Recommended Free Tools
Kanzi was not simply “drinking invisible tea.” Researchers staged a controlled, tea-party-like game to test whether the bonobo could track an imaginary object through a sequence of pretend actions. In the experiments, Kanzi usually identified where pretend juice or a pretend grape should still be—even though the containers were empty.
Kanzi was not simply “drinking invisible tea.” Researchers staged a carefully controlled, tea-party-like game to test whether the bonobo could keep track of an object that everyone knew was imaginary. In three experiments, Kanzi usually identified where pretend juice or a pretend grape should still be—even though the cups, pitcher, container, and jars were empty.
As an Amazon Associate I earn from qualifying purchases.
The study provides evidence for a specific cognitive ability: representing a “pretend object,” or maintaining a mental representation that is deliberately disconnected from what is physically present. It does not show that Kanzi understood a human tea party in every cultural sense, possessed human language, or imagined exactly as a human child does.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →What did Kanzi do in the pretend tea-party experiment?
Amalia P. M. Bastos and Christopher Krupenye tested Kanzi, a male bonobo (Pan paniscus) known for his experience with lexigrams, spoken prompts, and human communication routines. The setup resembled a child’s make-believe tea party, but the researchers’ real question was much more precise:
#1 Best Overall
- Factory sealed DVD
Could Kanzi track an imaginary object through a sequence of pretend actions and then find it when asked?
The containers were transparent and empty. The experimenter acted as though juice or grapes were present, moved them from one container to another, and then asked Kanzi to indicate where the pretend item remained. Because the locations were varied, Kanzi could not succeed simply by always pointing to the left or right.
Experiment 1: tracking pretend juice
The first test used two transparent empty cups and an empty transparent pitcher.
Free tools Windows power users keep installed
One-click scans. No signup required.
- The experimenter pretended to pour juice into both cups.
- She pretended to pour the contents of one cup back into the pitcher.
- The pitcher was placed out of sight.
- Kanzi was asked where the juice was.
If Kanzi was following the pretend sequence, he should have selected the cup that had not been “emptied.” He did so in 34 of 50 trials, or 68 percent.
That is above chance, but it is not perfect performance and should not be presented as an intelligence score. The significance lies in the pattern of controls surrounding the result: the cup positions changed, the containers were empty, and Kanzi had to follow an invented transfer rather than simply respond to a visible reward.
Experiment 2: did Kanzi know the difference between real and pretend juice?
A strong alternative explanation had to be tested. Perhaps Kanzi believed that the pretend juice was physically real. If so, his choices might reflect a mistaken search for hidden food rather than an understanding of make-believe.
Researchers therefore offered Kanzi a choice between a cup containing real juice and a cup associated with pretend juice. He selected the real juice in 14 of 18 trials, approximately 77.8 percent of the time.
That preference matters. It suggests that Kanzi could distinguish the pretend scenario from an actually filled cup while still tracking the imaginary juice in the first experiment. In other words, he did not appear to treat the empty cup as literally containing a drink when a real drink was available.
Experiment 3: tracking a pretend grape
The researchers repeated the basic logic with a different object: a grape.
- The experimenter acted as though she took a grape from an empty transparent container.
- She pretended to place it into one of two jars.
- She then pretended to empty one jar.
- Kanzi was asked to indicate where the grape should still be.
Kanzi selected the jar that should still contain the imaginary grape in 31 of 45 trials, approximately 68.9 percent of the time.
Repeating the result with pretend grapes helped address the possibility that Kanzi had learned one narrow response to the earlier juice routine. His performance was again above chance, although again far from flawless.
Why is this evidence for pretense?
Pretend play requires more than copying an action. The participant must maintain two layers of representation at once:
- Physical reality: the cups, pitcher, container, and jars are empty.
- Pretend reality: the imagined juice or grape has a location within the make-believe sequence.
Researchers sometimes describe this ability as secondary representation: a representation that is held apart from immediate reality and used to model a hypothetical situation. In humans, related abilities support pretend play, imagining possible futures, considering “what if” scenarios, and reasoning about what another individual may believe.
The real-versus-pretend control is especially important. A bonobo that simply believed an empty cup contained actual juice might be expected to choose it even when real juice was offered elsewhere. Kanzi’s general preference for the real drink suggests that he could keep the imagined object inside the pretend task without confusing it with a physical object.
What the study does—and does not—prove
The most defensible conclusion is that Kanzi could represent pretend objects under the conditions tested. The experiments were designed to rule out several simpler explanations, including a fixed-side preference and the belief that the hidden food was physically present.
That conclusion is narrower than saying Kanzi had a human-like imagination. The study did not establish that:
- all bonobos can perform the same task;
- all apes engage in pretend play in the same way;
- Kanzi understood the social or cultural meaning of a tea party as a human child would;
- Kanzi had human language or human consciousness;
- he could construct unrestricted imaginary worlds; or
- his performance was equivalent to a human child’s performance.
There was one subject: Kanzi. He also had an unusual life history, including extensive experience with human caregivers, spoken communication, lexigrams, and structured cognitive testing. Those circumstances made the experiment possible, but they also make generalization difficult. A crucial next step would be testing less enculturated apes and a larger number of individuals.
Why Kanzi was an unusual subject
Kanzi became famous through research on bonobo cognition, communication, and comprehension. He used lexigram keyboards—symbols associated with words or concepts—to request food, identify social partners, and ask caregivers to play. Ape Initiative has described Kanzi as knowing more symbols than the other bonobos at its center.
His background does not make the findings unimportant. It does mean the results should be interpreted as evidence about Kanzi’s ability in a highly scaffolded setting, not as a simple readout of what every wild or captive bonobo naturally does.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →The experiment used human prompts, pointing, transparent containers, repeated actions, and a cooperative test environment. Those features improved experimental control and made the task understandable, while also limiting claims about spontaneous, independent pretend play.
Kanzi died before the study was published
The footage and experiments concern events that took place before Kanzi’s death. Ape Initiative reported that Kanzi became unresponsive and died on March 18, 2025, at age 44, after receiving treatment for heart disease and regular cardiac monitoring.
Rank #4
Bastos and Krupenye’s paper was published posthumously in Science on February 5, 2026, in volume 391, issue 6785, pages 583–586. Its DOI is 10.1126/science.adz0743. Current descriptions should therefore use the past tense: Kanzi was tested, and he was a bonobo associated with Ape Initiative. He is not currently living there.
What viewers should look for in the video
Video coverage can make the experiment look like a charming animal version of a children’s tea party, but the important details are easy to miss. Watch for:
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware match- the use of transparent, empty containers;
- the difference between the experimenter’s pretend actions and the objects’ physical contents;
- the moment one imagined item is “removed” from the sequence;
- the change in cup or jar positions across trials; and
- the distinction between the main tracking task and the separate real-juice control.
It is more accurate to describe the footage as researchers staging pretend juice-and-grape tasks in a tea-party-like setting than to say that Kanzi was drinking invisible tea or spontaneously hosting a fully human tea party.
Read more about Kanzi
Readers who want background on Kanzi’s earlier life, lexigram communication, and the history of ape-cognition research may want to look for Kanzi: The Ape at the Brink of the Human Mind by Sue Savage-Rumbaugh and Roger Lewin. It is useful as background reading about Kanzi, but it predates and does not cover the 2026 pretend-object study.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How to support Kanzi’s legacy
Readers moved by Kanzi’s story can consider supporting Ape Initiative’s Kanzi Bonobo Fund. Ape Initiative describes its work as including bonobo science, education, outreach, and lifelong care for Kanzi’s bonobo family. This is a charitable-support option, not a purchase or a substitute for evaluating the peer-reviewed research.
Ape Initiative also documents personalized short videos featuring its resident bonobos during routine enrichment, training, or husbandry. Because Kanzi died in 2025, such a video should never be presented as a new Kanzi appearance or as a way to obtain footage of him.
The larger lesson
Kanzi’s performance does not erase the difference between human and bonobo minds. It does, however, challenge the assumption that the ability to separate an imagined situation from physical reality is exclusively human.
Best Value
The study’s strongest contribution is methodological as much as emotional: it turned an appealing story about a bonobo tea party into a test with repeated trials, changed locations, a second pretend object, and a real-versus-pretend control. Kanzi was imperfect, and the sample was only one animal. Even so, his above-chance choices suggest that, at least in this structured setting, he could keep track of something that was never there.
Frequently Asked Questions
Was Kanzi a chimpanzee?
Kanzi was a bonobo, not a chimpanzee. Bonobos are a separate great-ape species, Pan paniscus.
Was Kanzi really drinking invisible tea?
No. The experiments used empty transparent containers and pretend actions. Kanzi selected the expected location of imaginary juice or grapes more often than chance, but he was not physically drinking invisible tea.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesHow well did Kanzi perform?
In the main juice task, he chose correctly in 34 of 50 trials, or 68 percent. He selected the correct jar in 31 of 45 pretend-grape trials, approximately 68.9 percent.
When did Kanzi die, and was the study published afterward?
The paper was published in Science on February 5, 2026, but the experiments were conducted before Kanzi died on March 18, 2025, at age 44.
The Bottom Line
Bottom line: Kanzi did not prove that bonobos imagine exactly like people. The controlled experiments provide narrower—and valuable—evidence that one highly enculturated bonobo could track pretend juice and a pretend grape while distinguishing the make-believe scenario from real food.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
