Abstract
Reward alone (for correct choices) proved to be markedly inferior to punishment alone (for errors), while a combination of the 2 led to the most rapid learning.