COGNITION

No monopoly on thought

Intelligence and the power of abstraction were long considered a sole characteristic of human beings. Animals were deemed to act like robots, without thinking. In the 1970s, the cognitivist approach began to raise doubts - and continues to arouse disquiet.

© MPI EVAN
© MPI EVAN
© MPI EVAN
© MPI EVAN
© MPI EVAN
© MPI EVAN
Research work being carried out at the Wolfgang Köhler Primate Research Center, part of the Max Planck Institute for Evolutionary Anthropology, situated in Leipzig Zoo. The research focuses primarily on the great apes, studying in particular their cognitive development, learning and socialisation. 1. Chimpanzees in an observation chamber. The researcher is about to place a piece of banana in one of the two cups to test the primates’ causal understanding. 2.3.4. A bonobo, orang-utan and gorilla in the primates’ enclosure, where some of the aspects being studied are their social behaviour and their learning methods and abilities. © MPI EVAN
Research work being carried out at the Wolfgang Köhler Primate Research Center, part of the Max Planck Institute for Evolutionary Anthropology, situated in Leipzig Zoo. The research focuses primarily on the great apes, studying in particular their cognitive development, learning and socialisation. 1. Chimpanzees in an observation chamber. The researcher is about to place a piece of banana in one of the two cups to test the primates’ causal understanding. 2.3.4. A bonobo, orang-utan and gorilla in the primates’ enclosure, where some of the aspects being studied are their social behaviour and their learning methods and abilities. © MPI EVAN

Do animals think? The answer, for many years a matter of philosophy or religion, has varied through the ages. Scientists began to study the question in the 19th century, although this has done nothing to subdue the arguments or prevent deviations. Behaviourism dominated from the period between the two World Wars until the 1970s. According to this theory, animal behaviour can be explained by an unthinking automatic response to external stimuli. This response to the environment stems from animals' innate genetic programming, or from behaviour learned through repetition or conditioning, for example. There is no need to consider thought processes, and hence cognitive intelligence, to explain the performance of rats in a maze or chimpanzees finding hidden food.

"The cognitivist approach gained the upper hand in the 1970s," explains Josep Call (1) from the Max Planck Institute for Evolutionary Anthropology in Leipzig (DE). A number of factors led to this U-turn. Following years of observation, American ethologist, Donald Griffin, proclaimed the existence of an animal consciousness. Elsewhere, theoreticians of learning also started to study animals' memory and categorisation skills. In 1978, psychologist and primatologist David Premack had no hesitation in asking the question: "Do chimpanzees have a theory of mind?" In other words, are they capable of imagining that others - fellow creatures or experimenters - have knowledge, intentions, or even beliefs? (2)

Complexities of the mind

The hunt for animal cognitive abilities was on. But first it was necessary to agree on what was meant by cognitive abilities. The first hurdle was to equate thought processes as diverse as inference (understanding causal chains), the construction of abstract rules, ‘mental time travel' (episodic memory and the ability to project oneself into the future), and the theory of mind or metacognition (being aware of what one knows... and of what one does not know). Josep Call refuses to establish a link, much less a hierarchy, among these abilities, which he does not consider to be comparable.

According to Julia Fischer, professor of cognitive ethology at the University of Göttingen (DE), "inference is widespread in the animal kingdom and can be explained by simple mechanisms, as can the construction of rules. Mental time travel appears to be more sophisticated. Some animals are not ‘prisoners of the present'." That being said, much of the time the behaviourist theory is amply sufficient to explain the observations in animals. "The main methodological obstacle is still to definitely rule out any ‘behaviourist' explanation for the performance observed. That is no easy matter, especially for most experiments on the theory of mind," she adds.

What are the conclusions so far? The great apes are in the spotlight, particularly chimpanzees, which are by far the most extensively studied group. After noting their complex social bonds in the wild and being astounded to see them use tools, researchers subjected them to more controlled experiments. In 2004, one of the experiments concerned all four species of great ape. The aim was to ascertain whether they understood the concept during an experiment where, of two choices, only the one that made a noise when shaken contained food and not the other. Great care was taken to eliminate any possibility of a learned or behaviourist response. Several chimpanzees, bonobos (pygmy chimpanzees) and gorillas - but no orang-utangs - understood immediately that the food was causing the noise.

Of primates and rats

Thirty years after Premack asked his question, it is now generally agreed that chimpanzees do have a theory of mind, albeit incomplete. They appear to understand what others see and know and the purpose of their actions. That being said, Josep Call claims: "There is no evidence that they attribute desires or beliefs to others." Not only do the great apes use tools, they are capable of keeping them for later use - at least this was true of the bonobos and orang-utangs tested in 2006. The experiment, conducted at the Leipzig Zoo, was the first to demonstrate that animals have the ability to project themselves into the future.

Our close cousins are not the only animals to have such cognitive skills. In 2006, rats demonstrated their ability to understand a causal chain (an event that triggers another event) and even to take deliberate action to trigger the delivery of sugar water. Admittedly, "when they do this, rats are not establishing any sort of explanatory law, although they do recognise a causal chain," says Anne Reboul, linguist and philosopher at France's Institute of Cognitive Sciences in Bron (Institut des Sciences Cognitives). Nonetheless, it is enough for Julia Fischer to declare that "rats and great apes are able to accomplish sophisticated mental operations before they act." In 2008, Robin Murphy, a psychologist from the University of London(3), demonstrated what he claimed to be a rat's capacity for abstraction. The test was to see if rats were able to grasp that a particular sequence of sounds resulted in the delivery of food and then to transpose this rule to a similar sequence of different sounds. In short, rats needed to infer a general, abstract law from a specific situation.

Necessity knows no law

In 2007, American researchers demonstrated that rats subjected to tests responded when they knew that they could solve the problem but refrained (preferring a lesser consolation than the reward for success, but one that they were certain to receive) when they did not know the answer (for example, being able to differentiate between two different sound wavelengths). So the rats knew that they didn't know! Up until then, only the rhesus monkey had demonstrated such metacognition. Perhaps the rat is an exception? Not in the least. "It is simply a convenient species to breed in the laboratory," says Robin Murphy. "There is nothing special about rats, even if they are the cleverest at solving their own particular problems. All species, including humans, share the same base of cognitive skills, with each species developing the specific skills it needs." Indeed, the experts warn us against the common error of viewing non-human animals as a homogeneous group, since no species is alike.

Another question remains largely unexplored. For instance, why in every experiment do some individuals perform better than others? "We are just beginning to examine this subject," says Josep Call. "As we are not able to work with large numbers of great apes and know very little about their cognitive development, it is difficult to decide how much can be attributed to genetics and how much to an individual's history." So it seems that the cognitivist approach has its limitations. According to Julia Fischer, ‘fashion' plays a role: "It is bad form to publish an article showing that an animal does not know how to do something."

Patrick Philipon

  1. Josep Call, Past and present challenges in theory of mindresearch in nonhuman primates. Progress in BrainResearch, Vol. 164, chapter 19 (p 341), Elsevier, 2007 Josep Call & Michael Tomasello, Does the chimpanze have a theory of mind ? 30 years later. Trends in cognitive sciences, 12(5), 187, 2008.
  2. The term ‘belief' should not be understood in the sense of religious belief, but means belief in a reality (such as a hidden object). Belief is assessed by means of false-belief tests.
  3. Robin A. Murphy et al., Rule learning by rats, Science 319, 1849, 2008

TOP

Read more

Elephants take a look at themselves…

Happy, passing the mirror test. © Josh Plotnik
Happy, passing the mirror test. © Josh Plotnik

In 2006, researchers erected a gigantic mirror in the elephant enclosure at New York's Bronx Zoo. It was known that most animals treat their reflection as though it were a fellow creature. Only humans, great apes and dolphins recognise themselves in a mirror. So why test elephants? "To date, the few species that have demonstrated the ability to recognise themselves in a mirror are altruistic animals capable of understanding the needs of fellow creatures in trouble and of helping them. Elephants also share these attributes," explains Joshua Plotnik, one of the experimenters (1). A link between the two types of behaviour is difficult to establish and may entail an awareness of the self as a being separate from others.

The test was a success. After a short phase of exploring the reverse side of the mirror, the three female Asian elephants started to look at themselves, to perform movements to test their reflection and to examine parts of their body that they normally could not see, such as the inside of their mouths. One of the elephants even passed the ultimate test. After crosses had been painted on her forehead without her knowledge, she rubbed them out with her trunk when she discovered them in the mirror. Does this mean that elephants are capable of seeing themselves ‘from the outside'? Do they have any conception of the boundaries of their own bodies? There is no way of telling...

  1. Joshua M Plotnik et al., Self-recognition in an Asian elephant,PNAS 103 (45), 17053, 2006.


TOP