Dr Katie Slocombe, University of St. Andrew's, Scotland.
Part of the show Animal Communication, Sexual Signalling and Emotions
Chris - So tell us a bit about your work and how you got into what you're doing.
Katie - I'm interested in how chimpanzees use their vocal sounds to communicate with each other about things about the world. One of the really key things about our own language is that we use words to refer to things in the world. If I say table, I'm sure you can all visualise a table in your mind. You know what that word refers to, and that's a key part of our language. Without that ability, we wouldn't be able to talk about very much. We're interesting in finding out where this ability has come from. I started looking at chimpanzees.
Chris - Are they actually our closest relatives?
Katie - Yes.
Chris - So at a genetic level they're actually our closest relatives, but do you think they're our closest relatives as far as how they function as well?
Katie - I haven't had much experience of working with other apes but I think yes, they are most like us. The ability to refer to things in the world has been quite confusing until recently because there's actually quite good evidence that monkeys can do this. Baboons and vervet monkeys all give alarm calls for different types of predators. The listening monkeys seem to understand what they mean. When they give a snake alarm call, all the listening monkeys will stand up on their back legs and scan the ground as if they're looking. If they hear the eagle call, they'll all look up to try and find the eagle.
Chris - As Kat was saying earlier, there are regional differences in some animals. Is the same true for these monkeys? If you were to record those calls, take them to another part of the world where those monkeys also live, and play them the recording, would they be interpreted in the same way?
Katie - No, I don't think so with the monkeys. I think it's quite hard wired and there isn't much learning. All the monkeys have to learn to understand the meaning, so they basically have to have the experience of hearing the eagle alarm call and then an eagle appearing. They must learn this when they are youngsters, because young monkeys will give the eagle alarm call for anything that's falling. This includes leaves and small birds. The adults know that there's no actual risk to them, but the youngsters have to learn when it's appropriate.
Chris - So do the adults avoid the youngsters because they cry wolf too many times?
Katie - Yes.
Chris - Well that's intriguing because sometimes the youngsters are going to be right.
Katie - That could happen, but generally they won't continue giving the call. However, if an adult male gave an eagle alarm call, most other monkeys would then reply with the correct behaviour and give alarm calls of their own. If it's a juvenile, they might look up to check if there's an eagle there, but not assume that the youngster is definitely correct.
Chris - Now turning to chimpanzees for a minute, what do we see in chimps as a precursor to human language?
Katie - This is the confusing thing. Up until now we didn't know anything about vocal communication in apes. There's all this evidence that monkeys can do clever things with their alarm calls but there's nothing comparable for apes. I started to look at chimps both in the wild and in captivity. I looked at the calls chimpanzees make when they're finding food, and the idea was that in some monkey species, it's been shown that the monkeys give different calls depending on the quality of the food. They have one call for high quality food and one call for low quality food. Listening monkeys can then understand something about the type food that's available to them and maybe decide whether they fancy going to eat or not. I started by recording the noises that the chimps were making when they were eating different types of food, and then recorded each individual's preferences. I did this by giving them two choices of food and seeing which food they chose over the other. In the end we could list the foods in the order they liked them. They absolutely love bread, and bananas come second.
Chris - Ok, so you knew what order they like the food in and you know the sounds that go with that particular food. What did you do then?
Katie - We then wanted to test whether listening chimps could take information about the food quality from these vocalisations. To do this we set up a playback experiment. This means that we play the calls back from a speaker, when in reality, there's no food there at all. We could just look at natural reactions, but then you can't tell what they're picking up from visual cues or if they've just seen the food directly. The idea is to give the calls in complete isolation and to say on the basis of what they're hearing, what they can say about the outside world. Before we did that, we established two trees in the enclosure. One of these gave out bread, and one gave out apples, which was a low value food.
Chris - So when you played these vocalisations back, what happened?
Katie - When the chimp heard bread grunts, he would look for longer and more thoroughly under the bread tree than he would under the apple tree. When he heard grunts that had been given to apples, he would then go and look for longer under the apple tree than the bread tree.
Chris - So that suggests that these are meaningful sounds that mean a particular thing about a particular food to these animals.
Katie - Yes.
Chris - I know we keep coming back to this, but if you have different animals in different geographical settings, do the same sounds mean the same thing to them?
Katie - We've got preliminary evidence from the wild that the grunts do sound the same and a high value food will elicit a similar call. Obviously in the wild it's not bread they're grunting at; it's a type of fig. If it's quality, it should transfer across. The only possibility that we are yet to test, is that they may not be labelling food as high and low quality, but be labelling them as something as specific as bread and apples. If it is as specific as bread and apples then that obviously wouldn't transfer across groups because they must have learnt those associations. Hopefully after we complete the experiments in the summer, we'll learn how specific these vocalisations actually are.