290 



Why Living Things Behave As They Do unit v 



Fig. 272 Humafis learn by trial and error. This youngster probably havnnered his 

 fingers in the beginning, but with repetitiojt, such wasted ynotions become less fre- 

 quent. Why is this not a conditioned response? (38TH annual report, supt. of schools, 



NEW YORK city) 



these. Learning to open a door is an- 

 other. Learning to dress yourself, to 

 skate, and to ride a bicycle became prob- 

 lems at various stages of your growth. 

 There were countless responses for you 

 to learn. Most of them were not condi- 

 tioned responses built on our inborn re- 

 flexes. How did you learn them? Again 

 let us turn to animal experimentation to 

 find the answer. In the less complex an- 

 imals behavior is less complex and for 

 that reason more easily studied. 



A fish learns by "trial and error." In one 

 experiment some shade-loving fish were 

 kept in an aquarium which was half in 

 the sun and half in the shade. They usu- 

 ally gathered at the darkened end, espe- 

 cially since this was the end at which 

 they were fed. One day they were 

 gently forced to the sunny end of the 

 tank and kept there by a glass partition 



placed across the aquarium. This parti- 

 tion contained a small hole just large 

 enough for a fish to pass through. As 

 \'ou would expect, the fish swam back 

 and forth along the glass until, by acci- 

 dent, one struck the opening; it darted 

 through. The fish was quickly rewarded 

 by getting food and by being in the 

 shade. Up to this point the behavior of 

 the fish was reflex behavior; it succeeded 

 because of good luck. But when the ex- 

 periment was repeated manv times with 

 this same fish it srot to the hole with 

 fewer and fewer wasted motions and 

 therefore in a shorter time. The fish 

 modified its behavior; it learned. It 

 learned after many repetitions in the 

 course of what is sometimes called "trial 

 and error." 



Other examples of trial and error. Many 

 similar experiments in learning have been 



