Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

QUEST+: A general multidimensional Bayesian adaptive psychometric method.

Andrew B Watson1

  • 1New York University Abu Dhabi, Abu Dhabi, United Arab EmiratesNASA Ames Research Center, Moffett Field, CA, USAabwatson@me.com.

Journal of Vision
|March 30, 2017
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Letter identification and the neural image classifier.

Journal of vision·2015
Same author

Computing human optical point spread functions.

Journal of vision·2015
Same author

A formula for human retinal ganglion cell receptive field density as a function of visual field location.

Journal of vision·2014
Same author

A formula for the mean human optical modulation transfer function as a function of pupil size.

Journal of vision·2013
Same author

Modeling acuity for optotypes varying in complexity.

Journal of vision·2012
Same author

A unified formula for light-adapted pupil size.

Journal of vision·2012
Same journal

Analysis of human visual experience data.

Journal of vision·2026
Same journal

Pyramid-based Bayesian modeling for high-resolution behavioral analysis.

Journal of vision·2026
Same journal

Sensation without perception: The white whale effect and perceptual blindness in autonomous vehicles.

Journal of vision·2026
Same journal

Gaze behavior during closed-captioned movie viewing adapts to absent audio through more frequent switching between text and scene.

Journal of vision·2026
Same journal

In pursuit of saccade awareness: Limited volitional control and minimal conscious access to catch-up saccades during smooth pursuit eye movements.

Journal of vision·2026
Same journal

Dissociable effects of element-lifetime and stimulus-duration on local and global motion processing: An equivalent noise study.

Journal of vision·2026
See all related articles

QUEST+ is a flexible, computer-based tool that speeds up psychological and perceptual experiments by intelligently choosing the most informative test questions for each participant in real time.

Area of Science:

  • Psychophysics research within QUEST+ adaptive testing methodology
  • Computational cognitive science and mathematical psychology

Background:

Researchers often struggle to efficiently estimate complex psychological parameters using traditional, fixed-sequence testing protocols. This limitation creates a significant bottleneck in data collection for multidimensional perceptual studies. Prior research has shown that simple adaptive procedures can improve efficiency for one-dimensional tasks. However, these older methods lack the flexibility required for modern, multifaceted experimental designs. No prior work had resolved the need for a unified framework capable of handling arbitrary stimulus dimensions. That uncertainty drove the development of more robust, Bayesian-based computational approaches. This paper introduces a generalized framework to address these persistent challenges in psychometric testing. The current study builds upon established principles to provide a versatile solution for diverse research applications.

Purpose Of The Study:

The aim of this study is to introduce a generalized, multidimensional Bayesian adaptive psychometric method for diverse experimental applications. This work addresses the limitations of existing, narrow-scope testing procedures that cannot handle complex, multifaceted designs. The researchers seek to provide a unified framework that accommodates an arbitrary number of stimulus dimensions. They also intend to support various psychometric function parameters, including slope and lapse rates. This gap motivated the development of a tool that integrates multiple historical advancements into one procedure. The authors propose that their method will significantly accelerate data collection across cognitive and perceptual science. By offering a single, flexible solution, they aim to simplify the implementation of complex research protocols. The study establishes the theoretical and practical basis for this versatile, high-performance testing approach.

Keywords:
adaptive testingpsychometric functionBayesian inferencestimulus dimensions

Frequently Asked Questions

The researchers propose that this Bayesian framework optimizes efficiency by selecting the most informative stimuli based on previous trial outcomes. Unlike static protocols, this method dynamically updates parameter estimates to minimize uncertainty throughout the testing session.

This tool utilizes a multidimensional Bayesian approach to manage arbitrary stimulus dimensions and psychometric function parameters. It functions as a generalized extension of original adaptive testing procedures, allowing for the integration of diverse experimental designs within one unified software environment.

A Bayesian framework is necessary to handle the complex, multidimensional probability distributions inherent in these experiments. This mathematical structure allows the system to update its internal model of participant performance after every individual response.

Related Experiment Videos

Main Methods:

Review approach involves evaluating a generalized computational framework for multidimensional stimulus testing. The authors describe a flexible architecture capable of managing arbitrary parameters and diverse trial outcomes. This design integrates several historical developments into one unified, coherent procedure. The researchers utilize a Bayesian statistical engine to update participant models in real time. Their approach supports various experimental configurations, including pair comparisons and categorical ratings. The system architecture accommodates both linear and circular stimulus dimensions for broad applicability. By consolidating multiple testing strategies, the framework simplifies the implementation of complex psychometric tasks. This methodology provides a standardized toolset for researchers aiming to optimize their experimental efficiency.

Main Results:

Key findings from the literature indicate that this procedure successfully implements a wide variety of experimental designs. The authors report that the framework enables precise measurement of psychometric function parameters such as slope and lapse. Results show that the tool effectively estimates contrast sensitivity functions and increment threshold functions. The study confirms that the method supports Thurstone scale estimation using pair comparisons. Data collection is accelerated through the intelligent, adaptive selection of stimuli across multiple dimensions. The researchers demonstrate that the system handles both linear and circular stimulus inputs with equal efficacy. This unified procedure replaces the need for numerous specialized, narrow-scope testing protocols. The evidence highlights the versatility of this approach for diverse applications in perceptual science.

Conclusions:

The authors demonstrate that this Bayesian framework successfully accommodates a wide range of experimental designs. Synthesis and implications suggest that researchers can now implement complex threshold measurements with greater speed. This approach facilitates the estimation of multiple psychometric parameters simultaneously within a single procedure. The evidence indicates that the method effectively handles both linear and circular stimulus dimensions. By integrating various trial outcomes, the tool enhances the precision of contrast sensitivity and noise-masking assessments. The findings imply that this generalized procedure serves as a powerful alternative to specialized, narrow-scope testing techniques. Future applications may benefit from the increased flexibility provided by this multidimensional adaptive strategy. This work establishes a robust foundation for accelerating data collection across cognitive and perceptual science domains.

The procedure incorporates various trial outcomes to estimate parameters like slope and lapse rates. This data type allows the model to refine its predictions about participant sensitivity across different experimental conditions.

The method measures increment threshold functions and contrast sensitivity functions with high precision. These phenomena are captured by the model as it adjusts stimulus intensity based on the participant's ongoing performance.

The authors propose that their generalized method accelerates data collection across many areas of cognitive science. They claim this approach provides a single, flexible solution for diverse experimental designs that previously required separate, specialized procedures.