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

Validating cluster size inference: random field and permutation methods.

Satoru Hayasaka1, Thomas E Nichols

  • 1Department of Biostatistics, The University of Michigan, Ann Arbor, MI 48109, USA.

Neuroimage
|December 20, 2003
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

FEMA-Long: Modeling unstructured covariances for discovery of time-dependent effects in large-scale longitudinal datasets.

PLoS genetics·2026
Same author

Leveraging STRAW +10 criteria to evaluate menopause stage effects on sleep quality.

Climacteric : the journal of the International Menopause Society·2026
Same author

Widespread use of invalid statistical tests in biomedical machine learning.

bioRxiv : the preprint server for biology·2026
Same author

Scalable Bayesian Image-on-Scalar Regression for Population-Scale Neuroimaging Data Analysis.

Journal of the American Statistical Association·2026
Same author

Convergent and divergent brain-cognition development in early adolescence.

Nature communications·2026
Same author

The methodological foundations of lesion network mapping remain sound.

bioRxiv : the preprint server for biology·2026
Same journal

Investigating the Neural Origins of Ear-EEG: A Correlation Study Using Scalp EEG Source Reconstruction.

NeuroImage·2026
Same journal

Hysteresis effects in visual and auditory perception and the comparison of underlying neural mechanisms - an EEG study.

NeuroImage·2026
Same journal

Short-term audio-tactile training affects cortical auditory speech-envelope tracking for incongruent but not congruent stimuli.

NeuroImage·2026
Same journal

Dissociable Neurocognitive Mechanisms of State and Trait Anxiety in Working Memory: Threat-Induced Alterations in Decision Dynamics and Attenuation of Large-Scale Network Reconfiguration.

NeuroImage·2026
Same journal

Neuro-Ocular Amyloid Characterization in Alzheimer's Disease via Cross-Site PET-MRI and Hierarchical Cross-Attention Driven Multimodal Representation Learning.

NeuroImage·2026
Same journal

Whole-brain network dynamics underlying intolerance of uncertainty.

NeuroImage·2026
See all related articles

Random field (RF) theory cluster size tests in brain imaging are often conservative, especially with low smoothness or thresholds. Permutation tests offer a more reliable alternative across various settings for neuroimaging analysis.

Area of Science:

  • Neuroimaging analysis
  • Statistical methods in neuroscience
  • Brain image processing

Background:

  • Cluster size tests offer greater sensitivity than intensity-based tests for brain image analysis.
  • Random field (RF) theory is commonly used for cluster size tests, but its performance is not fully understood, particularly when RF assumptions are questionable.

Purpose of the Study:

  • To evaluate the performance of random field (RF) theory-based cluster size tests.
  • To compare RF methods against exact permutation tests under various neuroimaging analysis conditions.

Main Methods:

  • Conducted a simulation study of cluster size tests.
  • Varied parameters including image smoothness, statistical thresholds, and degrees of freedom.
  • Compared the performance of RF theory methods with permutation tests for Gaussian and t-images.

Related Experiment Videos

Main Results:

  • RF methods were generally conservative for Gaussian images, especially with low smoothness and low thresholds.
  • For t-images, RF tests were conservative at lower thresholds and required high thresholds and smoothness for adequate performance.
  • Permutation tests performed well across all tested settings, provided discreteness in cluster size was addressed.

Conclusions:

  • Random field theory cluster size tests can be unreliable under certain conditions in neuroimaging.
  • Permutation tests provide a more robust and accurate approach for cluster size analysis in brain imaging.
  • Specific recommendations are provided for choosing between permutation tests and RF tests based on data characteristics.