216 WALTER S. HUNTER 



In order to make sure that the animal got the full benefit of the 

 stimulus, the opening of the resonance box was directed down- 

 ward upon the apparatus at a height of 19 inches. The noises 

 were given just behind and slightly above the discrimination 

 box. Each stimulus was given from the time the rat entered the 

 box until the food was reached behind the base of the T. Pun- 

 ishment and reward were used with all of the rats. Because of 

 the noise, the inductorium was placed in a distant room of the 

 laboratory. 



The six rats of set one were required to associate a turning 

 to the right with the tone (c' 512 v. s.), one trial, and a turning to 

 the left with the noise (hand clapping), a second trial. These 

 stimuli were as nearly equal in intensity as the experimenter 

 could secure. The intensity of the tone was approximately 

 equal to that of a tone produced by dropping a solid rubber ball 

 of 102 gm. upon a rigidly held c' 512 tuning fork from a height 

 of 100 cm. This gives a good medium intensity. In view of 

 the rat's poor sensitivity to intensity differences in tone (see 

 below) , these measurements are only intended to give the reader 

 a fair notion of the intensity values employed. Five trials 

 daily were given. All six rats learned the discrimination in from 

 310 to 520 trials. All of the records show a gradual decrease in 

 the number of errors. Controls were instituted which indicated 

 that the auditory stimuli alone, and not extraneous cues from 

 the experimenter or cues from the order of presentation, » were 

 determining the reactions. Two series of tests were then used: 

 (1) The tone was withheld so that the rats received at one trial 

 a noise and at the next trial not tone, but the absence of tone. 



» When the rat is confronted with a difficult discrimination, only the greatest 

 care will prevent the " learning of the problem " on the basis of position factors. 

 These kinaesthetic cues are often of great complexity and are influenced by pun- 

 ishment. The subject will repay careful study. The following types of cases, of 

 which some have been previously described, were noted in the present tests: (1) 

 The rat may form a habit of going always to one side. Every day he will begin 

 in this wise: but after a series of punishments, he may change and go to the opposite 

 side for the rest of that day. (2) A rat may alternate between sides in the order 

 right-left. If this leads to severe punishment, he may at times reverse the alterna- 

 tion to a left-right order. (3) Rat No. 5 formed a habit of alternating after each 

 success only. He would go to the right and, if successful, would go to the left in 

 the next trial. Had he failed on the right, though, he would have continued to go 

 there until he succeeded. He would then have gone to the left where the same type 

 of performance would be again gone through. 



Behavior of this nature may give the appearance of true discrimination of the 

 stimuli presented by the experimenter. It can only be checked up by most carefully 

 chosen variations in the order of presentations. 



