For over four decades the Wisconsin Card Sorting Test (WCST) has been one of the most distinctive tests of prefrontal function. Clinical research and recent brain imaging have brought into question the validity and specificity of this test as a marker of frontal dysfunction. Clinical studies with neurological patients have confirmed that, in its traditional form, the WCST fails to discriminate between frontal and non-frontal lesions. In addition, functional brain imaging studies show rapid and widespread activation across frontal and non-frontal brain regions during WCST performance. These studies suggest that the concept of an anatomically "pure" test of prefrontal function is not only empirically unattainable, but also theoretically inaccurate. The aim of the present review is to examine the causes of these criticisms and to resolve them by incorporating new methodological and conceptual advances in order to improve the construct validity of WCST scores and their relationship to prefrontal executive functions. We conclude that these objectives can be achieved by drawing on theory-guided experimental design, and on precise spatial and temporal sampling of brain activity, and then exemplify this using an integrative model of prefrontal function [i.e., Miller, E. K. (2000). "The prefrontal cortex and cognitive control." "Nature Reviews Neuroscience," 1, 59-65.] combined with the formal information theoretical approach to cognitive control [Koechlin, E., & Summerfield, C. (2007). "An information theoretical approach to prefrontal executive function." "Trends in Cognitive Sciences," 11, 229-235.]. (Contains 2 tables and 4 figures.)