Host: Welcome to CET Talks, the International Accreditors for Continuing Education and Training’s podcast, where we convene thought leaders in the continuing education and training ecosystem to share ideas, research, best practices, and experiences that promote the creation of a world that learns better. Enjoy the episode.  

Randy Bowman: Hello, and welcome to CET Talks. My name is Randy Bowman, IACET’s president and CEO. I’m excited to be here with our learning track cohost, Meghan.  

Meghan Conan: Hi Randy. Happy to be back with you again for another great podcast. My name is Megan Conan. I am the program manager for ASTM International currently, and I have had a very longstanding relationship with IACET. I am thrilled to have been working with them for about the last 10 years, doing podcasts, working on some initiatives regarding distance learning, and my personal favorite, which is the Open Digital Credentials program, for which I’m fortunate enough to be able to be the instructor for that course, as well as whatever else they might need support in. So, happy to be here today with you, Randy, and with you all.  

Randy Bowman: Thanks, Meghan. I’ve got to ask; maybe you can help me diagnose this. I can’t tell you how many times in my life I’ve shown someone a much faster, easier way to do something on their computer. They nod, they smile, they say, “Wow, that’s great,” and then I watch them go back and do the exact same thing they’ve always been doing that’s causing all of their problems every single time. Meghan, what am I doing wrong here?  

Meghan Conan: Oh, Randy, I don’t think you’re doing anything wrong. I think something we all struggle with is the way we’ve always done it and getting into habits and routines, and it becomes harder, especially as adults, for us to change and move forward into new things and new ways of doing things.  

Randy Bowman: Well, I’m right there with you, and that’s why I’m so excited to have with us today, Julie Dirksen, and she’s going to talk to us about how we can design learning for behavior change.  

Meghan Conan: Yes, we’re so excited to have Julie with us today. Julie is a learning strategy consultant and an author. She’s written Design for How People Learn and Talk to the Elephant: Design Learning for Behavior Change, which I am super excited about, since I have read it myself. She specializes in applying behavioral science to learning interventions with an MS in instructional systems technology from Indiana University. Julie’s also a former adjunct faculty member from the Minneapolis College of Art and Design and has been honored as a Learning Guild Guild Master, which is a fantastic achievement. So, welcome, Julie. We’re very happy to have you with us today.  

Julie Dirksen: Thank you for having me.  

Randy Bowman: Well, Julie, you heard my predicament. How am I supposed to help these people? I know in your books, and in your articles, you often explore the idea that knowing doesn’t always lead to doing. Can you explain a little bit why so many learning interventions fail to change behavior, even when learners understand the content?  

Julie Dirksen: I described the core problem, which I think is what you’re talking about, is they know what to do, but they’re still not doing it. We have that all the time. We have it in safety training, we have it in our own lives. I personally have that relationship with sunscreen many days of the week, and I’m working on it; I’m trying to be better. As with so many things, you can have all the intellectual knowledge in the world and still not be motivated to do the behavior. I totally understand; I really do have good intentions about sunscreen, but sometimes it just doesn’t happen the way that it’s supposed to. As you described, habits are a huge piece of it because habits are these incredibly beneficial things. They create cognitively easy actions for us throughout the day.  

Basically, once something becomes a habit, the cognitive action that needs to execute that habit is much lower than, say, the first time you did a particular behavior. The big benefit of it is that you’ve got all this stuff that can run without a lot of conscious attention, but that’s also the problem, right? It runs without a lot of conscious attention and interrupting and changing it for something else is actually a pretty effortful thing to do, to change an existing habit. So, the example you’re describing, of people having these ways of doing things on their computer, and you can show them a much better way, but that still feels harder than the convoluted way that they’ve always done it. There’s a lot of interesting research that looks at glucose metabolic rate and how much brain fuel is being run to do something. It’s unquestionably the highest  during the first time you do it, when you’re really trying to wrap your head around it and understand it. So, that level of effort sometimes discourages us to make changes, even though the change is going to ultimately be a really beneficial one.  

Meghan Conan: That’s a fantastic response to that. You’ve got me spinning on all sorts of thoughts, and as I mentioned, I was fortunate enough to read your book, Talk to the Elephant, which I know is a nod to the idea of dual processing in the brain. But for our listeners who might be unfamiliar with that metaphor, what does the elephant represent in learning and behavior change?  

Julie Dirksen: Daniel Kahneman’s ‘thinking fast and slow,’ for example, has this dual process idea. He talks about system one and system two. I like the rider and the elephant, which comes from the psychologist, Jonathan Haidt, because I think it’s just easier for people to wrap their heads around, and it’s nicely visual, and it sticks with people. Basically, if you think about your brain, a big swath of your brain is concerned with things like all your automatic functions, like breathing and heart rate. It’s also concerned with perceiving the world, like seeing, hearing, and understanding what the immediate environment around you is like. It’s also concerned with moving your body around in the environment. Everything from gross motor control, just being able to walk around, all the way to fine motor control, being able to tap something in on your phone or something like that.  

Then we also have things like the limbic system, which are your amygdala and your hypothalamus, which concern feeling and emotional reactions and things like fight or flight. So, a big portion of your brain real estate is dealing with your automatic visceral feeling brain. It’s the part of your brain that reacts instantaneously if there’s a perceived threat or even something that’s odd in your environment. It’s the part of your brain that, again, runs without a lot of conscious attention because it’s got its automatic routines. It’s the emotional deep reluctance when you know you should turn off the next episode of the TV show and actually go to bed, yet the inertia’s got you trapped on the couch. Just one more show. And so in Haidt’s metaphor, you refer to that part of your brain as the elephant, right?  

It’s this large, feeling part of your brain that deals with things that are happening to you right now, and also your physical sensing, perceiving, and moving around in the world. And then we have these nice areas of your brain like your prefrontal cortex, which is right behind your eyes, and that activates when you’re doing things like executive function, impulse control, logical reasoning, or planning ahead for the future. So, it’s all these really super important cognitive functions that help us do things like save for retirement, decide to wear sunscreen even though it’s not that sunny out, make good choices about how you’re going to allocate your work time to the first thing in the morning, or any of those kinds of things. But we have both entities that are participating in this decision-making.  

So when your morning alarm goes off, and you’re thinking, “I should get up, I should get up, I should totally get up,” your elephant’s also like, “But it’s comfortable and nice and warm, and it’s still dark out, and why don’t we stay here a little longer?” So, depending on who wins, your rider or the elephant is all about whether the snooze button gets pushed on an alarm or not. Both of these entities are a valid part of our decision-making. The elephant isn’t lazy or stupid or any of those kinds of things. It serves a lot of very important functions. It helps keep us safe. It keeps us from expending a lot of effort that maybe isn’t useful. It guides a lot of our day-to-day reality. We wouldn’t be the humans we are without it. But at the same time, when we’re creating a lot of learning experiences, we tend to focus entirely on the message to the rider.  

What’s the logical argument for doing something? Or what’s the logical argument for changing the way that you do something on your computer and not recognizing that both entities play into decision-making? So, if you can figure out how a learning experience takes into account both the elephant and the rider, we tend to see improvement in things like a behavior that lasts a little bit longer. Like, okay, they tried it once in class, but now they went back to the workplace, and now they’re not doing it anymore. If we can really engage the elephant, that makes it much more likely that people will persist with a new behavior or a new habit or something that they’re changing about the way that they do their.  

Randy Bowman: That’s amazing insight and great reminders about how our brains actually work. Maybe it’ll help me be a little more patient with my users when they continue to go and do it the hard way. So many learning and development teams, they struggle to measure behavior change in any kind of meaningful way. What’s your advice for identifying indicators or metrics that actually matter?  

Julie Dirksen: Kathy Moore, who did a tremendous amount of work on learning analysis and things like scenario-based learning, she always had a thing where she would ask stakeholders upfront, “What is one metric that you’re already measuring that will change if this training material is successful?” I think that’s a really great upfront question because one of the things that it can do is indicate that maybe you’re not measuring anything that’s going to be significant in this, and maybe that’s a conversation that we need to have. The other aspect of measurement I think is important is people get really daunted by the idea of measuring actual behavior change at scale. So, if you’ve got an audience of 3000 people, the idea that you’re going to be able to measure 3000 people’s behavior is an expensive proposition. That’s a lot of coordination, effort, and a lot of expense.  

I think one of the things we can also look at is when we look at academic research, a lot of times they will look at a reasonable sample size for behavioral measure to get some feedback on the efficacy of their programs. Another thing I encourage people to do is, well, can you follow a cohort of 30 around and see if you can do maybe some AB testing or something that indicates what’s working and what’s not working in terms of the outcomes of the learning. Things like the Brinkerhoff Success Case model are another nice one that doesn’t involve quite as heavy a lift as actually collecting behavioral data from an entire audience. I think having some flexibility about how we’re thinking about it, because one of the questions whenever we’re trying to measure these things is, are you looking for proof or are you looking for data that you can feed back in and improve things?  

If you’re looking for proof, you may need to know the stats with the entire 3000 people in my audience. But if the crucial thing that you’re looking for is about what’s working and what’s not working, that I can feed into iterations or help identify what future training needs might look like, you can actually get good data. It’s not a hundred percent conclusive, but it can certainly help inform people’s efforts as they go forward and future iterations of material. One of the other nice things is that some of the AI tools are really good at analyzing natural language inputs. I’m kind of frequently frustrated when I get all these emails with just one 10-point scale of how likely you are to recommend. And I mean, they aren’t the worst. I suppose it indicates something, but it so rarely has anything to do with real feedback that I might want to give on the experience because it’s such a limited question. But open text inputs, hearing from people, or doing interviews have been typically quite costly. I think one of the nice things about some of the AI tools is it brings down the cost of doing analysis on some of those sorts of things. That can make it more feasible, as well.  

Randy Bowman: I completely agree with you on the net promoter score. I can go off on that for days.  

Julie Dirksen: Yeah, exactly.  

Meghan Conan: I’m right there with both of you on that one. You had talked earlier, Julie, about habit and making that change to the behavior by doing it and being conscious of doing it in the beginning. I’m curious; what role do you think the external environments, emotional cues, or people’s habits, in general, play when someone follows through on a learned behavior?  

Julie Dirksen: Well, we know that the environment makes a huge difference. And in fact, in many cases, that’s the best investment that you can do around behavior change. In a lot of cases, if we make the behavior much easier to do, we’re just going to get better results. Some of the persuasive messaging or trying to convince people or any of those kinds of things, we’re going to see some more variability there. But honestly, if the behavior is just much simpler and much easier to do, that will almost always get us some beneficial results. I do think whenever we’re looking at a behavior change, I think one of the crucial questions to ask is what can we do to make the behavior that we want people to do as easy as possible? The other thing that shows up constantly, I’ve literally never seen a behavior change challenge that didn’t have an element of delayed or absent feedback.  

So typically, if we think about some of these behaviors that we really want people to do—I use handwashing as an example quite a lot—if you don’t wash your hands adequately in a healthcare or food service setting, bad things happen. People get sick. But the likelihood that you’re going to know immediately that I didn’t wash my hands and so, therefore, I’m getting people sick, is basically impossible. Even if you’re in a healthcare setting, the bad thing might be some patients on the ward get sick, but you’re almost never going to be able to draw a dotted line between, I didn’t wash my hands at three o’clock on Thursday and therefore these patients got sick; our healthcare environments don’t work that way anymore. We don’t have that kind of individual care that might help you draw that conclusion. So, when we look at those kinds of things, those tend to be the most challenging behaviors, where there’s sort of no visible downside to it, or people don’t have enough experience of the upside.  

So in your computer example of, “Hey, I just showed you a way easier way to do it,” if their experience is, “I still feels harder than the way that they’ve always done it,” well, they’re experiencing that feedback, and feeling “It still feels easier to do this the way that I’ve always done it” is going to be maybe the winner for a while. But if we can take several steps out in the computer program itself and make it easier, which is really the providence of user experience designers who do a lot of work and research around how to make this user interface work as well as possible for people, then that often ultimately has a bigger effect. I mean, in the hand washing and healthcare example, compliance with hand washing guidelines used to be, I don’t remember what the dates are on this, but this is quite a while ago, but used to be estimated by researchers at about 40%.  

In modern healthcare settings, it’s probably closer to like 70%, although I will admit I haven’t pulled the stats in a while, so I could be out of date. But that 30% increase was actually much more attributable to changes to the environment, where they made hand washing easier, and largely the addition of and availability of the alcohol-based hand rubs that were just plain faster than washing your hands at a sink for 20 seconds. So, we saw a much bigger change in adjusting the environment than necessarily something that we would see in terms of trying to persuade people to take it more seriously and try harder.  

Randy Bowman: So some of the same concepts from, I think, Thaler’s Nudge book there. How do you make it easier for people to do the right things?  

Julie Dirksen: Right? Yeah, he talks about choice architecture and things like that. Some of the Nudge research has gotten a little bit wonky, but it can absolutely be part of a design intervention to try to find some ways to make the good choices easier to do.  

Randy Bowman: Get rid of the friction that’s there. We’ve talked about some ways of making it easier, and the need to get rid of that friction. What is one thing that our listeners, as soon as they finish listening to this podcast and subscribe to it, can then go and implement to make behaviors easier, instead of just making knowledge stickier?  

Julie Dirksen: Yeah, I mean, I really think the biggest thing, and I have some resources I’m happy to share around it, but really making sure that you understand why the behavior is happening. There’s a concept in psychology called bounded rationality, which is the idea that people’s behaviors make sense within the rational sphere around them. It’s usually us looking at the outside going, “Well, that’s dumb.” The issue is if you’re not solving the right problem. You can tell people louder, more emphatically to do the behavior, but if you don’t really understand why they’re doing it, is it access to materials? Is it that they don’t understand the downstream consequences? Is it that they haven’t practiced enough to really sort of build that habit muscle around it? Is it that they just don’t believe you? Is it that you think it’s easier or better to do it this way, but they actually have real reasons for not doing it?  

There was one anecdote—I was talking to a client where they were talking about safety glasses and the frustration that people weren’t wearing these safety goggles in a, I think, manufacturing setting or construction, I can’t remember. But one of the problems was that the glasses would fog up, and so the reason people wouldn’t wear them is because it actually impaired their vision and made them less safe. That’s not a training problem to fix; that is a different issue that needs to be addressed. So, ensure that you’re solving the right problem. I want people to do this new behavior, but maybe they just don’t feel confident. Well, confidence can come from a lot of things. It can come from them having some experience in a safe environment, or it can come from seeing somebody else do it and go, oh, okay, that’s not so scary. I can do that myself. It can come from other elements of social proof or other examples from norming to behaviors. But if we don’t know what the issue is, then we can’t create a good learning design. I think it’s a lot about looking at the question of what’s really missing here.  

One of the models that I talk about in the Elephant book a lot is COM-B, which is out of the University College, London Center for Behavior Change. At its top level, it’s a really simple question—is this person capable of doing the thing? Do they already have the capability? Well, if they don’t, then training’s a great answer. Do they have the opportunity to do it? And that might be that physical opportunity, or it might be the social opportunity. If we’re saying, this is a good behavior, but a lot of people in their environment are modeling just the exact opposite, yeah, you’re not going to fix that with training either. And then are they motivated to do the behavior? That gets into a whole bunch of interesting conversations around where and why, and how people are motivated around behaviors, which is probably the answer to a longer question.  

Meghan Conan: Julie, this has been a completely fascinating topic to me. I have to say one of my biggest pet peeves is that statement: “It’s the way we’ve always done things.” I think you’ve given myself and our listeners a lot to think about in terms of behavioral change and how it comes to learning. Here at IACET, we always like to end our podcast on one final question. What does a world that learns better look like to you?  

Julie Dirksen: I think the feedback question is the really big one. I think most people are really starved for good feedback; learning designers often don’t get enough feedback on the results of their efforts. And I think learners often don’t get enough feedback on their actions and help, especially when we’re talking about behaviors. Are they doing it right? Is somebody helping them course correct? Are they getting opportunities to practice in key ways? So, I think a system where the feedback all works better is the learning environment I’d love to see.  

Randy Bowman: Wow. Julie, thank you so much for such an informative and engaging discussion today. This was really too short. We could probably go on for another two hours about all of these great concepts. They’re just so helpful to so many people. What really stood out for me today is the biology behind learning, and how important it is to really understand how brains aren’t always wired to change just because we know better. As someone who comes from tech, I usually assume that once you show someone the better path, they’ll take it. But, Julie, as you reminded us, change isn’t about logic. It’s about emotional environments and deeply habitual behaviors that are already there. I appreciate that challenge and that reminder as I move forward, at least helping people use their technology better. Meghan, I’m sure you have lots of thoughts, too, but what was your key takeaway?  

Meghan Conan: So many thoughts, and I agree this conversation could go on, but I think one of the biggest things that stood out for me is thinking about the conscious change in the environment, how we can be better prepared to make those changes, and apply them to our world. As a developer of training myself, it’s given me some time to think about how I can ingrain that into our courses. How can I help provide some hints and tips to help change those behaviors or guide them towards the change, to make the updates that they want to make, and to apply those items that they’ve learned in their real life, even when it goes against the way they’ve always done things, as we like to say.  

Julie Dirksen: Yeah, absolutely. Enough practice to start getting comfortable with any behavior is one of those critical components that shows up all the time. I would say that that’s a challenge for a lot of instructional designers because there’s typically this attitude of “We don’t want to take people away from the floor for too long” and “We don’t want the training to go on too long.” But practice is such a crucial component that I do think that’s one of the key ingredients if you really want to see the behavior change.  

Meghan Conan: No, I agree completely. As we wrap up today’s discussion on Talk to the Elephant: Designing Learning for Behavior Change, we’d love to hear from you, our listeners. Have you ever designed a training that taught everything right but still didn’t change behaviors? And what would you do differently now, after listening today?  

Randy Bowman: Don’t forget, you can submit topic ideas, suggestions for guests, review our archives and more, at the CET Talks website available at cet-talks.org. Don’t forget to subscribe to this podcast on your favorite platform so you don’t miss any episodes. Thank you so much for joining us today.  

Host: You’ve been listening to CET Talks, the official podcast of IACET. Don’t forget to subscribe to the podcast on Spotify, Apple Podcasts, or wherever you listen to podcasts. To learn more about IACET, visit iacet.org, that’s I-A-C-E-T.org. Thanks for listening, and we’ll be back soon with a new episode.