Cockatoos with tools: learning from a teacher

Professor Alex Kacelnik, Fellow in Biology, and colleagues from the University of Vienna and the Max Planck Institute at Seewiesen have released findings from their latest study into the tool-making behaviour of Goffin's cockatoos which show that these birds can learn how to make and use wooden tools from each other. Their work is thought to provide the first controlled experimental evidence for the social transmission of tool use in any bird species.

Goffin’s cockatoo (Cacatua goffini) is a curious species of Indonesian parrot not known to use tools in the wild. At a laboratory in Austria the researchers had observed a captive adult male Goffin’s cockatoo named ‘Figaro’ spontaneously start to sculpt stick tools out of wooden aviary beams to use them for raking in nuts out of his reach. To investigate if such individual invention could be passed on to other cockatoos the team used Figaro as a ‘role model’, exposing other birds to tool use demonstrations, some with Figaro as ‘teacher’ and others without his ‘students’ seeing him at work.

In the experiments one cockatoo group was allowed to observe Figaro skilfully employing a ready-made stick tool, while another could see what researchers called ‘ghost demonstrations’ – either seeing the tools displacing the nuts by themselves, while being controlled by magnets hidden under a table, or seeing the nuts moving towards Figaro without his intervention, again using magnets to displace the food. The birds were all then placed in front of an identical problem, with a ready-made tool lying on the ground nearby.

Three males and three females that saw Figaro’s complete demonstration interacted much more with potential tools and other components of the problem than those seeing ghost demos. Remarkably, all three males in this group acquired proficient tool use, while neither the females in the same group nor males and females in the ghost demo groups did.

The successful birds did not just imitate Figaro’s movements: their tool-use techniques were themselves new. Figaro held tools by their tips, inserted them through the cage grid at different heights and raked the nuts towards him while adjusting the tool’s position as the target moved closer. The successful observers, instead, laid the sticks on the ground and propelled the nuts into their reach by a quick ballistic flipping movement. The latter technique was arguably more efficient for the test circumstances, which differed from those in which Figaro had made his first discovery; the pupils in this sense surpassed the teacher’s performance.

Two of the successful observers were later tested in the absence of ready-made tools, but offering them suitable tool-making material. One of them spontaneously started to make his own tools out of a wooden block, while the other initially failed, but then did so after a single demonstration session watching Figaro carve tools out of a block. Whilst inconclusive, this shows that learning to use tools may in itself stimulate the acquisition of tool-making, which is more distant from the target behaviour and closer to behavioural planning.

Professor Alex Kacelnik said: ‘There is a substantial difference between repeating a teacher’s behaviour and emulating his or her achievements while creating one’s own methods. The latter implies a creative process stimulated by a social interaction, while the former could, at least potentially, rely on simpler imitation. The cockatoos seem to emulate and surpass their teacher, which is what all good professors hope for from their best students.’