The science of life : fully illustrated in tone and line and including many diagrams, str. 814

BOOK 8

will attack only one special kind of prey ; insects that will starve to death on any plant but their usual habitat, not because it is unwholesome to them, but because they have only the instinct to eat one particular kind ; ants which have the instinct to make fungus-gardens; and so forth. Birds are similarly equipped, and for a long time, while great reptiles dominated the earth, this was the direction which vertebrate behaviour took as it elaborated and improved itself. Even such a close relative of man as the cat has instinctive reactions towards mice. Take a kitten a few weeks old, that has been brought up away from all other cats and fed solely on milk, and put it and a mouse together in a box; it will show all the signs of feline hunting excitement, growling and bristling, and will try to catch and kill its natural prey. But this sort of thing is less well seen in kittens than in birds, for mammals tried a unique mental experiment as they evolved ; they made for individual adaptability, and therefore cleared the mind as far as possible of fixed reactionsystems. In man it seems there is scarcely anything of the sort. He starts with no such equipment of directive dispositions.

The new-born infant will suck, and it will grasp and hold on to a finger or stick ; and it will show signs of fear if support is suddenly taken away from it and it feels itself dropping. In addition to this it manifests a very distinct fear reaction at loud noises, but it has no other specific fears at all. No other avoidances, and no other impulses but this simple grasping and suction. Careful experiments show that babies have no instinctive fear of particular kinds or sizes of animals. A fiveand-a-half months infant was tested with a black cat, a pigeon, a rabbit, a white rat, and a large dog; the smaller creatures she tried to get hold of, the dog she looked at with unafraid interest. When taken to the Zoo a week or so later, she showed no fear when brought quite close to camels, bears, zebras, ostriches, nor even when a monkey made threatening gestures through the wires. Similar tests on a seven-and-ahalf months child gave the same result. In the same way, it can be shown that babies have no fear of flame. Nor is there any specific instinct for catching any particular kind of animal, as with the kitten.

The child has to learn what to avoid and what to be afraid of. It begins by a positive grasping and mouthing reflex to almost everything that is presented to it. It reaches out, grabs it, manipulates it, and brings it

788

THE SCIENCE OF LIFE

CHAPTER 7

up to its mouth. If these positive reactions lead to a painful result, a selective inhibition will be set up.

Sometimes it takes a long time to effect this. A five months baby, for instance, which was tested by Watson, reached for a lighted candle and followed it about with her eyes. If allowed to put her fingers in the flame, the hand was reflexly withdrawn ; and the withdrawal was quick enough to save the child from pain, for she did not cry. If the fingers almost but not quite touched the flame, they were bent, but the whole hand was not snatched away. On the first occasion, the baby was tested twentyfive times, the painful stimulus being usually enough to cause this finger-bending, and two or three times enough for the hand to be pulled away. One week, and again two weeks later, the experiment was repeated. On neither occasion was there any noticeable inhibition of the impulse to reach for the flame. On the fourth trial, a further fortnight later, the reflex had begun to be modified ; the baby, now six months old, sometimes checked her hand halfway to the candle, sometimes would not reach at all but sucked her fingers instead. But sometimes she still reacted as positively as at the beginning. Six weeks later, however, she only reached out once in the primitive unconditioned way; and after this the reaction became habitually negative.

In this series, about a hundred and fifty stimuli, slightly or moderately painful, were needed to inhibit the original seizing reflex, though doubtless if the experimenters had been willing to let the child burn itself more severely, fewer trials would have been needed. As we have already mentioned, the human infant is instinctively afraid of nothing but loud noises and loss of support. But by utilizing this innate unconditioned reflex, it can be made afraid of any object you like. Watson, for instance, showed a baby a furry toy animal, and at the same moment his assistant made a loud noise by hitting a steel bar out of the baby’s sight. After this had been repeated three or four times, the baby had acquired a conditioned reflex in the same way as did Pavlov’s dogs; on seeing the toy animal, it showed “ conditioned fear,” even though no noise was made, and from now on it was afraid of the animal. What was more, it did not differentiate the toy clearly from other similar objects; as with Pavylov’s dogs, any stimulus resembling the conditioned stimulus worked, and the baby was frightened of other toy animals of about the same size,