between, we weren't even calling it UX at this point.
This was the mid-nineties. We were still calling it things like usability engineering and, human computer interaction against learning design, because if you want somebody in a workplace setting to do that, to take the right action, there are two main kind of ways that you can do it.
One is you can create a system that really supports the behaviors that you want them to do, or you can train them how to do it. And so, it's the classic knowledge in the head versus knowledge in the world tension, which is a chapter in Don Norman's design of everyday things. But where are you putting the information or the knowledge so that you can best support performance, which is, I think where this kind of gets really interesting when we look at things like software product design or something like that, is how much are you creating a system that should be usable, for a brand new user, right from the get go versus the system that's actually going to need to be learned.
And there's going to have to be some kind of training or some kind of learning curve to support people developing proficiencies and being able to use like an entire feature set or, things like that.
And so that's been a theme. The [00:08:00] name of my company is Usable Learning. And it was because I was had this sort of dual interest in UX or usability. And also on the learning side I did my master's degree in instructional technology at Indiana, but I spent about half my time over.
And like I said, wasn't even UX yet. It was still human computer interaction. And yeah, so that really got me interested. And then I spent about eight years with a consulting company doing e learning design. Which was really fun and had a great time trying to make, really good, interesting learning experiences in an online environment.
That whole field has been a really challenging one because every time the technology changes the learning experiences get stupid again. We keep getting to the point where we can do cool, interesting things and online learning, and then, All of a sudden we have to support mobile devices and it all goes back to text and pictures or, maybe video if you're lucky.
And that cycle has happened a bunch of different times. But then it turned into a situation where it really felt like so many people come [00:09:00] into you training other people through domain expertise, and there wasn't a good kind of first book to support that audience. Like, if you get started in UX, somebody hands you Steve Krug's don't make me think or something like that.
Or if you're trying to get started with graphic design, they hand you, Robin Williams, non-designers design book and we didn't seem to have one of those for instructional design or learning design. And so that's where design for how people learn came from. Was I really thought we needed a good 1st book to hand people who anybody could read.
It didn't need to be too academic in terms of terminology or language, but it still had a real strong underpinning in evidence-based practice. And what the research said was effective and so that's the. That's been the, that was the impetus for design for how people learn. And I think my only marketing plan for it was hopefully it would be the book that people would recommend to new people and we're somewhere between like 80 and 85 thousand copies worldwide.
So that seems to have worked out way better than I had any right to expect it to. And [00:10:00] then the short version of how I got to the new book was I kept finding situations where people knew what to do, but they still weren't doing it. So they had the knowledge. It wasn't a knowledge transfer problem.
And all my tool set from educational psychology and things like that about helping people learn or remember, weren’t really helping when somebody already has the knowledge, but the behavior is still not happening.
And so that drew me into behavioral science where we look at, well, what else is going to be causing people not to you know take behaviors that they know are are ideally the right thing to do, or continue to do behaviors that they know are the wrong thing to do. And how do we help support people and design for those scenarios where we're trying to help people with some of the more challenging behavior changes.
And so I got very interested in behavioral science. And that led to the new book, which is Talk to the Elephant: Design learning for behavior change.
Amy: Right.
Julie: Jonathan Haidt's metaphor from a book called The Happiness Hypothesis that came out in, I think, the mid [00:11:00] 2000s. And he talks about how your brain is there's a lot of these sort of dual decision-making Models. There's Daniel Kahneman's system two and, some things like that.
I like this one because as soon as I explain it to people, they, it absolutely grabs them, which is this idea that like, you think about your brain, like down by your spinal cord or some of the oldest parts of your brain. So, breathing and heart rate and reflexes, these really kind of primitive functions.
Then you get things like vision, hearing, gross motor control, fine motor control right in the middle of your brain, you get the limbic system, which has like the amygdala and the hypothalamus and emotional reactions and, fight or flight and different kinds of things. And so all of these areas, a big amount of your sort of brain real estate Activates when you're dealing with, like, sensing the world perceiving or understanding what's what inputs are coming in.
So your physical environment, your physical reality moving around in the world, controlling your body, all of these kinds of things. So this is all like what he refers to as the [00:12:00] elephant, which is your automatic emotional visceral brain. What's your experience moving around in the world or perceiving the world or interacting with it?
And what do you, what are your feelings about it? And then right up in front, we have areas like the prefrontal cortex, which tend to activate when we're doing things like executive function and reasoning and impulse control and projecting out into the future to consider consequences.
But that used to have this brain function that's your conscious verbal thinking brain, right? And that both of these entities participate in decision making. So when you're lying in bed on a cold morning and your alarm goes off, you're like, your rider is like, you should get up. You should get up. You should make sure that you get coffee and a decent breakfast before you have to like prep for the webinar. And, you should like not.
And then your elephant's like, but it's warm and comfortable and nice here. And so who who wins the battle for the snooze button?
Does your logical, Empirical, project out into the future self win? Or does your immediate like, this is nice. And I think I'll stay here a little longer self win [00:13:00] for some of those decision makings some of that kind of decision making.
And so, yeah, that whenever we're finding these challenging decisions you have both of these entities, or both of these systems that are sort of participating in the decision making, and we have a real tendency to just talk exclusively to the right, or we make the rational argument or show the data. We, make the intellectual case, or we talk about vague, hard to imagine, benefits to doing this behavior that may or may not happen in the future. Sometime we don't know when and your elephant's like, yeah, but right now, what do I care about?
Or what does my environment tell me, or what does my physical reality tell me is important. And so the new book, the title is Talk to the Elephant, which is if you're going to engage with people on these difficult behavior changes, which quite frankly, almost always involve waiting for some future benefit or consequence. And, your elephant has this present bias. It cares about what's [00:14:00] happening right now. And if you're talking about something abstract in future, you're always going to struggle to convince people of the utility of it because everybody's got 37 other things they can be worrying about every minute of every day.
And if you're number 38, you're just not good. Like, it's just not going to come up. They're not going to get that far down on the list. And so how do you push it up? The level of importance where the thing you're trying to communicate to your users or the thing you're trying to talk about, starts to really resonate and feel important to them.
Well, if you're not, that's where we really need to talk to the elephant, because how you feel about something is one of the main gauges we use for is something important.
Amy: I have so many questions but thank you for that explanation. Let's talk about how this plays out with real examples because I always find that useful.
So first of all, let's say we're building products, creating startups, designing games, that sort of thing. That's a nice insight. What [00:15:00] does that actually mean in practice to do A versus B, if I'm a creator?
Are there any products in the marketplace you look at and you say, yeah, that's good design?
Julie: Right. Even if it's something really mundane, like your, product for invoicing people or something like that.
When we're looking at this, one of the big classic problems with a lot of product development is the feature creep piece, right? We keep throwing more features in there because it gives us something to market and gives us something to talk about, but it's not necessarily answering immediate questions or immediate problems that people have.
So like one of the fundamental questions that I always have is, what's the immediate need that people have? And that could be in software products, but it's also how we talk to people about training programs, which is what I do a lot, right?
Amy: Right.
Julie: So one of the formulas that I kind of use when I'm thinking about how I'm communicating value is how big is it?
And that's usually where we focus. Like you can do amazing things with this product. Here's how it's going to revolutionize your invoicing and make life way easier and [00:16:00] whatever. But there's also how immediate is the benefit? I'm pretty obsessed with hyperbolic or temporal discounting which is this phenomenon that happens when a reward gets pushed out into the future. So, can we do a quick experience experiment with your audience just to see?
Amy: Sure.
Julie: Okay. So if I was handing out money right now, I could hand out either $10 or $11. So tell me whether you'd rather have the $10 or the $11.
And again, totally not a trick question, completely legitimate.
I got some votes for 11 dollars, which makes sense. Right? It's more money. Sure. Why not? Okay. So the next version of this question is, would you rather I'm giving out money, but I'll give you 10 dollars today or 11 dollars tomorrow.
Now, what do people want?
Okay. Now I've got to vote for 10 dollars, 11. Okay still some 11. Okay. So you guys are forward thinking. You're willing to wait.Okay. Next question. $10 today or $11 in a year.
What do people want?
Okay, pretty [00:17:00] conclusively, right? So it wasn't the size of the reward that changed. It was when I was getting, when are you going to get it, right? So if they can get it immediately, of course, everybody wants the extra dollar. If they have to wait a day and it starts to go a little bit.
And I've done this with lots of groups and it's usually 60-40, 50-50 split. And then if you have to wait a year, everybody's like, no, I'll take the $10 today. Every once in a while, I get one person who wants the $11 a year. And then they want to talk to me about interest rates. But the challenge there is whenever we're talking about things, we can't just ask how big is the reward in this case, it was an extra dollar, but when do they get it?
And so a lot of times we're talking about rewards that are delayed and the more that people delay a reward, the more it gets discounted. And this happens in my world with learning design quite a lot where I'm trying to teach people things they just don't need right away. If I'm teaching you a skill, you can use immediately.
It's pretty easy to allocate your attention. If you were watching a video to fix the leaky faucet that is currently leaking [00:18:00] that you immediately want to fix, you don't usually have trouble paying attention to that video. But if I'm teaching you, a tax principle, you're not going to use for six months.
It's much, much more difficult because your elephant is like, I don't understand why we're here. This doesn't seem like something I can use. Why do I care?
And so when we're creating learning experiences, the way that we deal with that is we either put the information make it available. So when they do actually have this tax problem in 6 months, they can just grab it. And it's easily available.
Or we try to create an interesting scenario or problem to solve that's immediate. And. Then, there's the elephants like, Oh okay, we're playing a game. I can do that. I'm trying to solve a problem. Sure. Now I can allocate my attention for this because this is really a consistent problem that we're seeing, which is, I need you to focus and pay attention and allocate, pay attention to my list of features and benefits or pay attention to this training that I'm trying to do about this product or things like that right now, but you can't use a lot of this stuff for a long time.
And so [00:19:00] when we have that scenario, we're putting all the weight on the rider. The rider is the only one who can project that into the future. And the rider is the only one who's going to care about something taxes that are happening in 6 months. And so the elephant is like, I think we should go look at Instagram because that seems way more interesting to me.
And so we have this tension between these things. As long as we're leaning so hard on the rider and not engaging the elephant at all, we're just making everything more difficult for the people in our audiences that we're trying to support.
Amy: So I feel like this connects to your comments earlier about gamification because part of what gamification was trying to do, and this still comes up, I'm still in meetings where people are like, Oh, if we gave them points for doing this thing, like all our problems would be solved, but it's just an idea and it seems like it's the right idea because it's an immediate reward instead of this for something that's [00:20:00] harder to explain or understand, maybe it's a boring task, whatever.
So why is it that adding those trinket like or, game like rewards doesn't solve this problem? Why doesn't it work?
Julie: Yeah. I mean, so often it's not about the intrinsic thing that that the person cares about, right?
If they can see intrinsic value that immediately helps them with something, you get attention all day long.
There's this myth that our attention spans are shorter than a goldfish. This is my favorite myth to debunk. And I think I have a video somewhere where I explain where this all came from and it was complete nonsense. I'd also like to know how somebody's measuring goldfish attention span.
I'd love to know what that actually looks like. But our attention spans, I mean, we've all had the experience of binging a Netflix show, or the example I always like to give is there's a movie theater near my house that runs All three Lord of the Rings movies back to back every Christmas. And people love it and it's 11 hours. So, we can pay attention for 11 hours if it's something we like and we're happy to engage with and we're motivated by.
So there [00:21:00] doesn't seem to be any actual like limit to people's attention spans. What there seems to be more likely is there's a limit to how long I can force myself to pay attention. Because that act takes a lot of effort and, eventually I'm going to get tired and need a cookie or a break or a nap or something like that.
And so anytime we're asking the rider to basically drag the elephant along for the ride, then maybe we are dealing with a limit of a few minutes or 10 minutes or whatever ideas we've had or heard around attention span limits. But if people are fully engaged, they can pay attention indefinitely, Game design absolutely looks at how to do this.
Amy: Yeah. So much of what you're talking about is good game design, just in time learning versus like, here's all the functionality in the app. Isn't that great? Just dripping it out, figuring out how to set up reward schedules that keep people engaged, how to introduce longer term challenges when you're ready for them, which is also good learning design.
Julie: Right. And that structured flow of goals where you get some easy wins and then you can move on and [00:22:00] do harder things once you're engaged with that. We know all of that is useful for learning design as well. And when we think about product stuff, like you want to be able to like open up a product and have some easy wins right away.
Right? Like you want to be able to do a couple of things that make your life easier immediately. And then if you're like, okay, I can see how this is helpful. Then maybe you're willing to invest a little bit more effort into learning some of the more obscure functions or setting it up in a way that you can actually, do some more complicated things with it.
And the more that people have that relationship with the product, the more product loyalty you're going to get, because they have put this effort into figuring out how to optimize it for the world and they've incorporated into their internal systems.
And so, that sort of structured flow of goals can be a really useful way to think about. What experience do I want people to? Do people want to have as they start to engage with my product? And what are they going to be able to do immediately versus what are they going to have to learn to do?
And there's always this trade-off between a more powerful tool [00:23:00] versus an easier to use tool. And Photoshop, incredibly powerful tool, but it is not easy to learn how to use. I've used Photoshop for decades at this point and I still can only use a tiny fraction of its functionality.
Fortunately, that's all I need for what I'm doing with it. But a much easier to use graphics tool is going to be less powerful. And so that's part of the trade off that we're dealing with. And what's the journey that as people start to come in and work with the things that we're building, where are they having early wins? Where are they having more satisfying accomplishments that they have to work on? All of those kinds of things become part of that experience.
Amy: Awesome. Yeah. And not easy to do, but that's good design. I also want to connect this back to your change ladder, but first I'm really interested in getting your overview of the key models that you're bringing up in your book because models are so helpful for looking at the world. They're like lenses.
And here's one. The [00:24:00] company model. So tell us about what is this and where did it come from? And how do you use it in your work, and in your own creating of models?
Julie: Yeah, the company model is part of a bigger model called the behavior change wheel, which was developed by Susan Mickey and at university college London center for behavior change.
And her team Robert West, the Atkins, and some other people worked on this. And combi is the idea that in order for behavior to happen, there need to be three kinds of crucial components. And then right in the middle of that piece of the company.
So is the person capable of making the chain of doing the behavior? And it could be their physical capability. So if it's using a physical tool and things like that, are they capable of doing the action? But then there's also psychological capability, which means do they have the knowledge?
Do they have the skills, which is anything they have to practice in order to develop some proficiency we'd call a skill area. Do they have the psychological stamina for certain things. For example, if you create a habit tracking app, that's the software products, like there's a certain [00:25:00] amount of stamina that's involved with just continually updating that day after day, and staying on top of it.
And so there's a stamina issue there as well. When we look at psychological capability. Opportunity looks at two different variables. It looks at physical opportunity, which is how well does the environment support the behavior. And that's all of the stuff that like UX people are frequently concerned with is how well does this system or how well does this environment support the things that we're trying to do.
And sometimes fixing the environment is way better and way more effective than trying to teach people. One of the examples I use a lot in the book is hand washing in healthcare because it's such a well-researched area. And the big jump in handwashing compliance in US healthcare came because of a lot of the physical changes they made to the environment.
So, making the handwashing stations easier to access and putting them right when the people walk into the room. And making the addition of the alcohol-based hand rubs was a big leap in handwashing compliance, because it was just easier and faster in a lot of environments. [00:26:00] So how do I make the physical environment, whether it's the software system, or literally the spaces around you work better for people and around the behavior?
Then there's the social environment. Do I see the behavior being modeled by other people? Does my social network support the behavior? Do I have accountability partners or people who are also engaged in this. And that's where we look at things like those community built, the communities and things like that around supporting the behavior. And then the last one's motivation and they break that into two categories as well.
One is reflective motivation, which is motivation I can reflect on or talk about. And so what are my goals? Do I have explicit goals around this? Do I see this as a competency? I want to develop. Is this a behavior I really want to do? How does this relate to things like values and identity?
And then automatic motivation is all the elephant stuff. It's habits and it's biases and it's automatic reactions and it's thoughts and feelings. And the stamina shows up there too, in terms of, like the impulse control when I'm tired or frustrated.
And so we look at all 3 of these. Are they capable of [00:27:00] doing it? Does the physical and social environment support it in terms of opportunity? And do they have motivation to do it? And then, depending on where the gaps are, we would need to adjust our strategy. So, let's say I'm doing a leadership curriculum, and I want people to be more comfortable giving feedback both ways between employees and managers.
Well, in some cases it might be people genuinely don't know good ways to give feedback. And so then that's a capability issue. And I do a training, but sometimes it's the opportunity, like there's social norms in place that are causing this to be a really fraught behavior. And I can train people all day long, but unless we're doing something to address the social environment, we're not going to change the behavior.
Amy: That model sounds close to identical to BJ Fogg's model.
Julie: Yeah, there's a lot of similarities. So when Combi was created in the behavior change wheel, they did a survey across, I think, 19 different behavioral models to try [00:28:00] to distill into common elements and BJ Fogg's, the BMAT and things like that. His focus has been a little bit more on specifically habit formation. And a lot of his stuff has had really interesting things around what's an opportune timing for trying to impact the behavior.
So it's similar, but it's got some sort of subtle differences as well. I think both have value. There's a lot of different ways that you can approach models. I like Combi because I feel like it really is the most comprehensive. And if you follow it all the way through, they look at different domains where you can intervene.
And then they have what's referred to as the behavior change technique taxonomy, which is like 93 different behavior change interventions that they distilled out of the research literature. And so you've got a really wide swath of solutions that you can pick and choose from.
And honestly, most of these behavior change challenges are complex. And so you're going to array a sort of whole kind of spectrum of options that you're going to look at rather than just a single solution. It's going to fix everything.
Amy: So that's great. Thank you so much for [00:29:00] explaining that model.
So are there other key principles or models in your book that are an important part of the cornerstones?
Julie: Yeah, I mean, one of the models that I know that we have talked about a little bit is based on some research that was also done out of Robert West and some other people where they were looking at risk assessment ladders and the idea, can you meet people where they are on the journey?
And I took some of the stuff that they did and created a slightly more general version of it. Because they were specific to things like texting while driving or smoking cessation or alcohol cessation . So their research was around using these. But just this kind of question of where are people in the process and how do you meet them where they are.
There's a bigger picture model. It's the kind of stage model that sort of says, people are in pre- planning, then they get to the point where they're executing and then maintaining. And people need different things, depending on where they are in the process.
And so that's one of the things that I use quite a bit, which is trying to [00:30:00] identify, where do we think our audience is? Is it that they genuinely don't know about the behavior at all? In which case education is perfect intervention? Or is it more of like, they know, but they aren't totally on board or not completely convinced, in which case we're dealing with persuasion.
Or some kind of something to address motivation. Is it that they are pretty much ready, but now we just need that little nudge or that little assistance or a little bit of coaching or a little bit of support in which case we're starting to look at things like, how do I create an environment or some support tools that can help you step over the threshold and get started with things.
Or how do I make sure that you're getting enough practice and feedback that you're able to perform at the level that you need to do?
Amy: Yeah, I feel that this is the moment 'cause this is talking about.
Julie: Right?
Amy: So by the way, for everyone listening this is something that you can get.
This is a worksheet that Julie's made available. So talk us through this. What is this? How do you use it?
Julie: Yeah. So is it that they don't know about the [00:31:00] behavior? It's education at that point really, or how do you help people?
Is it that they know about it, but they don't really get it or they aren't convinced or they have other priorities. And, number four is a tough one because competing priorities are always going to be at play. We are all busy people. And everybody's ready to start that new exercise program, but maybe just not today.
So even if somebody is convinced of the utility of the behavior, and they're convinced that they want to do it, it can still be challenging to prioritize it above other things. Is it about that I'm like, okay, I'm ready to get started, but I want to exercise and do strength building because I know that's going to make life better as I get older.
But I don't really know how to get started or I need some help, or I need some support. I'm not confident or I think it's too hard. And then we, moved down into 8, 9, 10 are things around like habit support.
I've started doing it. I've done it for 2 whole weeks. Now that's great. But you know, I just skipped a day. What happens then? And how do we create some kind [00:32:00] of feedback system, which is where BJ Fogg has done actually quite a lot of work in that. How do we create self reinforcing habit systems?
And so as we move through this, the support or the resources that somebody needs changes pretty dramatically. It might be appeals to identity or it might be talking about values in one place and then another place. I just really need somebody to take my hand and kind of walk me down the first few steps of the path.
Or it might be that I really need a feedback system or I really need an accountability partner. And so, one of the things we can look at, is there a way to meet people where they are? So you do some kind of assessment to figure out where they are to have a little survey tool in there to, that can be modified for different purposes that people can self identify where they are in the process.
And the answer is people can be more than one place. It's not a perfect, like, you always go from, 1 to 10 every single time. Sometimes it's a combination of different ones. Or is it about creating an environment with lots of resources so that wherever people are, they can find the thing that [00:33:00] they need or that can be helpful to them.
So this becomes looking at what's the whole ecosystem around somebody to support the behaviors that we're trying to help people with.
Amy: So this is very central for what you do, which is designing learning programs, but everything you're talking about, especially that last list, Sounds incredibly relevant for designing any kind of medical or healthcare interventions.
Does that come up a lot in your work?
Julie: Yeah, I've done quite a bit with healthcare over the years. Like I said, the hand washing stuff was one because I've worked on some of those curriculums. Done a lot on training for vaccination workers or, medical technology.
I used to way back in the day, know a lot about IV infusion errors. So all the places where people would make mistakes setting up an IV for a patient. If you skip a decimal point, it can be really bad. And so one of the things we were looking at for a lot of that stuff is how much of the guardrails are the things that you want to do to sort of change the behavior we try to build into the [00:34:00] system?
But the more that we build the systems to really control the behaviors, the less flexible those systems frequently become. And so, we always want to have that support of both this.
Amy: Yeah.
Julie: The human judgment, because they're the ones right there doing the thing and the system support.
So if we try to lean entirely on the system to control all the behaviors, we've all been in those, right? We've all been in the horrible phone tree, trying to get support where you know, the thing you need, you genuinely know, and it just doesn't want to give you that option because they're convinced that you should be going down this path.
And you're like no, it's not right. This is not what I need. I know what I need. And this isn't it. And so that's the danger of relying too heavily on, we'll just build the system so that people can't make mistakes. Well, people are complicated and the world is complicated.
And so, that inherent flexibility that you want to have in there means that you need to have humans who can learn and adapt in your systems as well.
Amy: Yep. So, [00:35:00] looking toward the future, Julie, I'm interested in what trends you're paying attention to.
What you see that's really shaping your world. And in particular, I'm interested on how you're approaching all the developments in AI in terms of your field, your work. It comes up in every conversation I have, people have pretty radically different approaches.
And what are you paying attention to looking toward the future?
Julie: The most exciting aspect of the AI piece to me is the idea that we'll be able to give people feedback at scale in learning environments. So that's the part I'm most excited about because historically, we have had an over reliance on multiple choice questions because they're easy to grade.
That's honestly the only reason why every online learning course that maybe anybody's ever taken has a multiple choice test in there somewhere. And they're not great. They're a limited tool and doing them well is actually really difficult, like writing good questions and stuff. And so the idea that people could generate [00:36:00] their own answers and get good feedback from an AI about whether or not they're answering a question well, means that we potentially, because the thing that has scaled easily in learning is content.
It doesn't matter very much whether 1000 people watch a video or 100,000 people watch a video. There's lots of infrastructure to support that, but giving meaningful feedback to 10 people versus 100,000 people has not scaled particularly well at all.
And so that's the part that I, interested in, I'm excited about. The part I'm more skeptical about, especially with the LLMs, the large language models, is trying to support expertise. So the main way that a lot of those are getting used, I think, well, now is somebody who does know how to do the job is using it to do some of their functions faster. So that person has the knowledge and the judgment to evaluate the outputs from a chat GPT or something like that and say, oh, this is right. But I need to fix this right. Now, they still need the stamina [00:37:00] discipline to actually check the outputs. And we know that's not happening because we've seen academic papers where people have accidentally copied the chat GPT prompt into the paper content itself.
Which, if anybody should be checking the outputs before they publish it, it should be academic papers. And that's a whole can of worms all on its own. So we know that people are not always checking the outputs before they hit copy and paste. And that's going to be a persistent issue because we also know that sort of by definition, these outputs are not reliable.
They're going to get better. They're already better, but it depends on if you think large language model hallucinations are a feature or a bug. And I think they might be a feature, not a bug, in which case they're never going away completely. And so we need to know who's validating these outputs.
And so if we wanna use the chat bots or the things like that as a way to support expertise. So an employee comes in and I can have them answering mortgage questions in a call center to go back to my first example, mortgage questions in a call center right outta the gate because they're [00:38:00] using a chat bot.
Well, that's great, except that this person doesn't have the expertise to evaluate what the chatbots telling him. And so then we need to figure out where is that governance coming from? Where is that validation coming from? So, if it's an expert using it, then they have the knowledge. They need the discipline, but they have the knowledge.
But if it's a new person trying to operate at a higher level without being trained, then how is that? Who's judging? How are we validating the outputs to make sure that, the right accurate information is still going out in the world.
Amy: Interesting. Any other than AI, anything else you're following?
Julie: The big one and actually the way that the AI plays into this is, we've known about the promise of the virtual reality environments for so long and it really just hasn't managed to get dragged across the line to being used. And I do think with the VR stuff, there's a couple issues.
One is like having safe environments for people to use them. But another issue has been the cost of authoring in [00:39:00] virtual reality. And I do think the AI, content creation stuff may be a big turning point in the next few years in terms of authoring in virtual reality environments, because we'll be able to, you'll have people who don't necessarily have to be programmers who will be able to author, fully realized virtual reality environments or experiences.
And so that's a big one because then you can actually have a lot of the physical cues that talk to the elephant in your learning design, as opposed to, abstract like reading text or, hearing lectures or things like that. And we know that has an impact on me.
Amy: What are you imagining there?
Julie: Well, one of my favorite researchers research groups is the virtual human interaction laboratory at Stanford. Jeremy balance and one of the principal people there. And they've done a lot of looking at how much does having the visual experience have an impact on behavior. And so one of my favorite ones that they've done is they wanted to look at would, knowledge or [00:40:00] experience have a bigger impact on using disposable paper, right? So the environmental impact of using disposable paper. So they had half the group come in and read about, like, trees being cut down to, support using more paper towels and whatever and then the other half went into a visceral environment where they were literally cutting down virtual trees with a virtual chainsaw.
There were sounds and I think they even had the floor wired to vibrate a little when the tree fell down and all this kind of stuff. So this visceral experience of the impact of it, and they brought them both at both groups out and they said, okay, well, you use less paper in the future.
And both groups said, yes, of course we will. It's Northern California. They're an environmental group. Sure, they would. And again, it's the writer who sort of projects out to future behavior. But then they would accidentally quote unquote, spill a glass of water and look at how many paper towels people use to mop up the spill, which is my favorite part of this experiment.
And the people who had been in the more visceral condition used almost 20 percent less paper than the people who had been in just the intellectual [00:41:00] condition. And so, there seems to be something I mean, I don't want to over extrapolate from a single study, but there seems to be something about having the experience.
We look at this in safety training all the time, right? People who know somebody who's been injured are going to be careful, as opposed to telling people, you should wear your eye protection or do lockout tag out for electrical or whatever. If you have personally had the experience, or someone who's had the experience, that is a very different level of adherence to the safety guidelines.
And somebody who's just had it explained to them with the PowerPoint slides and, somebody who tells them admonishes them that it's very important.
Amy: Yeah. That's so interesting.
This was so fantastic having you here, Julie, and getting to know you a little bit better. I just absolutely loved it.
Julie: Yay! Thank you so much for having me. And I'm still a little bit of an Amy Jo Kim fan girl, so this is very exciting for me too.
Amy: So thanks everybody. And Julie special thank you. Good luck with your move.
Julie: Yes. Thank you.
Amy: Thanks again, Julie. Have a [00:42:00] great day, everyone. Bye.
Outro: Thanks for listening to Getting2Alpha with Amy Jo Kim, the shows that help you innovate faster and smarter. Be sure to check out our website, getting2alpha.com. That's getting[number]2alpha.com for more great resources and podcast episodes.