Research results, Scientific Publications

For humans, death is surrounded by culture, emotion, ritual, and language. But the question can be framed in a much more basic way: what would an animal have to understand in order to recognize that someone has died?

Perhaps we do not need to begin with death in all its complexity. At its most elementary, recognizing death may begin with something simpler: understanding that a living being, once expected to act, move, respond, or behave in certain ways, has lost those functions, and that they will not simply return.

This idea lies at the heart of the minimal concept of death, a philosophical proposal developed within comparative thanatology, the study of how humans and other animals perceive and respond to death. According to this view, understanding death does not necessarily require a human-like idea of mortality, an abstract theory of life, ritual, language, or grief. At its most basic, it may depend on recognizing the persistent loss of expected functions.

A new study on Goffin’s cockatoos approaches this broad question through one experimentally tractable component of a minimal concept of death: whether the birds can learn that something which previously produced a reward has stopped working in a specific context, and flexibly adjust their behaviour.

The study, conducted by researchers at the Messerli Research Institute of the University of Veterinary Medicine Vienna, grew out of a collaboration between the philosopher Assoc. Prof. Dr. Susana Monsó, author of Playing Possum: How Animals Understand Death (Princeton University Press, 2024), and cognitive biologists Dr. Antonio J. Osuna-Mascaró and Assoc. Prof. Dr. Alice Auersperg, who study problem solving and technical cognition in Goffin’s cockatoos.

“Understanding death may sound like an all-or-nothing question, but it can be broken down into simpler components,” says Monsó. “One of these components is the ability to recognize that an expected function is no longer there, and that this loss is not just momentary. This study allowed us to test that component in a controlled way.”

Goffin’s cockatoos are a particularly interesting species for this question. They are highly innovative parrots, known for solving complex physical problems, manufacturing tools, transporting tool sets, and discarding tools when they are no longer functional in a specific context. In the new study, the researchers asked whether they could also learn about long-lasting non-functionality in a more abstract, non-tool context.

To test this, the team designed a touchscreen task. The cockatoos first learned a simple rule: if they touched a round button on the screen, they received a small food reward. They could also touch an arrow to move on to the next trial.

The researchers then introduced a complication. Sometimes, when the birds touched the round button, the screen produced a very salient flashing event. From that moment on, the button stopped working, but only when it appeared against that particular background. Against other backgrounds, the same button could still work.

The question was whether the cockatoos could learn more than “this button no longer works.” They had to learn a more flexible rule: when this flashing event occurs, this particular context is no longer useful. More importantly, they had to apply that expectation to new backgrounds in later sessions.

Osuna-Mascaró, co-first author of the study, describes the experiment as follows: “The experiment may look complex, but the idea is quite simple. Imagine that a button works perfectly, and then, after something happens, it stops working, but only in one specific situation. We wanted to know whether the cockatoos could learn not only that this particular button was no longer worth trying in that context, but also what the event meant more generally: when a new background flashed, would they expect that this new background had now stopped working too?”

This is exactly what the researchers found. With experience, the cockatoos learned to skip the backgrounds in which the button had stopped working, instead of continuing to insist. But they did not stop using the button in general: they continued to press it in contexts where it still worked. In other words, they had not simply lost trust in the touchscreen, but were adjusting their behaviour to the specific context.

The birds’ performance was not perfect, and there were strong individual differences. Some individuals adapted faster than others. Some also showed visible reactions to the flashing event, such as vocalizations or crest raises.

Eleonora Rovegno, co-first author and the researcher who conducted the tests, describes how the cockatoos reacted to the button turning non-functional: “It was striking to see how differently individuals reacted when the button suddenly stopped working. Although these emotional reactions were not part of the statistical analysis, some birds showed very visible responses, such as vocalizations, crest raises, begging calls, or stronger arousal. These observations made the task feel very real from the birds’ perspective: something that had just worked suddenly no longer did.”

The researchers emphasize that the study does not show that Goffin’s cockatoos understand irreversibility in a human-like conceptual sense. Rather, it shows that they can learn and flexibly respond to persistent, context-specific loss of function within the experimental setting.

Monsó warns about the limitations: “We should be careful not to jump from this result to the conclusion that the birds can understand death. However, the ability to recognize that an expected function is no longer there, and to respond appropriately to that loss, is one of the capacities that may be relevant for understanding how animals respond to death and other forms of permanent change.”

Because the task was implemented on a touchscreen, rather than requiring tool use, the experimental design could potentially be adapted to other species. Touchscreen methods are already used in comparative cognition research with a wide range of animals, which may allow future studies to test how other species adapt to this paradigm.

Auersperg, as head of the Goffin Lab, adds: “What makes Goffin’s cockatoos such an interesting model is that they repeatedly surprise us by carrying knowledge from one context into another. In the wild and in captive studies they learn about tools, objects, and physical problems, but here we reduced the setup to make it about functionality itself. By stripping away the physical problem and using an abstract touchscreen task, we can begin to ask how broadly how different animals understand that something which once worked may cease funtioning.”

The study adds to a growing body of evidence that Goffin’s cockatoos are highly flexible problem solvers. It also offers a new experimental approach for studying one of the cognitive components relevant to the minimal concept of death.

---

Wissenschaftlicher Ansprechpartner:
Dr. Antonio J. Osuna-Mascaró
Co-first author
Email: Antonio.OsunaMascaro@vetmeduni.ac.at
Alternative email: tayosuna@gmail.com
Telephone number: +34 626 035 544
Bluesky: @biotay.bsky.social
Twitter: @biotay

---

Originalpublikation:
Antonio J. Osuna-Mascaró, Eleonora Rovegno, Susana Monsó, Remco Folkertsma, Alice M.I. Auersperg (2026) Sensitivity to permanent non-functionality in Goffin’s cockatoos (Cacatua goffiniana). https://www.nature.com/articles/s41598-026-57007-1

---

Weitere Informationen:
https://www.vetmeduni.ac.at/en/university/infoservice/press-releases/press/sensitivity-to-permanent-non-functionality-in-goffins-cockatoos