Observations from the past have suggested that great apes engage in make-believe. However, current experimental studies indicate that our nearest living relatives are capable of keeping track of non-existent items.
Kanzi consistently identified the correct location of the pretend juice and grape.(Image credit: Ape Initiative)Subscribe to our newsletter
For the initial time, researchers have experimentally demonstrated that bonobos (Pan paniscus), who are very similar to us along with chimpanzees (Pan troglodytes), can participate in pretend play—a capacity once thought to be exclusive to humans.
Similar to 2-year-old humans, Kanzi, a remarkable bonobo with an understanding of English, was able to keep track of imaginary juice and grapes during simulated tea parties, as detailed in a study published on Thursday, February 5, in the journal Science.
“This outcome truly amazed us,” stated co-author Christopher Krupenye, an assistant professor of psychological and brain sciences at Johns Hopkins University, in comments to Live Science. “What we are observing here is that… something that appears to be intrinsically human and emerges early in our development is also shared with our closest relatives,” he remarked.
This suggests that the human ability to visualize things that are not physically present might have originated prior to the divergence of humans and bonobos from their last shared ancestor over six million years ago, Krupenye indicated.
Imagined realities
Prior anecdotal accounts had hinted at great apes, both in captivity and in the wild, engaging in make-believe. For instance, a young wild chimpanzee in Guinea was seen manipulating a discarded human-made leaf cushion by placing it on its head. Additionally, a captive bonobo was observed “plucking” and “consuming” blueberries from a picture of actual blueberries.
However, given that these anecdotal instances could be attributed to alternative interpretations, such as the apes genuinely believing the pretend items were real, Krupenye and his associate Amalia Bastos, a comparative psychologist at the University of St Andrews in the U.K., aimed to investigate the question of “can animals truly pretend?” within a controlled, experimental framework.
Since Kanzi could comprehend and react to English, he was the clear candidate for the initial investigation, Krupenye explained.
Initially, Kanzi was trained to identify the container holding juice. He was presented with two transparent bottles, one filled with juice and the other empty, and was instructed to point to the juice’s location. Correct responses were rewarded with some of the juice. Kanzi achieved a perfect score across 18 repetitions of this training phase.
During the test trials, an experimenter placed two empty transparent cups next to each other on a table before Kanzi. Subsequently, they mimicked pouring juice from an empty pitcher into each cup, then poured the imaginary juice from one of the cups back into the pitcher. Kanzi was then asked to indicate the cup containing the juice, but he received no feedback on his accuracy and no reward.
Kanzi successfully pinpointed the location of the pretend juice 68% of the time, suggesting his capacity to track the non-existent liquid.
Kanzi passed away aged 44 in March 2025. (Image credit: Ape Initiative)
However, the possibility remained that he merely believed the empty cup contained actual juice. To investigate this scenario, the team conducted a second experiment involving a cup filled with juice and an empty cup placed on a table. They simulated pouring juice into the empty cup and then held the empty pitcher over the full cup without performing the pouring action.
Krupenye noted that if Kanzi genuinely believed both cups held juice, he would have selected them with equal frequency. Yet, when asked which cup he preferred, Kanzi chose the cup with real juice 77.8% of the time, indicating his ability to clearly differentiate between actual and imagined juice.
“That, in a way, gave us confidence that we were truly observing an ability to track imaginary or make-believe objects,” Krupenye commented.
Bastos expressed that she remained somewhat uncertain at this juncture—Kanzi’s capacity to indicate the location of the pretend juice could have been coincidental. Consequently, the team replicated the identical procedure but used a pretend grape. Kanzi correctly identified the imaginary grape’s location in 68.9% of trials.
“By the conclusion of the third experiment, I was highly assured that our observations were accurate,” Bastos stated.
While the research is constrained by the fact that only one bonobo was tested, it nonetheless represents the initial clear substantiation that great apes can participate in pretend play, as communicated to Live Science via email by Laura Simone Lewis, an evolutionary anthropologist and psychologist at the University of California, Santa Barbara, who was not involved in the study.
“This marks a significant advancement for our discipline, as it furnishes direct evidence supporting anecdotal accounts from the wild suggesting that our great ape relatives can utilize their imaginations for various activities, including make-believe,” she commented.
This research establishes Kanzi’s comprehension of shared pretense initiated by humans, but not his ability to generate pretend scenarios himself.
“I believe it would be a substantial overstatement to claim, based on this, that we are witnessing something akin to what is observed in 2-year-old children, where pretense production, including actions like drinking from empty cups, is typically routine,” Paul Harris, a psychologist at Harvard University not associated with the study, informed Live Science.
Krupenye and Bastos aspire to extend the exploration of pretend play to other great apes. “If the existing anecdotes hold true, it should follow that other apes also share this capability,” Krupenye suggested.
Primates Quiz: What do you know about our closest relatives?