RELATIVE VALUES OF REWARD AND PUNISHMENT 

 IN HABIT FORMATION 



J. D. DODSON 



From the Psychological Laboratory of the University of Minnesota 



CONTENTS 



History of problem 231 



Present problem 237 



Preliminary experiment 241 



Conclusions 247 



Experiment proper 247 



Curves of learning 249 



Summary of tables 256 



Differences in learning in males and females 262 



Interpretation of curves of relative values of hunger 262 



Values of different strengths of electrical stimuli 266 



Comparison with earlier results 268 



Summary of facts to be explained 268 



Some laws of learning which have been suggested 269 



Relation of rate of learning to retention 274 



Retraining 275 



Conclusions 275 



HISTORY OF THE PROBLEM 



For investigations of animal behavior under laboratory con- 

 ditions, the first need is some form of stimulation which will 

 serve as a motive for the proper performance of the required 

 act. The forms of stimuli most commonly used are rewards 

 and punishments. But there are differences of opinion among 

 students of comparative behavior as to which of these two motives 

 facilitate most the learning process. At the suggestion of Prof. 

 R. M. Yerkes the writer has undertaken to answer, for at least a 

 sensory habit in one species of animals, this important question. 



The use of reward and punishment as motives for securing 

 desired forms of behavior is as old as the companionship of 

 human and infra-human organisms. Wherever man has at- 



231 



