A new study shows that cockatoos’ ability to invent tools is due to their high general intelligence

by GrrlScientist for Forbes | @GrrlScientist

A Tanimbar corella (also known as Goffin’s cockatoo or Goffin’s corella, Cacatua goffiniana) at Abenteuer Zoo, Metelen, Germany.
hecht1969 / Creative Commons Attribution 2.0 Germany license.)

People had long (and mistakenly) thought that designing and using tools was what set us apart from other animals. We even become rather smug about this, too. Then in 1990, we were astonished to learn that wild chimpanzees use tools. After we calmed down, we decided, ok, chimpanzees can use tools, but they’re our closest living relatives, so that’s allowed. But everyone else is still stupid compared to us primates — especially those birdbrains.

Then, Betty.

Betty the crow became an international avian Rock Star after she set us straight.

In 2002, a paper was published in the journal, Science, that just blew people away. This paper described how Betty, a New Caledonian crow, Corvus moneduloides, bent a straight piece of wire into a hook specifically to retrieve a small basket containing a piece of food from a vertical tube in a puzzle toy (ref). (Ironically, Betty spontaneously crafted a hooked tool after her mate stole her tool and wouldn’t give it back.)

But when human children were tested on a similar task, they could not work out any suitable solutions to retrieve the food morsel until they were eight years old — THIRD GRADE (That’s Year 3 to you Brits)! Slowly, it dawned on some of the more intelligent people in the crowd that birds are actually quite clever, and that human arrogance is probably the one thing that sets us apart from other animals.

After the “Betty commotion” died down somewhat, we took a closer look at wild New Caledonian crows and saw that even in the wild, tool design and use is a big part of their culture. Further studies revealed that wild New Caledonian crows often make hooked tools to retrieve food, and so we decided these birds really aren’t as clever as all that: Betty and her friends were just doing what New Caledonian crows naturally do because their specialized tool manufacture and use behaviors are instinctive. Thus, we started to question just how innovative tool manufacture and use really is in a species that does this all the time.

Scientists then looked at tool manufacture and use in rooks, C. frugilegus(ref). Rooks are another corvid species but they are not specialized tool users — and just to be sure they lacked any contaminating influence from an undiscovered birdie culture of tool manufacture and use, this study focused on rooks that were raised in captivity. Yet, when tested, these birds also showed “insightful tool design and use” (including hook design) for solving food fetching challenges.

This victory by brainy birds was short-lived because primate nay-sayers came to the rescue, arguing that rooks have an innate behavioral predisposition for tool design and use. So … rooks are not really clever, either: they’re just doing what rooks naturally could do (in secret).

Are cockatoos capable of solving the puzzle toy challenges?

How could we figure out how innovative birds might possibly be? As study subjects, cockatoos seemed to fit the bill: they are not related to corvids, but they share many cultural and ecological characteristics that could favor the evolution of creative thinking as well as tool use. At the same time, cockatoos lack morphological and behavioral adaptations for tool manufacture and use, and they don’t appear to manufacture or use tools in the wild. (But keep in mind that there is one recently documented exception.)

Of course, this lack of tool design and use could be because cockatoos have the equivalent of a built-in Swiss Army Knife on their faces in the shape of a powerful, curved beak along with a fat, stubby tongue in their mouths that works much like an opposable thumb.

A research group at the University of Vienna that has been studying avian intelligence and cognition stepped up to investigate. This group has been working with a flock of Tanimbar corellas, also known as Goffin’s cockatoos, Cacatua goffiniana, for years. Goffin’s cockatoos are good subjects for studies such as these because they are opportunist island birds that feed on a variety of foods, at least some of which are only seasonally available. Thus, the cockatoos’ problem-solving and tool-making abilities probably are the consequence of flexible behavioral and learning abilities that arose as they went about their daily lives over millennia.

The research team, led by Isabelle Laumer, a doctoral candidate who studies how birds think and learn, set out to investigate how cockatoos approach challenges such as those posed by the puzzle toys solved by Betty and her friends. In these experiments, Ms. Laumer’s goals were twofold:

  1. To investigate whether innovative tool bending or unbending can arise in a bird species that apparently lacks any ecological pre-dispositions for bending a material during tool-related foraging or nest building; and
  2. To identify the effect of individual pre-experience with hooks for success in the food fetching tasks

To examine these questions, Ms. Laumer and her team designed a similar series of tests to those used to test the corvids. The researchers placed a favored food morsel (a cashew nut) inside one of two long transparent plexiglass tubes (so the cockatoos could see the nut but not reach it) and the tubes were either lying horizontally or standing vertically. The researchers offered the test cockatoo either a bent or straight green pipe cleaner (depending upon the task that the parrot was faced with), and a green piece of string as a distractor.

“As in the Betty studies, we confronted our animals with a vertical tube containing a reward basket with a handle and a straight piece of pipe cleaner. In a second task with a horizontal tube containing a reward at its centre and a piece of pipe cleaner that was bent at 90°”, explained Ms. Laumer who conducted the study at the Goffin Lab in Vienna.

“Retrieving the reward from the vertical tube thus required the birds to bend a hook into the straight pipe cleaner to fish the basket out of the tube. The horizontal tube in turn required the birds to unbend the bent piece of wire to push the food out of the tube.”

Figure 2. (a) Pre-experience A in the vertical (above; wire bent to hook) and horizontal condition (below; wire straightened). (b) Pre-experience B in the vertical (above; pre-bent hook and string inserted) and horizontal condition (below; straightened wire and string inserted).
(Credit: Isabelle Laumer et al, doi:

The cockatoos were divided into two groups. The first group received stepwise “scaffolding of experience” where the birds were taught how to manufacture and use special hook tools to retrieve the food morsel. The other group of cockatoos, which served as the control group, were given the same number of opportunities to manipulate the puzzle toys but were provided no scaffolding education.

In this test paradigm, if the cockatoos are fast learners and had good generalization skills, they would benefit from scaffolding and do better than the control group. On the other hand, if cockatoos are creative thinkers and are naturally innovative, the control group of birds would spontaneously create the appropriate tools independent of any scaffolding.

Five adult female and eight adult male captive-bred Goffin’s cockatoos voluntarily participated in these studies. They all had prior experience making straight stick tools but none had ever worked with bendable materials such as pipe cleaners, nor had they ever made or used hooks prior to these experiments.

All trials were HD video recorded and after each session the bent wires were photographed and stored. To avoid providing any involuntary body language cues to the test bird, the experimenter was present behind the camera during testing and wore mirrored sunglasses, avoided any head movements and did not speak.

Yes, cockatoos are as crafty as corvids

The researchers found that their Goffin’s cockatoos can craft a functional hook from a pliant material, thus meeting the first goal for this study. Despite the cockatoos’ apparent lack of a predisposition for tool making, three of the 13 birds tested individually could bend straight pieces of wire into hooks and four birds straightened bent pieces of wire. Two of the study cockatoos were successful in retrieving the food reward and one cockatoo became consistently successful at both tasks — bending and unbending the wire — to retrieve the cashew nut reward.

Those cockatoos that became consistently successful at one or both tasks showed a learning effect in the overall efficiency of accomplishing them: although the time spent crafting the tool remained roughly unchanged, the time spent probing for the food reward decreased.

When the corvids’ performances were compared to those of the cockatoos, the researchers found that the corvids caught on to the tasks faster — Betty the crow solved the challenge for the first time in her second trial, for example — demonstrating that the crows’ culture of tool design and use was helpful in achieving their goal. In contrast, the cockatoos appeared to be natural experimenters: their functional hook designs appeared to improve over time, with later tools having nearly perfect hooks, whereas the corvids’ hooks did not show any structural progress.

“Most hooks were bent inside the birds’ beaks at the far distal piece of the straight pipe cleaner and had steeper angles and thus more obvious hooks than the tools produced by Betty the crow,” Ms. Laumer said. “Nevertheless, they used different individual strategies to fixate the rest of the tool during the bending process.”

The second goal of this study was to identify whether previous experience with hooks affected task success of the birds. Ms. Laumer reported that having the experience of working with the pliable pipe cleaners — either straightened or bent into a hook — helped some cockatoos develop their crafting skills, but on the other hand, individual pre-experience was not essential for success in either task. Ms. Laumer noted that one of the three cockatoos, Moneypenny, who became an expert hook-bender, was from the control group and thus, did not receive any scaffolding education.

“[O]ther than corvids or the humans tested on this task, some of our birds completely lacked any experience with premade hooks,” Ms. Laumer explained. “As one of those animals still mastered the task we can assume that this type of experience does not seem to be necessary for them to find a solution to the task.”

“These findings are surprising as our cockatoos are neither specialized to tool assisted foraging, as the New Caledonian crows, nor are they bending sticks during nest construction, but breed in pre-existing tree holes,” pointed out the senior author, Alice Auersperg, who is head of the Goffin Lab.

“Considering that the tasks were solved by a limited number of birds combined with the fact that they used individual techniques for making their hooks supports the assumption that Goffin’s cockatoos have to actively invent the solution to the problem rather than retreating to inborn stereotyped behavioral routines,” Professor Auersperg explained.

This study highlights that cockatoos are not relying on instinct as corvids appear to do, but instead, they appear to be capable of creative problem-solving.

“Thus, our results suggest that their hook bending from pliant material does not require strong hereditary predispositions from specialized tool use/manufacture or nest building, but seems to arise from more general modes of cognitive processing,” Professor Auersperg said.

“It seems that, at least for now, and for this particular species, we can get the innovative aspect of hook bending off the hook.”


I. B. Laumer, T. Bugnyar, S. A. Reber, and A. M. I. Auersperg (2017). Can hook-bending be let off the hook? Bending/unbending of pliant tools by cockatoos, Proceedings of the Royal Society of London B: Biological Sciences 284:20171026 | doi:10.1098/rspb.2017.1026

Also mentioned:

Alex A. S. Weir, Jackie Chappell, Alex Kacelnik (2002). Shaping of Hooks in New Caledonian Crows, Science 297(5583):981 | doi:10.1126/science.1073433

Christopher D. Bird, and Nathan J. Emery (2009). Insightful problem solving and creative tool modification by captive nontool-using rooks, Proceedings of the National Academy of Sciences of the United States of America 106(25):10370–10375 | doi:10.1073/pnas.0901008106

Enjoy my writing? Please give me a few handclaps to recommend this piece. Follow me on Medium for more like this.

.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..

GrrlScientist is very active on twitter @GrrlScientist and you can follow all her writing by subscribing to her TinyLetter

Originally published at Forbes on 13 September 2017.

PhD evolutionary ecology/ornithology. Psittacophile. scicomm Forbes, previously Guardian. always Ravenclaw. discarded scientist & writer, now an angry house elf