Beyond p-values: averaged and reproducible evidence in fMRI experiments

Area of Science:

• Neuroimaging
• Cognitive Neuroscience
• Functional Magnetic Resonance Imaging (fMRI)

Background:

• Functional magnetic resonance imaging (fMRI) studies may overlook subtle activation regions due to modest response magnitudes.
• Conventional p-value thresholding methods can fail to detect these small but significant brain activity areas.
• Understanding reproducible activation patterns is crucial for robust neuroimaging research.

Purpose of the Study:

• To present brain-activation patterns that are reproducible across experimental modalities and robust to pattern generation methods.
• To highlight the limitations of conventional p-value approaches in detecting subtle fMRI signals.
• To explore the neurophysiological basis of reproducible brain activation patterns.

Main Methods:

• Utilized a reproducibility analysis method, referencing Liou et al.
• Investigated within- and between-subject brain-activation patterns.
• Compared findings with conventional group-averaged data analysis.

Main Results:

• Identified reproducible within- and between-subject brain-activation patterns.
• Demonstrated that reproducibility analysis can detect patterns missed by conventional p-value methods.
• Observed both increased and decreased responses in the precuneus and posterior cingulate, contrary to group-averaged findings.

Conclusions:

• Reproducible patterns offer a more complete picture of brain activity than conventional averaged data analysis.
• There is a neurophysiological basis for these reproducible activation patterns.
• Findings have potential applications in scientific and clinical neuroimaging studies.