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

Comparing methods of analyzing fMRI statistical parametric maps.

Jonathan Marchini1, Anne Presanis

  • 1Department of Statistics, University of Oxford, Oxford OX1 3TG, UK. marchini@stats.ox.ac.uk

Neuroimage
|June 29, 2004
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

Rare protein-coding variation and the genetic architecture of height in >1.4 million individuals.

medRxiv : the preprint server for health sciences·2026
Same author

Population-scale repeat expansions elucidate disease risk and brain atrophy.

Nature·2026
Same author

Humans with function-disrupting variants in the myostatin gene (MSTN) have increased skeletal muscle mass and strength, and less adiposity.

Nature communications·2026
Same author

Rare coding variants in CHRNB3 associate with reduced daily cigarette smoking across ancestries.

Nature communications·2026
Same author

Computationally efficient meta-analysis of gene-based tests using summary statistics in large-scale genetic studies.

Nature genetics·2025
Same author

Variant Classification Using Proteomics-Informed Large Language Models Increases Power of Rare Variant Association Studies and Enhances Target Discovery.

Genetic epidemiology·2025
Same journal

Brain-Inspired Large Model Mindreading.

NeuroImage·2026
Same journal

Light on Broken Networks: Resting-State fNIRS as a Tool for Connectivity Mapping.

NeuroImage·2026
Same journal

Criticism-Evoked Rumination Is Linked to Dynamic adjustments of the Left Superficial Amygdala in Adolescents.

NeuroImage·2026
Same journal

GeNED.ar cohort: Neuroimaging Resource for Aging Studies in an Admixed Population from Argentina.

NeuroImage·2026
Same journal

DTI-ALPS index correlates with poor neuromodulation outcomes of bilateral STN-DBS in Parkinson's disease patients: a prospective cohort study.

NeuroImage·2026
Same journal

Decoding neuronal criticality firing patterns for large brain based EEG models.

NeuroImage·2026
See all related articles

Comparing statistical parametric map (SPM) analysis methods, posterior probability thresholding offers the most power for detecting activated regions. Type I error control provides the strictest control over false positives in neuroimaging analyses.

Area of Science:

  • Neuroimaging analysis
  • Statistical modeling
  • Brain activity detection

Background:

  • Statistical parametric maps (SPMs) are crucial for analyzing neuroimaging data.
  • Delineating activated brain regions involves various statistical thresholding approaches.
  • Understanding the performance characteristics of these methods is essential for accurate interpretation.

Purpose of the Study:

  • To compare the performance of three main thresholding approaches for SPM analysis: type I error control, false discovery rate (FDR) control, and posterior probability thresholding.
  • To evaluate these methods under realistic simulation conditions mirroring real datasets.
  • To determine the optimal approach for delineating activated brain regions.

Main Methods:

  • A simulation study was conducted to compare different statistical thresholding methods.

Related Experiment Videos

  • The study utilized default settings for each approach to mimic typical usage.
  • Performance was evaluated based on statistical power and type I error rates.

    • Main Results:

    • Posterior probability thresholding demonstrated the highest statistical power.
    • Type I error control thresholding yielded the lowest type I error rates.
    • False discovery rate (FDR) control thresholding offered intermediate performance, with potential to approximate posterior probability thresholding under specific parameters.

    Conclusions:

    • Viewing activation delineation as a classification problem provides a valuable framework for comparing methods.
    • The choice of loss function is critical, as it dictates the penalty for different types of errors.
    • Posterior probability thresholding is recommended for maximizing detection power, while type I error control is best for minimizing false positives.