What A Tea Party With A Bonobo Taught Us About Imagination

Is the capacity to imagine unique to humans? Scientists thought so—but a pretend tea party with a bonobo named Kanzi suggests otherwise.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  WNYC is supported by Bear. Science is about asking questions, testing, trying again. It's not easy, but it's relentless. It evolves, and that's why science delivers. Learn more at sciencedelivers.com.

• 0:14Ira Flatow?

  Science Friday is supported by MathWorks, creator of MATLAB and Simulink software for technical computing and model-based design. MathWorks, accelerating the pace of discovery in engineering and science. Learn more at mathworks.com.

• 0:33Speaker 02

  Hey there. You're listening to Science Friday. I'm Sci-Fri producer Kathleen Davis. Today on the show, let's play pretend. Our ability to imagine is part of what makes us who we are, not just as individuals, but also as humans. It turns out, though, that we might not be the only species that's capable of playing pretend. In a string of experiments, scientists sat down, set the table, and hosted pretend tea parties with a bonobo named Kanzi to see if he'd play along. And he did. For a long time, the capacity to imagine was thought to be a uniquely human ability. So once again, the great apes have proved us wrong. Joining me is study co-author Dr.

• 1:17Speaker 02

  Amali Abostos, a cognitive scientist at the University of St. Andrews in Scotland. Welcome to Science Friday. Thanks so much for being here.

• 1:25Speaker 03

  Hi, Kathleen. Thank you so much for having me.

• 1:27Speaker 02

  So let's talk about what inspired this study. I mean, a great ape tea party sounds like something I would dream about if I took too many melatonin gummies before bed. Where did the idea for this come from?

• 1:39Speaker 03

  It actually came from Kanzi himself. So when I first met Kanzi, we were there as a research group to try and figure out what sorts of studies we might want to run with this group of bonobos, not just Kanzi, but all his friends that lived with him as well. And the very first thing that happened was... I walked up to the glass and a PhD student was next to me. And Kanzi pointed at me, pointed at the PhD student and pointed at one of the symbols on his lexigram. Kanzi has all the, had all these symbols that he could use to communicate what he wanted. And when we looked at the lexigram, it said tickle. Uh, and so he was basically asking me and, you know, the PhD student to tickle each other.

• 2:20Speaker 03

  So because Kanzi was a celebrity in the field, you know, if Kanzi tells you to do something, you just, you just go ahead and do it.

• 2:27Speaker 02

  Forget about HR, no one cares about that.

• 2:30Speaker 03

  Exactly. So we pretended to tickle each other and Kanzi seemed very entertained by that. And then he asked us to chase each other. So we, you know, pretended to chase each other around as well. Um, and that sort of led me down this path of wondering if, you know, apes uh, pretend play or have a sort of idea of what pretending might actually entail. And that's why we ended up with this study.

• 2:53Speaker 02

  Yeah. So, I mean, set the table for me. How did you do this experiment?

• 2:58Speaker 03

  So luckily for me, um, they had done this in kids before. And so all we had to do was look at the child literature. It sounds simple, but it's a, it's a little bit more complicated than it sounds. But the way that they, um, proved that children pretend play was basically by presenting two and three year olds with a pretend tea party. So they might have, for example, a, uh, you know, teapot and they pretend pour imaginary tea into one cup, then the other. And then they'll empty one of those cups and ask the kids, where is the tea? And if they're keeping track of all these sort of displacements, they will point at the cup that still contains this imaginary tea.

• 3:39Speaker 03

  Now Kanzi was not a fan of tea as most bonobos aren't. So we, so we substituted tea for juice, uh, but it's very much the same idea. So we presented him with this, uh, situation first where we had two empty glasses and we fill them alternately with imaginary juice. And then we pour one out and we ask Kanzi, where's the juice? And at that point, uh, he pointed more often than expected by chance at the glass that still contained this imaginary juice.

• 4:09Speaker 02

  So when, so when you say that, you mean that. He pointed to the, you know, correct juice-filled container, the imaginary juice-filled container more often than you could attribute to, you know, he was just randomly pointing, right?

• 4:23Speaker 03

  Exactly, exactly right. That's how we measure a lot of sort of choice paradigms with animals is whether what they've done is different from, you know, 50/50, which is chance behavior.

• 4:35Speaker 02

  I mean, how do you know that he's able to play along and imagine versus like picking the cup that he knows you want him to pick?

• 4:48Speaker 03

  Absolutely, yes. Excellent question. So there's a few different things that we can do to help that. So first of all, I wasn't the one pouring the juice. The person who was sort of doing the experiment, running the experiment, was a trained researcher who knows how to work with apes and works with Kanzi regularly, but was naive to the experimental hypotheses. So the people actually collecting the data didn't know what was expected of Kanzi and didn't know why we were doing what we were doing. They're just sort of running the procedure, which is mad, especially if you're sort of pretending to pour things into containers, right?

• 5:22Speaker 02

  They were just along for the tea party.

• 5:24Speaker 03

  Exactly, and not aware of what they were going along with.

• 5:29Speaker 03

  But, you know, even despite that, the fact that we use naive experimenter, we also considered that maybe, you know, Kanzi was 44 years old at that point. Maybe he couldn't see very well, maybe he thought there was real juice in there, maybe he was just confused. And so we decided to run a second experiment where we directly compared real juice and fake juice. And now, instead of asking him where the juice was, we asked him which one he wanted. And he definitely could tell the difference between real and fake. So it's not just that he was confused the first time around.

• 6:02Speaker 02

  So Kanzi understood that there was a tea party happening and he's playing pretend, but...

• 6:09Speaker 02

  I mean, the next level up would be, could he have hosted his own tea party?

• 6:14Speaker 03

  Oh, I wish we had had a chance to investigate that. That would have been the next step, right? So what we did do is show that he could track the imaginary displacements of these imaginary objects that were performed by a human. So the next step in the research would have been to ask this question of whether Kanzi himself might have manipulated, you know, imaginary objects and sort of carried out the imaginary tea party himself, if you will. But the only reason we couldn't do that is that Kanzi passed away before we could even ask that question.

• 6:47Speaker 02

  Kanzi passed away last year, right?

• 6:49Speaker 03

  Right, so he passed in March 2025, and the last time that I visited him was the autumn of 2024.

• 6:57Speaker 02

  Oh, were you surprised that Kanzi was playing along?

• 7:02Speaker 03

  Um, I was surprised that he continued to engage with it, because the really interesting thing about this experiment is we couldn't reinforce Kanzi for either choice. So we couldn't give him food or juice or anything, or even let him know that he was right when he made the correct choice, because otherwise we might train him to perform as we wanted in the study. Every time he made a choice, the experimenter just said, "Okay, Kanzi," and just moved on to the next trial. And, you know, you might think that Kanzi would get fed up at some point and go, "Oh, God, it's one of those pretend trials again. I'm not going to get anything. I'm just gonna up and leave," you know? But he kept engaging, which is, which is really interesting. I think shows that he enjoyed it to some extent.

• 7:41Speaker 02

  Mm-hmm. I mean, going into this experiment, how much did we know about imagination in other animals?

• 7:49Speaker 03

  Yeah, so we already had some anecdotal observations that apes, both in the wild and in captivity, behave in ways that look like they're pretending. So, for example, chimpanzee females in the wild, they'll carry sticks and logs around as if they're infants, sort of cradling them, putting them on their backs as if they're babies. And this could be, you know, that it could be the case that they're pretending that these are babies, but it's really hard to tell without testing things empirically what is going on. You know, they might be mistaken, they might think that it's a baby, but it's not. And so it's really hard to tell exactly what is going on. And that's why we need experimental studies to differentiate between the different hypotheses.

• 8:32Speaker 02

  Mm hmm. I mean, do you think that it's possible other animals outside of great apes have the capacity to imagine?

• 8:39Speaker 03

  I think it's certainly possible. I think the hardest thing is how to test it, right? Because even with Kanzi, we're hedging our bets, right? We're we're using an ape that specifically understands a bunch of English words. And we can prompt him, right? We prompted him with language. We said, hey, Kanzi, let's play a game. Let's find the juice.

• 8:56Speaker 02

  Hmm.

• 8:56Speaker 03

  If you were to just present this task without the prompting, who's to say that the animal would understand that this is a pretend play situation, right? They might get frustrated and leave because they don't see the point. Um, so really the question is, can we come up with other methods that work for less enculturated apes? Or work with, you know, other animal species as well?

• 9:17Speaker 02

  Mm hmm. I mean, I'm curious, what is the upside to imagination? Like, is there some sort of evolutionary benefit that we get from being able to play pretend?

• 9:27Speaker 03

  I think so, yeah. So my intuition is that, at least in humans, pretending is a way to prepare you for the future without any real risks, right? So you can go to a pretend tea party, you can accidentally pour pretend tea on your imaginary guest, you know? And you, you can, you know, have as many faux pas as you want. And you're gonna come away with that unscarred because it's not real life and it's totally fine. So, uh, you know, just like the chimpanzee females in the wild, they're carrying these sticks around as babies. Maybe that's a way to prepare for real babies when the time comes. And so it might just be that when children and juveniles pretend to do things, it's sort of a way to prepare for the future without any of the risks of actually taking the real actions.

• 10:12Speaker 02

  Yeah, I'm curious, this seems like, at least knowing Kanzi seems like it was very impactful for you. I mean, did it feel like you were losing like a friend?

• 10:23Speaker 03

  Yeah, it really does, and especially with such a special ape like Kanzi. You know, there's no other bonobo out there that's quite like him. He's got, he had such a unique sort of rearing history and such a unique understanding of, you know, the sort of language, for one. He understood two to three hundred English words. You could ask him questions and he would perform things that, you know, it looked like he understood the questions you were asking of him. And you don't really get that with any other ape. So he was so unique in our field and he had such a huge personality as well. I have to say, I got spat on by Kanzi more than once.

• 11:06Speaker 03

  So you miss them, you know, you get used to their quirks and their personalities. And it's a, it's a big gap when they, when they go.

• 11:11Speaker 02

  What did this experiment mean to you? Do you think about like humanity or, you know, our innate abilities differently after doing this?

• 11:21Speaker 03

  Oh, yeah, I mean, it's, it's something our field continually comes up against is that a sort of idea that humans want to feel special and humans want to feel like there's something that differentiates us from other animals, right? And every time you come up with something new. So in the 1900s, we were convinced that what made humans unique was the fact that we could use tools. And then, you know, along comes Jane Goodall in 1950 and she goes to work with wild chimpanzees. And she shows not only do they use tools, they actually make tools. And so man, the toolmaker, no longer applies as a differentiator between us and other species, right? And so it's just something that continually happens in our field as we propose that something makes us special,

• 12:06Speaker 03

  something is human unique. And then along come scientists like myself and Chris. And we ask the question, is that really the case? And we test it out. And, you know, sometimes it's not the case that it's unique. And that changes our perception of ourselves and other animals as well.

• 12:21Speaker 02

  Yeah. Well, what a lovely note to end on. Amalia, thank you so much for joining me. And shout out to Kanzi. Thank you so much for this lovely conversation.

• 12:30Speaker 03

  Thank you. Thank you for having me.

• 12:32Speaker 02

  Dr. Amalia Bastos is a cognitive scientist at the University of St. Andrews in Scotland. And if you want to see Kanzi in action at his tea party, we've got the video for you. Head to our website, sciencefriday.com/imagine. This episode was produced by Rasha Aridi, and we'll be back in your feeds tomorrow with a special Valentine's Day episode. Catch you next time. I'm Kathleen Davis.