In this paper, we describe the development of the Wisconsin Card Sorting Test (WCST). We trace the history of sorting tasks from the studies of Narziss Ach on the psychology of thinking, via the work of Kurt Goldstein and Adhemar Gelb on brain lesioned patients around 1920 and subsequent developments, up to the actual design of the WCST by Harry Harlow, David Grant, and their student Esther Berg. The WCST thus seems to originate from the psychology of thinking ("Denkpsychologie"), but the test, as it is used in clinical neuropsychological practice, was designed by experimenters working within the behaviorist tradition. We also note recent developments suggesting that, contrary to the general impression, implicit learning may play a role in WCST-like discrimination learning tasks.