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

Fitting distributions using maximum likelihood: methods and packages.

Denis Cousineau1, Scott Brown, Andrew Heathcote

  • 1Université de Montréal, Montréal, Québec, Canada. denis.cousineau@umontreal.ca

Behavior Research Methods, Instruments, & Computers : a Journal of the Psychonomic Society, Inc
|January 12, 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

Error Cancellation During Early Task Performance.

Experimental psychology·2026
Same author

Baroreflex Activation Therapy In Heart Failure-the Barostim-Enabled Neurohormonal Intervention for Improving Treatment of Heart Failure (BENEFIT-HF) Trial Rationale and Design.

Journal of cardiac failure·2026
Same author

Positive Bias in Value-Based Decision Making: Neurocognitive Associations with Resilience.

The Journal of neuroscience : the official journal of the Society for Neuroscience·2026
Same author

The Creative Space Theory as a map to explore the mind.

Possibility studies & society·2026
Same author

Assessment of student confidence, knowledge, and attitudes after use of interactive computer modules in a university-level cardiovascular physiology course.

Advances in physiology education·2026
Same author

An illustrative guide to expressing cognitive theories using evidence accumulation modelling.

Behavior research methods·2026
Same journal

Real-time language comprehension research using the Apple-Psych system.

Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc·2014
Same journal

Constraints on the perception of synthetic speech generated by rule.

Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc·2014
Same journal

Perception of synthetic speech produced automatically by rule: Intelligibility of eight text-to-speech systems.

Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc·2012
Same journal

Ninety-three pictures and 108 questions for the elicitation of homophones.

Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc·2011
Same journal

Free-association norms for the Spanish names of the Snodgrass and Vanderwart pictures.

Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc·2005
Same journal

French normative data and naming times for action pictures.

Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc·2005
See all related articles

Quantile maximum probability (QMP) offers an efficient and unbiased method for fitting response time (RT) distributions. This approach, when applied to common RT models, provides reliable parameter estimates for cognitive modeling.

Area of Science:

  • Cognitive Psychology
  • Psychometrics
  • Computational Neuroscience

Background:

  • Response time (RT) models are crucial for understanding cognitive processes.
  • Analyzing the entire RT distribution, not just means and variances, provides more powerful tests.
  • Nonparametric methods are ideal but often difficult to implement.

Purpose of the Study:

  • To review and compare distribution fitting methods for RT data.
  • To evaluate the quantile maximum probability (QMP) method against maximum likelihood (ML).
  • To assess the impact of different distribution functions on parameter recovery.

Main Methods:

  • Distribution fitting using maximum likelihood (ML) and quantile maximum probability (QMP).
  • Evaluation of QMP with common RT distribution functions: ex-Gaussian, Gumbel, lognormal, Wald, and Weibull.

Related Experiment Videos

  • Comparison of ML software packages: PASTIS, QMPE, DISFIT, and MATHEMATICA.

    • Main Results:

    • QMP demonstrates low bias and good efficiency across various distribution functions.
    • The choice of distribution function significantly impacts parameter recovery and efficiency.
    • Lognormal and Wald distributions showed non-linear parameter dependencies, increasing bias and reducing efficiency.

    Conclusions:

    • QMP is a viable and efficient method for fitting RT distributions.
    • Careful selection of distribution functions is critical for accurate cognitive modeling.
    • Guidelines are provided for linking descriptive RT models to cognitive theories.