414 MILDRED WEST LORING 



3. Failure of the subject to speak into the voice key so as to 

 make the clock stop. 



4. A few cases where 3 occurred for the experimenter. 



These results corroborate the conclusions drawn from the 

 averages. It will be noticed that to a great extent the modes 

 bear out the same conclusions as do the averages, especially for 

 the increase in reaction time for nouns over adjectives. 



The relatively large number of reaction times of four seconds 

 and over, for some of the subjects, must indicate either that the 

 stimulus words need further elimination or that these subjects 

 had particularly slow reaction times, therefore bringing a larger 

 percentage of reactions into this four second group. Both con- 

 clusions are true to some extent. Some further elimination is 

 still necessary to accommodate the lists to the vocabularies of 

 even university freshmen. On the other hand subjects VII, 

 VIII, and XI were naturally slow in reaction no matter how 

 simple the stimulus word. 



Conclusions for experiment II 



1. The reaction tune for the noun-adjective association is 

 longer than for the adjective-noun association. 



2. The reaction time for both adjective and noun stimulus 

 words increases directly with the number of syllables in the 

 stimulus word. 



3. The above conclusions agree with those made in experiment 

 I. They hold also for a group of difficult adjectives and nouns 

 such as were used in the practise period of this experiment, 

 although the reaction tunes in this case are greatly increased 

 above those obtained on the large mixed group used in experiment 

 I or the selected group of this experiment. 



4. Distribution curves plotted from the results for adjective 

 and noun stimulus words indicate that the modes for the reaction 

 times follow in general the laws of the averages. 



