Kant’s denial that psychology is a properly so-called natural science, owing to the lack of application of mathematics to inner sense, has garnered a great deal of attention from scholars. Although the interpretations of this claim are diverse, commentators by and large fail to ground their views on an account of Kant’s conception of applied mathematics. In this article, I develop such an account, according to which the application of mathematics to a natural science requires both a mathematical representation and (...) a metaphysical validation for the positive use of this representation to achieve a priori knowledge about nature. The second condition—that of metaphysical validation—has been overlooked in the literature. I show that psychology’s falling short of natural scientific propriety consists not in our lacking sufficient mathematical tools for the representation of inner states, according to Kant. After all, we can represent the mere temporality of inner states with the line and their intensities with numbers. Rather, the problem is that metaphysics does not validate the further use of such mathematical entities for the achievement of a priori knowledge about inner phenomena. (shrink)