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

Stopping rules in Bayesian adaptive threshold estimation.

Rocío Alcalá-Quintana1, Miguel A García-Pérez

  • 1Departamento de Metodología, Facultad de Psicología, Universidad Complutense, Campus de Somosaguas, 28223 Madrid, Spain. ralcala@psi.ucm.es

Spatial Vision
|August 3, 2005
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

Psychometric functions for temporal discrimination: Duration or log duration?

Behavior research methods·2026
Same author

Goodness-of-fit Tests for Categorical Models of Psychological Processes: Fixing the Occasional Failures of Asymptotic Theory.

The Spanish journal of psychology·2025
Same author

Are alternative variables in a set differently associated with a target variable? Statistical tests and practical advice for dealing with dependent correlations.

The British journal of mathematical and statistical psychology·2024
Same author

Are the Steps on Likert Scales Equidistant? Responses on Visual Analog Scales Allow Estimating Their Distances.

Educational and psychological measurement·2024
Same author

Both reduced ovarian reserve and severe semen alterations are overrepresented in couples seeking assisted reproductive technology treatment for the first time: a cross-sectional study.

Reproductive health·2023
Same author

Accuracy and precision of responses to visual analog scales: Inter- and intra-individual variability.

Behavior research methods·2022

Sequential procedures for threshold estimation do not outperform fixed-length methods in yes-no tasks. In two-alternative forced-choice (2AFC) tasks, sequential methods offer minimal benefits, and their indices lack diagnostic value for accurate estimation.

Area of Science:

  • Psychophysics
  • Statistical modeling
  • Computational neuroscience

Background:

  • Sequential procedures in threshold estimation rely on dynamic stopping rules.
  • The diagnostic value of indices for accurate estimation is often assumed.
  • Bayesian sequential procedures offer an alternative to fixed-length methods.

Purpose of the Study:

  • To compare the performance of five Bayesian sequential procedures against fixed-length procedures.
  • To evaluate indices for use in dynamic stopping rules for threshold estimation.
  • To assess the impact of psychometric function characteristics on sequential procedure efficacy.

Main Methods:

  • A simulation study was conducted comparing Bayesian sequential and fixed-length procedures.
  • Five indices were evaluated: probability interval width, posterior standard deviation, absolute change, average change, and sign fluctuations.

Related Experiment Videos

  • The study considered both yes-no and two-alternative forced-choice (2AFC) tasks, varying psychometric function parameters.

    • Main Results:

    • Sequential procedures did not outperform fixed-length procedures in yes-no tasks.
    • In 2AFC tasks, sequential procedures showed minimal improvement over fixed-length methods, primarily with short, undependable runs.
    • The differences in performance diminished when procedures reached a sufficient number of trials (approx. 70) for dependability.
    • None of the evaluated indices demonstrated sufficient diagnostic value to justify their use in dynamic stopping rules.

    Conclusions:

    • The diagnostic value of common indices for dynamic stopping rules in sequential threshold estimation is questionable.
    • Sequential procedures, particularly those not based on sign fluctuations, do not consistently outperform fixed-length procedures.
    • Implementation challenges further limit the utility of sequential procedures as alternatives to fixed-length methods.