PERSEVERANCE IN REACTIONS PRIMATES AND RODENTS 37 



attention until I read Watson's (7) contention that advantage 

 and disadvantage as such have nothing to do with habit forma- 

 tion. It was then too late to use most of my subjects for an 

 experimental investigation of the problem that is dealt with in 

 subsequent pages. 



The repetitive tendency that finds expression in reactions that 

 I have classified as belonging to types " D " and " E " may 

 play an important role in habit determination. If one may 

 safely generalize to the effect that what an organism has done 

 most frequently in past encounters with a situation it is most 

 apt to do, all things else being equal, in reencounters with that 

 situation it necessarily follows that whatever leads to the repe- 

 tition of a given activity before a definite habit is formed must 

 be regarded as a possible determinant of the ultimately estab- 

 lished habit. We have found that various factors which are 

 subject to but limited and imperfect control may operate at 

 any time to precipitate persistent repetition of an unsuccess- 

 fully directed activity, even though no past experience and no 

 constant reactive value of the situation could be regarded as 

 directly accounting for such repetition. Inwardly arising adven- 

 titious stimuli or unpreventable causes of excitement or dis- 

 traction that operate from without may thus account for the 

 relative frequency of a given activity, and this frequency may in 

 turn act as a factor in determining the composition of the ulti- 

 mately established habit. Reference to the calculations for 

 White Rat 5 (appendix) affords the following example: 



At the beginning of White Rat 5's twentieth trial she had 

 entered alley No. 1 10 times, alley No. 2 16 times, alley No. 3 

 22 times and alley No. 4 18 times. During the twentieth trial 

 a quite incalculable factor precipitated a Type E reaction, so 

 that she entered alleys No. 1 and No. 2 7 times each. At the 

 end of her twenty-fifth trial she had entered alley No. 2 ?>2 

 times and alley No. 3 only 31 times. At the completion of her 

 series of 100 trials alley No. 2 had been entered 161 times, whilst 

 alley No. 3 had been entered only 133 times and alley No. 4 

 only 56 times. In other words, during four preliminary trials 

 (these necessarily enter into the calculations for table 8) and 

 19 formal trials this subject had shown a preference for alley 

 No. 3 over all other alleys, and a preference for alley No. 4 

 over alley No. 2; but an episode of repetitive behavior which 



