How did you feel when you first saw Betty making a unique new tool?
Amazed! Please ask your readers to take a look at the video of Betty (). Only the great apes, and perhaps only humans alone, are thought to foresee the needs of a task and create a solution to it, but from watching Betty it seems that she can do this as well.
I am extremely conservative in attributing mental powers like this to animals, but Betty made a tool of rather complex shape for a problem that she hadn’t experienced before using a material that she had no experience of working with.
Advertisement
Betty tried to obtain the food with the straight wire but when that proved impossible she took it away, wedged it in a crevice in her tray and bent it to the appropriate hook shape. Then she went over and used it as a tool. She even corrected the shape of the hook. That implies that she saw a problem, conceived a solution to the problem and then went on to execute it. In the wild they do use twigs shaped as hooks, but the manufacturing process is totally different.
How would Betty stack up against non-human primates in her toolmaking skills?
Very well. Chimpanzees and capuchin monkeys can certainly learn how to use tools. But there are precious few examples of equally clear planning of original solutions to new problems by individuals. The use of hooks is not at all easy for either non-human primates or children to acquire. Chimpanzees may learn to use hooks but they have to learn by practice whether the angled tip of the hook has to be beyond or before the object they are trying to move. They don’t necessarily understand the physics of the manipulation but simply repeat the activity. Now, Betty could have learned something about hooks from experience in the wild with branches that have twigs sticking out of them. Nevertheless she certainly managed to transfer that knowledge to an entirely new kind of material.
Does that mean that Betty in some sense understands the physics of what she is doing?
It is very difficult to give a definite yes or no answer. From the very first time she made a hook, the impression you have is of a very determined manipulation. One has a notion that she understood what she was doing, but to make a strong case from that you need a long programme of experiments. You need to provide variation in the kind of tasks you give the animal, so that the same underlying principle has to take very different forms. Then you may be able to attribute understanding of the physical principle behind the activity.
One example of this is an experiment in which we allowed Betty to familiarise herself with two pieces of material that were left lying around on the floor of the aviary. One was a rigid T-shape and the other a floppy one. After she had played with these instruments for no particular purpose we gave her a task in which only the rigid T would extract a trapped object. She seemed to understand the needs and picked up the rigid T in the first trial of every session. This is just a start and we need to do a large number of experiments. The crucial experiments are going to be developmental. These animals live for a long time and they learn through their lives. We need to study how these skills and concepts are acquired to establish how social transmission, individual learning and genes contribute to the behaviour of adult animals. Fortunately, Betty seems keen to breed in captivity, and we are now watching the growth of her one-week-old first offspring, and incubating eggs from other crows. These newborns could help solve the riddle of how nature and nurture interact constructively to shape these extraordinary behaviours.
Are these birds really intelligent? Or are they only “intelligent” in one very special way?
To answer generally, we humans have the illusion that we are intelligent in all sorts of directions, but Chomsky was one of the first to highlight that humans’ ability for language is pre-programmed. Even human minds are organised with specific abilities, so I am not surprised that birds have specific abilities too. As the philosopher Daniel Dennett said, the task is to characterise the kind of mind that each species has rather than think that the way we think about the world is necessarily either the only one or an unconstrained one.
To answer in more concrete terms, we are still exploring whether these animals are exceptionally clever in tasks that are unrelated to the use of tools or whether they are just particularly well developed for anything that involves manipulation of objects to obtain food.
But why should only crows from New Caledonia have these abilities?
It could be a result of exceptional ecological circumstances. New Caledonia is an island that does not have any woodpeckers, so there are none of the birds that would usually extract food that lies hidden in broken logs. Moreover, it has no large native mammals, so the crows will have had limited chances to gain protein from carrion. That may have given the crows an extra opportunity to acquire these abilities. But have they done so genetically or culturally? Do all these crows start using tools automatically even if they have never seen another crow doing it? Or is tool use passed on from one crow to another?
Some studies in the wild suggest that there are regional differences in the pattern of tools that crows make (91av, 17 August 2002, p 44). That means cultural factors are important and there may even be a kind of “cultural ratchet” by which improvements are built up as time goes on. We can explore how much they can do without cultural input by rearing crows in the lab, and that’s what we are doing right now.
Crows aren’t the only birds that you have been working with…
Not at all. I have been working with starlings for a quarter of a century. One thing I am interested in is how animals take decisions with respect to risk.
A simple decision for a bird would be whether it is better to take the “safe” option of hunting for worms on the ground, where there is a relatively regular distribution of prey, or hunting for insects by hawking from a perch, where you might get a very high pay-off if a swarm passes by, or a very low pay-off if one doesn’t.
Generally speaking we find that when an animal wants more of something there is a preference for safer outcomes rather than variable outcomes with the same average pay-off, in other words, the animal is risk averse. But when making choices about things they would like to reduce, such as delays in obtaining food, or in humans monetary losses, subjects prefer variable outcomes over fixed outcomes: that is, they have a positive risk appetite.
We’ve been trying to understand the origins of some of the apparent irrationalities in decision-making that we have discovered. Our work ties up with experimental economics.
Is it true that you have set out to commercialise what you learned from starlings?
Absolutely true. Organisms, whether human or otherwise, don’t always behave as you might expect from abstract models of optimal performance. Individuals who take decisions on the boards of major corporations are using a brain that was designed for an entirely different kind of task. They make errors and sometimes they structure the problems in the wrong way by not understanding the implications of risk, or by not presenting it in a way that could be incorporated into some rational attitude for the corporation.
We began working with one business and found many ways in which decision-making could be improved. Because that worked well, we created a small company called ORRA [Oxford Risk Research and Analysis] to help businesses avoid some of these irrationalities we have been studying. And although I’ve been an academic all my life I am really enjoying a taste of the “real” world.
