Psychophysical Testing with Bayesian Active Learning

Jacob Gardner Washington University in St. L, Xinyu Song Washington University in St. Louis, Kilian Weinberger Washington University in St. Louis, John Cunningham Columbia University, Dennis Barbour Washington University in St. Louis
Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence, PMLR R13:644-653, 2015.

Abstract

Psychophysical detection tests are ubiquitous in the study of human sensation and the diagnosis and treatment of virtually all sensory impairments. In many of these settings, the goal is to recover, from a series of binary observations from a human subject, the latent function that describes the discriminability of a sensory stimulus over some relevant domain. The auditory detection test, for example, seeks to understand a subject’s likelihood of hearing sounds as a function of frequency and amplitude. Conventional methods for performing these tests involve testing stimuli on a pre-determined grid. This approach not only samples at very uninformative locations, but also fails to learn critical features of a subject’s latent discriminability function. Here we advance active learning with Gaussian 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 stimuli simultaneously. We evaluate our method in both simulated and real auditory detection tests, demonstrating the merit of our approach.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR13-st-l15a, title = {Psychophysical Testing with {B}ayesian Active Learning}, author = {St. L, Jacob Gardner Washington University in and St. Louis, Xinyu Song Washington University in and St. Louis, Kilian Weinberger Washington University in and University, John Cunningham Columbia and St. Louis, Dennis Barbour Washington University in}, booktitle = {Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence}, pages = {644--653}, year = {2015}, editor = {Meila, Marina and Heskes, Tom}, volume = {R13}, series = {Proceedings of Machine Learning Research}, month = {12--16 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r13/main/assets/st-l15a/st-l15a.pdf}, url = {https://proceedings.mlr.press/r13/st-l15a.html}, abstract = {Psychophysical detection tests are ubiquitous in the study of human sensation and the diagnosis and treatment of virtually all sensory impairments. In many of these settings, the goal is to recover, from a series of binary observations from a human subject, the latent function that describes the discriminability of a sensory stimulus over some relevant domain. The auditory detection test, for example, seeks to understand a subject’s likelihood of hearing sounds as a function of frequency and amplitude. Conventional methods for performing these tests involve testing stimuli on a pre-determined grid. This approach not only samples at very uninformative locations, but also fails to learn critical features of a subject’s latent discriminability function. Here we advance active learning with Gaussian 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 stimuli simultaneously. We evaluate our method in both simulated and real auditory detection tests, demonstrating the merit of our approach.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Psychophysical Testing with Bayesian Active Learning %A Jacob Gardner Washington University in St. L %A Xinyu Song Washington University in St. Louis %A Kilian Weinberger Washington University in St. Louis %A John Cunningham Columbia University %A Dennis Barbour Washington University in St. Louis %B Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2015 %E Marina Meila %E Tom Heskes %F pmlr-vR13-st-l15a %I PMLR %P 644--653 %U https://proceedings.mlr.press/r13/st-l15a.html %V R13 %X Psychophysical detection tests are ubiquitous in the study of human sensation and the diagnosis and treatment of virtually all sensory impairments. In many of these settings, the goal is to recover, from a series of binary observations from a human subject, the latent function that describes the discriminability of a sensory stimulus over some relevant domain. The auditory detection test, for example, seeks to understand a subject’s likelihood of hearing sounds as a function of frequency and amplitude. Conventional methods for performing these tests involve testing stimuli on a pre-determined grid. This approach not only samples at very uninformative locations, but also fails to learn critical features of a subject’s latent discriminability function. Here we advance active learning with Gaussian 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 stimuli simultaneously. We evaluate our method in both simulated and real auditory detection tests, demonstrating the merit of our approach. %Z Reissued by PMLR on 04 October 2026.
APA
St. L, J.G.W.U.i., St. Louis, X.S.W.U.i., St. Louis, K.W.W.U.i., University, J.C.C. & St. Louis, D.B.W.U.i.. (2015). Psychophysical Testing with Bayesian Active Learning. Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R13:644-653 Available from https://proceedings.mlr.press/r13/st-l15a.html. Reissued by PMLR on 04 October 2026.

Related Material