What are you looking at, cockatoo? Does social learning depend on social relationships?

As a lecturer, I know my students but I do find it hard to keep track of which students are doing which modules and who graduated one vs two years ago. Complex group living requires animals to know and keep track of various relations between themselves and others. These could be seen in the form of agonistic interactions which can lead to the formation of dominance hierarchies but can also include affiliative interactions between related and unrelated individuals. This is a lot to take on board, and it is thought that keeping track of the individuals and social relationships within a group is much more challenging than, say, remembering feeding areas and places with shelter; it has been suggested that it is complex social environments that lead to high levels of cognition.

Just like us, non-human animals are good at learning from each other and are able to gather different sorts of information. This ability can be very sophisticated, such as learning about complex motor patterns, or it can be something simpler, such as when the presence of another animal interacting with an object makes you more attracted to that object later.  Interactions with group members provide opportunities for transmission of information, and learning from others is considered a short-cut to novel information. However, it is only useful when that information is current: In a changing environment it may therefore be better to learn about new situations yourself. Therefore, ideally social learning should only be used selectively, namely when it is likely to be reliable and useful.

So how do you know what to learn from others and what to learn individually? Does it matter who it is that you’re learning from?

In a recent paper published in the Psychonomic Society’s journal Learning and Behavior, researchers from the University of Vienna and the Messerli Research Institute asked just those questions about Goffin cockatoos. The researchers not only wanted to investigate whether cockatoos were more likely to interact with an object (and these guys really love objects!) if they had seen another cockatoo interact with it, but also whether it mattered who that other cockatoo was. Author Alice Auersperg explains “The aim was to investigate the effect of social relationships on social learning mechanism during unrewarded object play”.

This question has been investigated in some other big-brained bird species but is of particular interest in Goffin cockatoos because, as well as living in a variety of group sizes, they exhibit an incredible amount of object play. In fact, when in captivity, they spend most of their day interacting with objects and have been observed spontaneously manufacturing tools, as in this video:

Further, the birds are able to learn to use a tool through socially transmitted information. This make them ideal birds for this sort of question.

And they look impressive, too:

In order to pursue those questions, the researchers needed to know who hangs out with whom and is dominant over whom. So the researchers watched the birds, and they did so for a long time.

Altogether 4090 minutes of observational data were collected, by recording both affiliative and agonistic interactions alongside proximity measures. Using those measures, the researchers were able to create a dominance hierarchy and were also able to see which birds were affiliated to each other.

When combined with kinship data, those observations allowed the researchers to create pairings in which birds differed in dominance, in which they were related or in which they were affiliates (we don’t really use the term “friends” but it gives you the idea) for the main social cognition experiment.

In this main study, birds were presented with four objects and the observer birds (both birds from a pairing were observers and demonstrators across different trials) got to observe their demonstrator interacting with the object.

The demonstrator was then removed and the experimenters looked at which object the observer birds subsequently interacted with…… perhaps surprisingly the observer birds did not pay attention to the choice of the other birds at all! The observers ignored the demonstrator’s choice, regardless of whether it was a dominant animal, an affiliate or even kin.

So what was going on? The birds formed rapid individual preferences that were not based on what they had seen the pair partner do. This suggests that their exploratory behaviour is largely uninfluenced by the other members of the group.

Those data are in apparent conflict with the fact that birds are known to be able to learn use a tool by observing others. How can this discrepancy be resolved? The authors suggest that the switch between social and individual learning will depend on the benefits involved. In this case, the birds did not receive a food reward for their interactions and this might explain the lack of transmission in this context.

As the Goffin is used as a model species for physical cognition it is highly important for us to understand the mechanism involved in their information gathering processes. We found that the birds have a linear rank hierarchy with males at the top similar to those found in some primates and strong afffilative relationships to certain individuals. Nevertheless social learning does not seem to influence their object exploration if no food is involved. This is in contrast to previous findings on some corvids.” explains Alice.

Let’s hope the rewards of passing their exams is considered motivation enough for my students to learn from observing me.

Article focused on in this post:

Szabo, B., Bugnyar, T., & Auersperg, A. M. I. (2016). Within-group relationships and lack of social enhancement during object manipulation in captive Goffin’s cockatoos (Cacatua goffiniana). Learning & Behavior. DOI: 10.3758/s13420-016-0235-0.

You may also like