Data visualization for inference in tomographic brain imaging

Pernet, Cyril R; Madan, Christopher R

Abstract

Tomographic brain imaging has a rich iconography. Whilst figures are prepared for scientific communication (i.e., directed to other researchers) they also often end-up on magazine and journal covers (i.e., directed to a lay audience). Scientific figures should however not be just glossy illustrations of what is in the text. One of the primary roles of figures is to carry information that cannot be easily explained in words or summarized in tables (Rougier et al., 2014). Poor scientific figures are figures that not only fail to convey additional information, but also figures that convey or induce incorrect information, especially for non-specialists. Here we provide a guideline on which visual information to display and in which context, to improve information content and minimize false inference. We first discuss the use of slices versus renders and in which situations they should be used. We next reiterate the need for unthresholded statistical maps (Jernigan et al.,2003) along with (i) the highlighting of significant areas on such maps (ii) the necessity to plot results in all regions of interest, and (iii) the choice of colour scales. Together, these measures provide additional contextual information and should prevent readers natural tendency to falsely infer differences in activations or absence of activations. Additional recommendations are also given to convey information about hemispheric asymmetry and effect sizes.

Citation

Pernet, C. R., & Madan, C. R. (2020). Data visualization for inference in tomographic brain imaging. European Journal of Neuroscience, 51(3), 695-705. https://doi.org/10.1111/ejn.14430

• Journal Article Type: Editorial
• Acceptance Date: Apr 17, 2019
• Online Publication Date: Jun 4, 2019
• Publication Date: 2020-02
• Deposit Date: Apr 25, 2019
• Publicly Available Date: Jun 5, 2020
• Journal: European Journal of Neuroscience
• Print ISSN: 0953-816X
• Publisher: Wiley
• Peer Reviewed: Peer Reviewed
• Volume: 51
• Issue: 3
• Pages: 695-705
• DOI: https://doi.org/10.1111/ejn.14430
• Public URL: https://nottingham-repository.worktribe.com/output/1840206
• Publisher URL: https://onlinelibrary.wiley.com/doi/full/10.1111/ejn.14430
• Contract Date: Apr 25, 2019