A STUDY OF PERSEVERANCE REACTIONS IN 

 PRIMATES AND RODENTS 



G. V. H-\MILTON 



Montecito, California 



INTRODUCTION 



The present study is a continuation of a previously published 

 one (1), which was also devoted to an investigation of per- 

 severance (trial and error) reactions. The method that was 

 employed in both studies requires the subject to find his way 

 out of an enclosure from which there are four apparently pos- 

 sible exit places. During a series of 100 trials each of the four 

 exit places affords escape 25 times, and during any given trial 

 only one of these places can be used for this purpose. The 

 right exit place to try for escape from the enclosure varies from 

 trial to trial, i.e., it is useless, during any given trial, to seek exit 

 from the enclosure by way of the place that afforded escape 

 during the immediately preceding trial. While the experimen- 

 ter's order of varying the right place to try for exit contains 

 this one negatively directive feature, it contains no specifically 

 directive positive feature that has thus far been discovered by 

 any of the subjects. The appearance of the apparatus, as it 

 is seen from within by the subject, affords no clue whatsoever 

 as to whether a given exit place is a right or a wrong one to 

 try until actual trial is made. It follows, therefore, that this 

 method operates against the formation of a specifically adap- 

 tive habit, i.e., a habit which would involve no unproductive 

 quests of an avenue of escape from the apparatus. 



The relation of this method to Yerkes' (3) multiple choice 

 method requires more explicit definition than it has yet been 

 given. Yerkes' method, which is of later development than 

 mine, embodies some of its secondary features but has a quite 

 different intention. Both methods exclude all specifically direc- 

 tive sensory stimuli, and require the subject to vary the place 

 through which he escapes from the apparatus from trial to trial. 



