Related Experiment Video
Updated: May 9, 2025
07:45
Imaging and Quantification of Intact Neuronal Dendrites via CLARITY Tissue Clearing
Published on: April 20, 2021
3.4K
Go Figure: Transparency in neuroscience images preserves context and clarifies interpretation.
Paul A Taylor1, Himanshu Aggarwal2, Peter A Bandettini3
1Scientific and Statistical Computing Core, NIMH, NIH, Bethesda, MD, USA.
Arxiv
|April 29, 2025
Summary
Transparent thresholding in neuroimaging visualizations includes subthreshold data, improving interpretation and reducing bias. This approach enhances reproducibility and clarifies results by providing essential experimental context alongside significant findings.
Area of Science:
- Neuroimaging
- Scientific Visualization
- Data Analysis
Background:
- Traditional neuroimaging figures display only statistically significant regions.
- This practice introduces bias in results interpretation and meta-analyses, contributing to non-reproducibility.
- Subthreshold data, crucial for experimental context, is often omitted.
Purpose of the Study:
- Advocate for a "transparent thresholding" method in neuroimaging.
- Integrate statistically significant and subthreshold data in visualizations.
- Enhance the interpretation and reproducibility of neuroimaging findings.
Main Methods:
- Developed and advocate for the "transparent thresholding" approach.
- Presented four examples illustrating the benefits of transparent thresholding.
- Demonstrated software packages supporting transparent thresholding.
Main Results:
- Transparent thresholding removes ambiguity and reduces hypersensitivity to non-physiological features.
- This method aids in identifying potential artifacts and improves cross-study comparisons.
- Transparent thresholding significantly reduces non-reproducibility biases.
Conclusions:
- Transparent thresholding offers a balanced approach to visualizing neuroimaging data.
- It improves clarity, reduces bias, and enhances the reproducibility of scientific findings.
- Wider adoption of transparent thresholding is encouraged for better neuroimaging research.
