246 THE WONDER OF LIFE 



gradual descent. If the animals had ideas on the subject, 

 they did not seem to use them. They learned by ' trial 

 and error,' as we often do ourselves. But Professor Thorn- 

 dike made an important step in suggesting that the pleasure 

 of the meal that rewarded escape served to ' stamp 

 in ' the immediately antecedent association between the 

 picture of the interior of the cage and the successful impulse 

 that led to the succession of muscular movements effecting 

 release. This is Professor Thorndike's ' sense-impulse ' 

 theory of learning. 



When Thorndike's cats were shut up in boxes which 

 could be easily opened in a particular way, they seemed 

 to get out by accident. On subsequent occasions they 

 did not take quite so long, and they gradually learned the 

 trick. Dogs were quicker, and monkeys quicker still. 

 In most cases the method seems to be the same — a chance 

 discovery, and subsequently a gradual ehmination of the 

 ineffective attempts. But there appear to be some cases 

 where it looks as if the animal had an intuition of the Une 

 of efiective trial, as if it ' had a notion ' of the best thing 

 to do. 



Experiments, especially in getting out of labyrinths, have 

 been made with rats and guinea-pigs, chickens and sparrows, 

 and some other creatures. The story is in most cases 

 essentially the same. The animals learn more or less 

 quickly to profit by their mistakes and to conquer the 

 difficulties of the situation. In some cases (Watson's 

 white rats) the learning appears to depend in great part on 

 a muscular memory of the efEective sequence of movements, 

 for the elimination of sight, hearing and smell and a good 

 deal of tactihty did not seem to make much difference 

 to the education in the maze. 



