The forgotten history of signal detection theory.

Abstract

Signal detection theory is one of psychology's most well-known and influential theoretical frameworks. However, the conceptual hurdles that had to be overcome before the theory could finally emerge in its modern form in the early 1950s seem to have been largely forgotten. Here, I trace the origins of signal detection theory, beginning with Fechner's (1860/1966) Elements of Psychophysics. Over and above the Gaussian-based mathematical framework conceived by Fechner in 1860, nearly a century would pass before psychophysicists finally realized in 1953 that the distribution of sensations generated by neural noise falls above, not below, the threshold of conscious awareness. An extensive body of single-unit recording and neuroimaging research conducted since then supports the idea that sensory noise yields genuinely felt conscious sensations even in the complete absence of stimulation. That hard-to-come-by insight in 1953 led immediately to the notion of a movable decision criterion and to the methodology of receiver operating characteristic (ROC) analysis. Over the ensuing years, signal detection theory and ROC analysis have had an enormous impact on basic and applied science alike. Yet, in some quarters of our field, that fact appears to be virtually unknown. By tracing both its fascinating origins and its phenomenal impact, I hope to illustrate why no area of experimental psychology should ever be oblivious to signal detection theory. (PsycINFO Database Record (c) 2020 APA, all rights reserved).