— 
D.—ZOOLOGY 97 
experiments that the monkey had learned to respond not to the red circle 
by itself but to it in combination with the blue triangle—that is, to the 
correct member of a complex comprising these two sensory data. 
The whole trend of modern work on the perceptions of animals is to 
show that they do not normally respond to simple physico-chemical 
stimuli, but to more or less complex whole-situations, and if to parts of 
the whole-situation, then to these parts in their relation to the whole. 
_ This is the essence of the principle of Gestalt—response to elements in 
the perceptual field as parts of the pattern of the whole. The principle 
of the whole is thus valid for the perceptual field just as it is for executive 
behaviour. 
These few examples of modern work on the perceptions of animals 
emphasise the need for extreme care in establishing exactly what it is in 
the surrounding world to which animals respond. We must not assume 
a priori that behaviour is determined by a concatenation of simple physico- 
chemical stimuli ; we must drop all metaphysical theory and try to find 
out by careful experiment just what animals do respond to. We shall 
often find that they respond to images or patterns, or to classes of objects 
that have for the animal the same functional significance, or to bare 
relations. 
Response to relations is clearly demonstrated in some very thorough 
work recently carried out by Kliiver!°on the perceptual world of monkeys. 
His general method was to train his monkey to draw in one of a pair of 
boxes differing in some particular, for instance in weight. When he had 
established a positive response to the heavier of a pair, he varied the 
difference between the boxes, using two others of quite different weights 
from the original pair. He found by this method that his monkeys 
would respond to the bare relation ‘ heavier than,’ quite irrespective of 
the absolute weight of the boxes used, provided of course that they were 
not too heavy for the monkeys to move. Other experiments of the same 
type showed that the monkeys had a power of practical generalisation, 
that many objects differing in shape and colour yet produced the same 
response—they were, from the monkey’s point of view, functionally 
equivalent. This method of studying the equivalence or non-equivalence 
of perceptual objects promises to be a very fruitful one for investigating 
the behaviour of animals. 
In the short compass of this address I have been unable to give more than 
the very slightest sketch of a method for the study of animal behaviour 
which is, I think, likely to be the method of the future. It is, I maintain, 
a perfectly objective method, dealing with observable fact, and it is free 
from any metaphysical preconceptions. 
I have been concerned to point out two things. One is that it is time 
biology shook itself free from the limitations imposed upon it by a blind 
trust in the classical doctrine of materialism. This doctrine is not in 
harmony with the modern development of philosophical thought, nor 
with the modern development of physical science, and it is not well 
_ adapted to the study of living things. 
% 10 H. Kliiver, Behavior Mechanisms in Monkeys, Chicago, 1933. 
yj 
¥ E 
j 
' 
