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

Combined permutation test and mixed-effect model for group average analysis in fMRI.

Sébastien Mériaux1, Alexis Roche, Ghislaine Dehaene-Lambertz

  • 1CEA, Service Hospitalier Frédéric Joliot, Orsay, France.

Human Brain Mapping
|April 6, 2006
PubMed
Summary

This study introduces a new statistical test for group analyses that handles varying individual uncertainties. The proposed likelihood ratio test offers greater sensitivity than traditional methods, especially in neuroimaging research.

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

When Can Brain Connectivity Track the Working Mind? A Large-Scale Benchmark of Dynamic Functional Connectivity Across Cognitive Paradigms.

bioRxiv : the preprint server for biology·2026
Same author

Initial organization and progressive expansion of the math-responsive brain network during the first school years.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same author

Replicability of multivariate brain-behaviour associations depends on clinical profile.

Communications biology·2026
Same author

ABCD-ReproNim: An educational program for responsible and reproducible analyses of ABCD data.

Developmental cognitive neuroscience·2026
Same author

Impact of Education and Music Training on the Development of Abstract Thinking in the First Years of Schooling.

Open mind : discoveries in cognitive science·2026
Same author

Open neuroinformatics infrastructure ecosystem for federated multisite studies.

bioRxiv : the preprint server for biology·2026

Area of Science:

  • Neuroimaging analysis
  • Statistical modeling
  • Brain imaging data analysis

Background:

  • Classical one-sample t-tests assume equal variances, which is often violated in group analyses.
  • Heterogeneous within-subject uncertainties in estimated effects can reduce statistical power.
  • Existing methods may lack sensitivity when variances are unequal.

Purpose of the Study:

  • To generalize the one-sample t-test for group average analyses with heterogeneous within-subject uncertainties.
  • To develop a more sensitive statistical test than standard t-tests and permutation-based versions.
  • To provide exact specificity control using a sign permutation framework.

Main Methods:

  • Developed a maximum likelihood ratio test statistic based on a Gaussian mixed-effect model.

Related Experiment Videos

  • Calibrated significance levels using a sign permutation framework (similar to Holmes et al.).
  • The method does not assume homoscedasticity (equal variances).

    • Main Results:

    • The proposed likelihood ratio test demonstrated potentially greater sensitivity compared to standard t-tests.
    • Results from the Functional Imaging Analysis Contest 2005 dataset support the test's effectiveness.
    • The test maintains exact specificity control under mild symmetry assumptions.

    Conclusions:

    • The generalized t-test effectively accounts for heterogeneous within-subject uncertainties in group analyses.
    • This new method offers improved sensitivity for neuroimaging and other group studies.
    • The likelihood ratio test provides a robust alternative when assumptions of classical tests are not met.