MADE for U of T | Ep. 15 | Multimedia learning today

Richard Mayer (RM): There’s just a lot of opportunities now for how we can communicate with learners, so I think that’s a plus. But basically, I think the same principles of instructional design apply to all these kind of new media opportunities we have. We’re still dealing with the same human brain, the same learning system that we humans have, so I do think we still want to take kind of a learner-centered approach. Think about how people are processing the information, and use that to design the best instructional messages we can with any of these media.

Inga Breede (IB): What are some of the common misconceptions that you’re still seeing about multimedia learning, even among experienced educators and designers?

RM: Yeah, if we look at our own, instruction, we could probably all be pretty self-critical. But, let’s see, there’s so many, so many problems, it’s hard to choose. I would say, probably having just too much “glitz,” too much extra material, let’s say, if we’re dealing with the screen. Too much on the screen. I mean, we can put a lot of cool things on the screen, but people can really only look at one fixation point at a time. So, we’re not really getting that information across if there’s just too much going on, and it’s going on too fast. So, I would say that’s one of the big problems I see, just what I would call just too much glitz on the screen.

And also, when we have words and images, which is kind of my thing, multimedia, often they’re separated, so you can’t really tell what the words are referring to in the graphic, so you have to keep looking back and forth between the words and the graphic. That wastes cognitive resources so that, they can’t be used, really, for understanding the message. So, this kind of separation of words and graphics, I think, is another problem, I see.

IB: You talked a bit about this “glitz,” we have all of these things that we can add, and really it’s about taking some of that away. So, I’d like to now talk a bit more about some of the specific principles. One of the things that’s been so influential about your work is how it translates cognitive science into these, clear, research-based principles that designers and instructors can actually use. So, I’d like to spend a bit of time focusing on some of those, starting with coherence. The coherence principle warns against those seductive details, the glitz. So, in an era of high production video and rich media, again, the abundance of tools available to us, how can designers make principled decisions about what to remove when the instructor, and we get this a lot, they want to make their content more engaging?

RM: Right, and we certainly want to motivate learners to want to learn. But I do think when we look at what we’re going to put into an instructional message, we want to kind of focus it on whatever the instructional objective is. So, anything that’s on the screen that’s not really supporting the instructional objective is something that probably doesn’t need to be there. So, that’s basically the coherence principle, that we want to weed out anything that’s extraneous to our instructional objective. But, I do agree that we want to engage students, so there are certainly ways to do that.

I mean, one whole area of research called emotional design, tries to design instruction so it kind of, elicits a positive emotional reaction from the learner. But we can do that without adding extraneous glitz. We can take the actual content that’s important and make that more pleasant.

So, for example, one thing we’ve done in a lesson on how a virus causes a cold, that was originally done with line drawings. We kind of redid them so that the virus has kind of a fierce facial expression and little spikes coming out of it, and the host cell has a little kind of confused facial expression. So, adding facial expressions to the characters, it creates what we would call “emotional design,” because that it makes it more pleasant to learn.

IB: More connected?

RM: Yeah, more connected, but at the same time, it’s relevant to the storyline, because that virus is aggressive, and the poor host cell is about to get attacked. So those are not seductive details, they are relevant to the story. I think when we want to add something that is engaging, we want to make sure it’s consistent with the storyline we’re trying to get across, not some extraneous features.

IB: And the segmenting principle, which covers breaking a complex multimedia message into smaller, meaningful parts. With the learner in control of that, and the pacing, what does effective segmenting look like in a real-world multimedia material like lecture videos? So, we do quite a few lecture captures. How would that look like?

RM: Yeah, I think that is a fantastic question. I don’t think we quite know what is the appropriate length of a segment yet. I don’t think there’s sufficient research to really tell us exactly, but we know it’s pretty short. So, I would say it’s probably good to break video into meaningful segments, at points where we’ve kind of covered, like, a topic. That’s a good place to pause. Maybe have some generative learning activities, like, asking questions, or asking people to summarize, or something like that.

Exactly how long that segment should be, my intuitive feeling is, it’s like 5 to 10 minutes at the most. But that is still a research question. I think we’ve got to look at. Who’s the audience? And also the content. Where are the meaningful breaks in the content? Because we kind of want people to be able to digest one part, make sure they really understand that before we go on to the next part.

IB: And you touch on this idea of, like, a shorter amount of attention. I mean, I see it with my own kids, with, social media platforms, and I think some of us in this call today, we’ve had instructors ask “Hey, can I put something on TikTok or Instagram?” So, if an instructor asks a designer to do that, from a multimedia learning perspective, what are the opportunities and risks that you see in using these platforms for education?

RM: Yeah, I think there are a lot of exciting possibilities, and I admire people’s creativity in trying to use some of the social media platforms that we have. I do think, though, probably the same principles apply. Some of the platforms have more constraints on how long the message can be, but I think we want to take it from the viewpoint of the learner: how can we design something that’s going to be consistent with how they’re going to process the information. But I do think we can apply the same principles to those platforms.

IB: And then going back to another principle, which is the personalization principle. And this is now where I introduce the topic of, (sarcastically) I don’t know if you’ve heard about AI, but, you know, there’s this growing interest in using AI to personalize learning experiences. So, how do you see AI, artificial intelligence, aligning or conflicting with the personalization principle as you define it?

RM: Yeah, well you know, my research community, you know, is totally focused on AI. This is the question we’re all trying to figure out. I think all of us in education are, like, how is AI going to influence education? How can we design instruction and use AI, you know, to improve what we’re doing? So, we’re all trying to figure out how it can write grants, and how it can do research on this.

I do think AI has just a lot of possibilities that we can’t even imagine probably right now. We’re just starting to grapple with where this is going, but it sure seems like it’s going to have a huge impact. I do think AI could definitely support the personalization principle. In fact, you know, for the last 50 years, we’ve had this dream of personalized tutors that could interact with you, who know you, who know what kind of how you think, what you’re interested in, how you like to interact. And they could kind of be your personal guide whenever you’re trying to learn something. And now, that possibility is right here in front of us, so that’s kind of like the personalization principle on steroids.

(Inga laughs)

RM: So I do think, yeah, we can design AI experiences based on personalization, and we know from our research things like tone of voice, gesture, oh, even, well, AI maybe doesn’t give us the perspective, but even a first-person perspective when you’re giving demonstrations, those all kind of increase, I think, the feeling of personalization, the feeling that the instructors working with you. You’re kind of part of this. So, I have a lot of optimism for where AI could go, but I think a lot of times when new technologies come on board, we get obsessed with all the great things it can do, instead of how do people learn? So, my focus is on applying the science of learning to education, and technology can be a tool, but I think we have to take a learner-centered approach, think about how do people process information, and use the technology to help people, you know, improve their cognition, instead of making us have to adapt to whatever technology’s doing.

IB: So then, looking ahead, what research questions about AI and multimedia learning do you think are most important for instructional designers, and instructors, and researchers to explore next?

RM: Yeah, there are so many questions. I personally am interested in, like I said, using it as kind of a tutor. It’s almost like having your own personal TA, if you have a college course. Usually, students never ask the professor a question, never ask the TA a question, but if you have, like, an AI as your teaching assistant, then you’ve got a source to go to that’s kind of low stress. I’m just curious about how we could, in lessons, maybe have people interact with AI really as a, what I would call a “generative learning activity.” So, AI could ask questions, or could answer your questions, or could help engage you in activities. So, I think what we have there is just a lot of possibilities that we just need to explore.