What Is Intelligence

In terms of SETI activities, we can reasonably define a species as intelligent if it can build a radio telescope. The problem with this definition is that mankind apparently became intelligent only about 50 years ago! So although in a practical sense it might be a good definition, it fails on philosophical grounds. There must be a better way of capturing the essence of intelligence.

A common approach is to define intelligence in terms of certain mental tasks that we find difficult, such as playing a decent game of chess or solving an algebraic equation. However, it is not much more difficult to write chess-playing programs or automatic equation-solvers than it is to perform the activities themselves. And this software manifestly does not possess intelligence. The sorts of activity humans and other animals do without thinking are much more difficult to program. No one has yet come close to programming a robot capable of navigating the world outside or of coping with the various challenges everyday life throws up. If finding food and avoiding danger are any measure of intelligence, then the average rodent is much more intelligent than the smartest robot. So if we want to appreciate what intelligence really means, and whether humans are unique in this regard, it might help if we understood something of animal intelligence. Unfortunately, if it is difficult to define intelligence in humans, it is even more difficult to define intelligence in animals.

Most people, if asked to rank non-marine animals in terms of intelligence, would probably rate man as the most intelligent animal, followed perhaps by apes, down through dogs and cats, down further to the likes of mice and rats, down even further to birds, and so on. It is a comfortable picture for the human ego: we are at the top of the tree of intelligence, our closest relatives are clever, our pets are quite bright, and the animals we do not particularly like are stupid. Implicit in this picture, though, is the notion of evolution as progress from a "less evolved" state (rats, say) to a "highly evolved" state (us), with intelligence being the scale against which one can measure progress. This is simply wrong.

In the first place, we have no reason for supposing intelligence (however defined) is the sole criterion by which we can rank animals. Why not instead use visual acuity, or speed, or strength? Indeed, why try to rank animals in this way at all? We should not view evolution as a ladder, with ourselves at the top and all other animals below us because they are not yet "evolved enough" to possess intelligence. Apes, birds, cats, dogs, mice and men are all equally "evolved," since we share a common ancestor that lived hundreds of millions of years ago. The various species have adapted to their environments in different ways; our species has certain characteristics that make it successful, but so has every other species on the planet. These species are all equally successful, since they have passed the critical test: they all have survived. If we want to assign different levels of intelligence to different animals, then we need a better gauge than our prejudices.

When biologists try to measure the intelligence of animals, they face an almost impossible task. Measuring the IQ of humans in a non-culturally biased way is difficult enough. But if tests on humans are biased, how can we possibly test the intelligence of different animal species? How can we factor out the differences in perceptual ability, manipulative ability, temperament, social behavior, motivation and all the other variations between species? Does a monkey fail to complete a maze because it is brainless or because it is bored? If a cat fails to press a lever that produces a food reward, should we conclude that the cat is stupid or is simply not hungry? Does a rat fail an intelligence test because it is dense, or because the test demanded visual discrimination (at which rats are poor) rather than discrimination between smells (at which rats excel)? These sorts of questions make it exceptionally difficult to be sure that we are testing an animal's cognitive ability.

Suppose we try to account for as many cross-species variables as we can think of in these cognitive tests. (For example, biologists might want to investigate how many list items an animal can remember, or whether an animal can recognize a face; either of these tasks might tell us something about cognitive processes in animals. The investigator would have to ensure the details of the test were different for different animals. The tests for pigeons and for chimpanzees would have to be different, if only to take into account their different physical abilities.) Suppose further that we define intelligence, general intelligence, to be a measure of how well animals score on such fundamental cognitive tests. Then a surprising fact emerges: most animals perform at about the same level! Of course there are some differences between species, but the differences are much smaller than one might expect. Chimpanzees can remember about seven items from a list at one time — but so can pigeons (so no more cracks about "bird brains"). Monkeys can quickly discern whether pile A contains more food treats than pile B — but so can cats. In fact, if intelligence is defined as the ability to perform these basic non-verbal tasks, then one can argue that to a first approximation all birds and mammals, including mankind, are about equally intelligent! This conclusion is still controversial; but if it turns out to be true we should not be surprised. After all, every species, including mankind, has to negotiate the same perilous world; we all have to eat and drink and find mates. The basic cognitive skills enabling animals to perform these tasks might well be common to all species.

On the other hand, one can equally take the opposite approach: maybe intelligence in animals consists precisely in all the factors we deliberately omit in cognitive tests. To use a computing analogy, we should not just consider the processor (the brain) but also the attached input and output devices (the senses and manipulative abilities of an animal). After all, a chimpanzee has hands which enable it to perform tasks that a cow simply cannot attempt. From this viewpoint there might be little general intelligence residing in the brain; rather, intelligence should be defined in terms of specialized intelligence — adaptations that enable particular species to succeed in their particular ecological niches. Support for this view is that the ability to learn (which is surely a large part of intelligence) seems to be specialized. Many animals can learn a particular task with ease but find it impossible to learn a logically equivalent task. It appears that an animal's learning ability depends upon the hard-wired behaviors already present in its brain. In this view, all animals are differently intelligent. It simply makes no sense to ask whether a bonobo is brighter than a homing pigeon: both creatures possess specialized intelligence enabling them to succeed in their particular environments.

These two seemingly opposite views of intelligence — that either general intelligence or specialized intelligence is the important factor — are perhaps merely two faces of the same coin. The lesson is that, cognitively, animals are both similar and yet different. In the case of mankind, much as we might like to think otherwise, our similarities with other animals are clear: we are simply not much better than many other animals at tasks that investigate fundamental, non-verbal cognition.

Nevertheless, it is impossible to deny the profound difference that exists between mankind and every other species. We may not be atop some evolutionary ladder of intelligence, but we are the only species capable of constructing abstract systems of thought. Only a member of our species can reflect upon his own thoughts and the thoughts of others. Only Homo sapiens is in the slightest bit interested in defining intelligence or wondering precisely what it means. Indeed, with an appropriate definition, one can quite reasonably exclude all other species and say that mankind alone is intelligent.

0 0

Post a comment