Why do some words stay with us but others fade away? Ada Aka, an assistant professor of marketing at Stanford Graduate School of Business, is determined to find out. “Words are incredibly powerful. Think about this: they shape how we perceive the world around us, but also how the world stays with us over time,” Aka says.
Aka’s research examines what makes words memorable and how language impacts our perception and decision-making. She and her colleagues recently conducted an ambitious study in which hundreds of undergraduates performed recall and recognition tasks over the course of 25 sessions. The results? Some words, regardless of where or how they appear, are simply stickier than others.
“Certain words are intrinsically more memorable,” Aka tells Think Fast, Talk Smart host and Stanford GSB lecturer Matt Abrahams. Concrete words, like “mountain,” are more likely to be recalled than abstract ones. Emotional words, particularly those related to loss or social connection, also stand out. Even informal language, like “oops,” can increase recall.
Aka’s research also explores how different communication styles affect the information we retain, including what we learn from AI-powered large language models. Her findings reveal a surprising tradeoff: The more conversational and engaging an LLM sounds, the less likely people are to remember precise details of what it says.
“If you’re teaching something quite technical, for example, it’s not good to use things like, ‘Let me tell you how this works,’” she says.
Aka’s work also extends to branding, where the most effective slogans blend creativity, humor, and psycholinguistic variables like emotionality. “Alignment and fit between a brand’s personality and values — as well as the message that this little multi-phrase or singular sentence slogan gives and how those two fit with one another — seems to be one of the more important factors,” she says.
For anyone crafting a message, Aka’s research proves that choosing our words carefully can be the difference between being heard and being forgotten.
Full Transcript
Note: This transcript was generated by an automated system and has been lightly edited for clarity. It may contain errors or omissions.
Matt Abrahams: The words we choose matter. They can help us be memorable, or they can lead our messages to be forgotten. My name’s Matt Abrahams and I teach strategic communication at Stanford Graduate School of Business. Welcome to Think Fast, Talk Smart, the podcast.
Today I’m really excited to speak with Ada Aka. Ada is an assistant professor of marketing at Stanford GSB. Ada’s research explores consumer memory, judgment, and decision making to uncover psychological mechanisms for consumer behavior and memory. Ada, I’m so glad you could join me, thanks for being here.
Ada Aka: Thank you for having me, Matt. It’s a pleasure.
Matt Abrahams: Shall we get started?
Ada Aka: Sure.
Matt Abrahams: Excellent. I’ve always been fascinated by words and word choice, and your research explores these topics. Specifically, you study what words seem to be remembered better than others. Why is word choice so important, and why is word recall interesting to study?
Ada Aka: Yeah, words are incredibly powerful. Think about this: they shape how we perceive the world around us, but also how the world stays with us over time. And in my work, as well as in the memory literature, we’ve found that carefully chosen words, they’re going to be taking the attention. So at a presentation, they’re going to be very important, but at the same time, if they stick with us, they’re going to influence our future decisions, our future thoughts and actions, so they’re going to be quite important from that sense as well.
We can say that words are the building blocks of how we perceive the environment that’s around us. And then methodologically, I’m a researcher, so one of the advantages of words and word lists is that over the years, what we have found is everyday experiences and memories are quite complex. I think you can relate to this from your own experiences. That being said, if we look at more constrained, very high internal validity or lab environments with words and word lists, the memory dynamics actually mimic how people remember in everyday life as well. So I think that’s a very nice, clean way for us to uncover the deeper mechanisms of the memory dynamics through words and word lists.
Matt Abrahams: So it sounds like you’re fascinated by words in and of themselves, but you also use words as a tools to better understand memory.
Ada Aka: Exactly. So they’re interesting, but at the same time, methodologically, they’re powerful themselves.
Matt Abrahams: Wow. At some point, I want to get more into the, to the memory piece, but for now, I’d love just to stay at a higher level and can you share some of your research results? Which types of words tend to stick in memory better than others?
Ada Aka: Let me tell you a little bit about the types of studies that we run to give you context of how these things work. We were involved in a very large mega study with undergraduate students over maybe a six month period, where the students didn’t come once, but they came repeatedly, 25 times, to our experiments.
Matt Abrahams: Wow, that’s a lot.
Ada Aka: Exactly, and in each of these sessions, they saw the same set of 576 words in randomized order within different lists. So at the end of these sessions, they either did a recall task, they were trying to remember as much as they can, or they did a recognition task. We showed them words, they said whether this was a word that they’ve previously seen or not. And what you can do is, at the end of everything, average all of these probabilities to say, what are the words that stick in people’s minds over time? And then, that’s the point that I think is quite fascinating: certain words are intrinsically more memorable than others, beyond where they were presented, what they were next to, or who the person was even. And those types of things, I think we can look at two different buckets, what we call psycholinguistic variables that relate to language related properties of the words, things like concreteness, a word like mountain is going to be more memorable. Of course, emotions matter quite a bit as well, both in terms of valence and arousal. And then, um, contextual diversity was another variable that stood out. So how many different contexts things appear relate to how much you’re going to remember those words later on.
So that’s just one bucket, but we can also think about things relating to semantic meanings of the words. There, we see emotion popping up a lot again, so words relating to death, for example, or words that relate to religion are some of those things that people remember. And then social processes: perhaps you can think from a more evolutionary perspective, why these types of words might be relevant. And in terms of the, the methods that we use, basically, we have an algorithm that you give it any type of work you’re interested in, you can get a prediction, a fairly accurate prediction on how much other people will remember that particular word.
Matt Abrahams: All right, thank you for the methodological breakdown. Help me understand: What are some words that we should be thinking about, or types of words? So it sounds like words that evoke emotion.
Ada Aka: Definitely.
Matt Abrahams: And words that are concrete.
Ada Aka: Exactly.
Matt Abrahams: Give me some other examples. So when I’m thinking about choosing the right word, I should lean towards words that are more like what?
Ada Aka: This is going to sound very context dependent, but actually informal language also stood out as being some of the words that tend to be more memorable. So if your context allows for it, I would also say incorporating, scattering some of these informal language words like even a word like oops, for example, might be relevant in terms of catching your attention. And later, making you remember not just that word, but that’s what’s around that particular word as well.
Matt Abrahams: That’s really interesting because a lot of people feel that when I’m communicating in an important situation, I need to use more formal, more rigid language. And what I’m hearing is if we actually want to be more memorable, maybe more conversational, colloquial language makes sense.
Ada Aka: I think there’s a right balance of if you want people to remember exact information versus more of the gist of what’s happening. And I think every once in a while, these types of words could bring back the attention. So it can help you with certain situations.
Matt Abrahams: To be clear, if I want people to remember the specifics, I would be more formal. Is that what you’re implying?
Ada Aka: Exactly. That’s right. In fact, that relates very nicely to some of the recent work that I’m doing. As you know, large language models, they’ve been everywhere. And I’m interested in studying how does learning from a large language model, particularly one that’s quite chatty, quite conversational in nature, influence what people remember from that interaction.
Matt Abrahams: Oh, how cool. So you’re studying not just how people remember eachother’s language, but how we remember language and conversations that come from our conversations with AI.
Ada Aka: That’s exactly right. Yeah. And we find more of a knowledge gap here. The more chatty or the more conversational the LLM large language model is, the less likely that people remember concrete, specific information. It’s not. The gist, I can’t speak to that, but if you’re teaching something quite technical, for example, it’s not good to use things like, let me tell you how this works, etc.
Matt Abrahams: Oh, interesting. So just cut to the facts.
Ada Aka: Exactly.
Matt Abrahams: If at least you’re an LLM. I wonder if, is that true for humans too? Do you know?
Ada Aka: I might, I don’t have necessarily empirical evidence from this particular study, but I would guess that, um, stories, for example, we have a lot of research showing the importance of stories, so it might be different for how humans interact.
Matt Abrahams: Interesting. So we might have a different way of remembering things that come from technology versus from other people. As we craft messages, can you provide some advice and guidance on how best to make those messages memorable?
Ada Aka: Yeah, if we move from individual words and intrinsically memorable words, of course, they’re going to be valuable. But I think I can rely a little bit more on my research on slogans. So brand slogans and what makes a particular brand slogan memorable. There we find that alignment and fit between a brand’s personality. And values perhaps, as well as the message that this little multi phrase or a singular sentence slogan gives, and how those two fit with one another seems to be one of the more important factors.
So I would say that depending again on the context and what you are trying to create in terms of the effect that you want people to lead with, that’s going to be one of the more powerful methods to incorporate.
Matt Abrahams: So, word choice that’s congruent with the actions or the feelings you want people to have.
Ada Aka: Exactly.
Matt Abrahams: Tell me more about your research with slogans. I’m fascinated by that. I mean, people, we’ve talked a lot on this show about personal branding, but certainly companies and corporations use slogans to help with their brands. What did you learn in your research?
Ada Aka: I think those two are quite related, actually, if you think about yourself as the brand, what are some of the values that you want others to think about you? There we see some of these psycholinguistic variables we talked about earlier in terms of being concrete, being emotional. And then creativity and humor stand out as additional linguistic features that make particular slogans memorable. In fact, one of the most memorable slogans, I think, is Skittles taste the rainbow.
And you might see, even with exemplars, even with just a few examples from the top, from the bottom, you can see some of these patterns that emerge. And once we do these research, we always control for what the brand is. So it’s not because, for example, McDonald’s, I’m loving it. It puts so much advertisement into their process such that their slogan is memorable. Of course, they’re going to be more memorable, but even after controlling for that, the word choice as well as how these come together influences memorability.
Matt Abrahams: The podcast is named Think Fast, Talk Smart, which is grammatically incorrect. And as a grammarian, it bothers me, but I think it has some impact. Apple is famous for think different, and again, grammatically incorrect, but it stands out. And so I’m curious, in part of the Skittles example you used, it stands out because obviously you can’t taste a rainbow. So there’s something that’s different, unique, stands out, isn’t quite right. Did your research show that that can be helpful as well, something that’s a little bit off makes us or helps us remember it more?
Ada Aka: Yeah, differentiation and distinctiveness are some of the measures that we include in our studies, and they do show benefit for, for these slogans. So I think you did a pretty good job in terms of thinking what the podcast is labeled as.
Matt Abrahams: Beyond studying memory and words, you also study decision making and judgment. Are there any particular counterintuitive ideas that come from that area of your research that our listeners might benefit from in terms of their communications with their family, their friends, or their coworkers?
Ada Aka: One counterintuitive finding from our research is that people, we’re not great at predicting what’s going to be memorable. If you ask me to create a sentence that’s going to be memorable, or if you ask me if someone told you this, how likely that you would remember over time, neither of those things that we can do very well. And there’s a positive correlation between how well something is going to be memorable and people’s predictions about it, but this is such a weak correlation that I didn’t necessarily expect a priori.
On the other hand, if you use these psycholinguistic features we talked about, or the semantic, the meaning of those words or phrases or messages, these tend to do much better at predicting future memorability. So I thought that was quite counterintuitive. Even though we interact with, we communicate, we hear, we remember every day, we’re still not necessarily very good at predicting what’s going to be memorable. Perhaps we concentrate on the wrong things. Maybe there are a few ideas we have in our minds and we only think about that variable when determining memorability, but in our work, we see that it’s not a good mapping between what we think versus what’s actually going to happen.
Matt Abrahams: So perhaps the takeaway from that research is we should test things out.
Ada Aka: Definitely.
Matt Abrahams: And we should see because what we think might be memorable might not actually be memorable.
Ada Aka: Exactly. And there’s so much psychology work concentrating on how algorithms or regressions or models predict way better than humans themselves instinctually predict these types of circumstances, and it’s no different for memory either.
Matt Abrahams: All right. Well, there you have it. Yet more evidence that we have to be focus grouping and practicing and testing our messages out with our audiences to see not just if they’re understandable, but if they’re they’re memorable along the way. I want to come to this notion of framing just for a second. It just popped into my head to talk about this. We have choices in the words we use to describe things, and have you done any work in the notion of framing and what that means? I’ll give you an example. So I teach a very simple structure for communication when you’re trying to persuade. Problem, solution, benefit. We can reframe the problem, though, as an opportunity, right? You’ve heard that an opportunity is just a problem turned inside out. That changes, though, more than just the words I use. It changes the way people feel about it. Any thoughts about framing in terms of those choices?
Ada Aka: Yeah. In terms of our models, I said that we can now have this machine that can make predictions for any given word. If we look at that, we actually see that more negatively framed words to be more memorable. So if you’re talking about even something like pain versus pleasure, pain tends to be more memorable. And that has some psychological behavioral relationships to things like prospect theory or loss aversion. But the idea here is it has to still be consistent with the general message of your class. But if you individually look at words, I would suggest that the problem framing would be better than the solution framing.
Matt Abrahams: I see. So, so negative tends to be more memorable than positive.
Ada Aka: Exactly. Mechanistically at least.
Matt Abrahams: Right, and that taps into prospect theory, risk aversion, those kind of things. One of my favorite studies on framing, divided people into groups and presented them with this issue that said you have a terrible disease, but they offered a treatment that had a 67 percent failure rate or a 33 percent success rate. Same idea mathematically, but people’s behavior was very different in terms of choosing it simply because of the way it was framed. So I find that stuff really fascinating.
Ada Aka: Definitely. Framing effects are very large and very interesting to study related to the memorability aspects as well.
Matt Abrahams: At the GSB, you teach a class called Customer Experience Design. Can you share some of the concepts you cover and two or three takeaways from the class that might help all of us?
Ada Aka: In CX Design, we focus on creating meaningful and memorable experiences. And some of the key concepts that we study relate to looking at customer journeys as well as experiences through every touch point. So of course we can study a certain point in time in great depth, but what we’re teaching our students is to be able to see this whole journey that includes a very large timeframe to be able to do a comprehensive and complete analysis of that. And at Stanford Design School, we have amazing researchers and amazing practitioners there. So we collaborate pretty frequently with them. And what we teach our students is the idea of design with empathy. How can you put yourself in the shoes of your customers, but do this not just at once during the process or prioritize a certain moment in time, but instead create this coherent branded experience that lasts over time. So that’s going to be one of the core principles that we talk about.
And then in terms of creating memorable moments, just like earlier, in terms of mispredictions of what is going to be memorable. We talk a lot about prototyping and testing out the individual differences or situations. We also concentrate on memorability and creating the memorable moments, but also prototyping and testing our initial hypotheses, our initial expectations. Sometimes it’s personalization, it’s a personal touch that you add. Other times you can leverage the recent advances in AI and these large language models, for example, to be able to find what this particular customer might be saying or might be interested in seeing in the future. So integrating a lot of different things that incorporate AI and prediction models, trying to create these personalized, memorable approaches is going to be a theme that keeps coming up in my class.
Matt Abrahams: I like this idea of looking at the experience as the whole journey. It’s not just that one moment. When I think about communication, many people might orchestrate a meeting and focus a lot on the agenda, but they don’t think about the room or what the invitation to the meeting was like or what the follow up is. So it’s thinking about the entire journey and experience that’s important, and then understanding or realizing that we’re not super good at predicting where that most memorable moment will come and we can actually learn from and leverage what AI can do to help us with that. That’s great. And I think all of us need to be thinking about our communication in the context of the complete experience that takes place.
Before we end, I’d like to ask you three questions. One I make up just for you and the other two are similar across everybody I interview. Are you up for that?
Ada Aka: Sure.
Matt Abrahams: Across all that you study and teach, if you were to provide guidance on how to make our messages and ideas stand out and be memorable, what one or two concepts would you really want us to understand?
Ada Aka: I think one of the core dimensions is going to be the fit between the context, the message, and in fact, the individual words that you choose within that message. If you think about fit, it applies so much more broadly. The fit between the context as well as the message and their alignment between the two. The more likely that the message resonates with a person, the more likely that they’re going to remember. Additionally, the words and ideas within the message have to be coherent as well.
And this goes back to a different memorability related phenomena called semantic congruence. The idea is quite simple. If you remember one thing, the next thing that you’re going to remember tends to be quite semantically related to it. So if you have a message, maybe multiple sentences, you make the person remember one of it, then that particular remembered message can cue the rest of the information that you’re trying to provide so you can use some of these memory dynamics together to be able to create these memorable messages. So fit was one of them, we talked about storytelling framework creating these resonating and emotionally appealing concrete stories that people can take and walk away. And then the last one is repetition. So trying to bring these ideas alive over and over again and making sure that the things that you mentioned: fit storytelling framework as well as repetition keeps coming back.
Matt Abrahams: So we have to say the same thing repeatedly. We have to use frameworks. Stories are wonderful frameworks. And then ultimately we have to have fit in what’s congruent and that’s what really helps.
Ada Aka: That’s a great summary.
Matt Abrahams: Question number two. Who is a communicator that you admire and why?
Ada Aka: Um, I think I will say [Hermann] Ebbinghaus for that, who’s a pioneer of memory research. And the reason for that is he studied forgetting curves. So he didn’t study memorability, but he talked about when and why people forget things that they learn. And the way he revolutionized the field or our understanding of memory is his conveying of this so clearly, such that, many people, in fact, do not forget the learnings from his work. He worked by himself, so he studied memory using these words and lists on himself. And I think that’s quite the dedication and power to be able to go through the process, but also explain it in such a clear way that sticks in mind.
Matt Abrahams: I love this idea of studying so people don’t forget what you say. I think I need to spend more time doing that myself. All right, our final question. What are the first three ingredients that go into a successful communication recipe?
Ada Aka: Emotions. Emotions and messages that evoke feelings. Things like humor or curiosity or creativity. Some of the themes that emerge in our work with slogans as well as words. That’s going to be my number one. In terms of two, ensuring that the message aligns seamlessly across different parts. So we can label this as coherence and then emphasize the fit with that as well. And then third, I’ll still go back to repetition using memorable words and reinforcing your key ideas. So emotions, coherence and fit, and repeat.
Matt Abrahams: Excellent. So it’s about having some emotional connection, making sure it fits, repeating those points. And if we do those things, our messages will be memorable and we’ll get our point across.
Ada, this has been a fantastic conversation. You’ve really illuminated a lot of factors around word choice and how to choose those words appropriately given the circumstances we find ourselves in. And you did so in a memorable way. Thank you so much.
Ada Aka: Thank you for having me.
Matt Abrahams: Thank you for joining us for another episode of Think Fast, Talk Smart, the podcast. For more tips on language and word choice, please listen to episode 80 with Jonah Berger. This episode was produced by Ryan Campos and me, Matt Abrahams. Our music is from Floyd Wonder, with thanks to Podium Podcast Company. Please find us on YouTube and wherever you get your podcasts. Be sure to subscribe and rate us. Also, follow us on LinkedIn and Instagram. And check out FasterSmarter.io for deep dive videos, English language learning content, and our newsletter. Please consider our premium offering for extended Deep Thinks episodes, Ask Matt Anything, and much more at FasterSmarter.io/premium.
For media inquiries, visit the Newsroom.
