Psychophysical detection testing with Bayesian active learning
Jacob R. Gardner, Xinyu D. Song, Kilian Q. Weinberger, Dennis L. Barbour, John P. Cunningham · 2015
Psychophysical detection tests are ubiquitous in the study of human sensation and the diagno-sis and treatment of virtually all sensory im-pairments. In many of these settings, the goal is to recover, from a series of binary observa-tions from a human subject, the latent function that describes the discriminability of a sensory stimulus over some relevant domain. The audi-tory detection test, for example, seeks to under-stand a subject’s likelihood of hearing sounds as a function of frequency and amplitude. Conven-tional methods for performing these tests involve testing stimuli on a pre-determined grid. This approach not only samples at very uninforma-tive locations, but also fails to learn critical fea-tures of a subject’s latent discriminability func-tion. Here we advance active learning with Gaus-sian processes to the setting of psychophysical testing. We develop a model that incorporates strong prior knowledge about the class of stimuli, we derive a sensible method for choosing sample points, and we demonstrate how to evaluate this model efficiently. Finally, we develop a novel likelihood that enables testing of multiple stim-uli simultaneously. We evaluate our method in both simulated and real auditory detection tests, demonstrating the merit of our approach. 1