Behaviour having consequences that are judged to be intelligent. The consequences of behaviour are due both to the behaviour itself and the environment that it influences. When these are judged in relation to some criteria of intelligence, both the behaviour and its appropriateness in the environment in which it is observed should be taken into account. Various criteria for judging behaviour in terms of intelligence have been used. Normally, the criteria relate to what we think about human intelligence. In other words, we judge animal behaviour to be intelligent if it is similar to human intelligent behaviour. This is not good scientific practice. For one thing, it is not possible to devise a fair intelligence test for an animal. Not only are some species good at some types of problem solving, and other species other types, but the performance of animals depends very much upon the testing apparatus. For example, rats (Rattus norvegicus) perform much better at problems involving visual discrimination, if they are required to jump at the symbol of their choice (e.g. in a Lashley jumping stand), instead of running at it (e.g. in a Y maze). It is not possible to find the best testing apparatus for a given species, because nor all possibilities can be tried. It is not fair to compare animals tested in different circumstances, nor is it fair to test animals of different species in the same apparatus.
Other criteria for judging the intelligence of the consequences of behaviour include economic criteria (i.e. in terms of the animal's own energy economy), appropriateness or success in nature (i.e. in terms of fitness), and in terms of theoretical optimality criteria. In general, in attempting to define intelligent behaviour, what matters is the behavioural outcome, not the nature of the mechanism by which the outcome is achieved. This is particularly true in assessing the role of cognition in intelligent behaviour. Cognition provides a way of coping with unpredictable aspects of the environment, but we should ask what advantage an animal would gain from a cognitive solution to a problem. A tried and tested rule-of-thumb (a risk-averse solution) may be better than an essentially experimental (risk-prone) cognitive solution. Thus, in judging animal behaviour, we must distinguish between cognition (a possible means to an end) and intelligence (an assessment of performance in terms of some functional criteria).
Subjects: Zoology and Animal Sciences.