The influence of increase and decrease of the amount of reward upon the rate of learning

Five multiple-choice experiments are reported. In one, for example, a Spanish word was followed by 5 English words, one of which was the equivalent of the Spanish word. The subject was to find the right word by successive choices and to fixate it. Four trials were given in each experiment. The relative influence of 0.1, 0.2, 0.4 and 0.8 cent for each right choice, given both immediately and at the close of the whole experiment, was studied. The word "Right" or "Wrong" was spoken by the experimenter after each choice. Increase of the reward up to 0.4 cent acted to increase learning, but increase to 0.8 cent seemed "to do as much or more harm (by excitement or otherwise) as it does good by arousing and maintaining attention to the work, satisfaction at success, etc." Learning is not proportional to the money reward. The giving of 0.1 cent per success at the time is superior to a vague promise of "a little money" for doing the whole experiment. The values of increasing and decreasing series of rewards do not differ greatly.