BrainMap taxonomy of experimental design: description and evaluation

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Neuroimaging

Background:

• Coordinate-based, voxel-wise meta-analysis is a growing field in human functional brain mapping.
• Rigorous meta-analyses depend on clear selection criteria for experimental designs.
• The BrainMap database offers a coding scheme for experimental designs, serving as a potential taxonomy.

Purpose of the Study:

• To describe and evaluate the BrainMap experimental-design taxonomy.
• To compare taxonomy fields with data-filtering choices made by experts in functional neuroimaging meta-analyses.

Main Methods:

• The study assessed 15 publications comprising 46 voxel-wise meta-analyses, pooling data from 351 original publications.
• BrainMap filter implementations were graded by ease-of-use (A-C) and stage-of-use (1-3).
• Quality and content filters were tabulated separately, with mandatory quality filters identified.

Main Results:

• All authors unanimously adopted the five mandatory BrainMap quality filters, indicating high agreement on data quality.
• Non-mandatory quality filters (group size, imaging modality) were universally applied (Stage 1, Grade A).
• Content filter choices showed a high proportion of accessible filters (Stage 1, Grade A: 68%; Stage 2, Grade A: 14%; Stage 2, Grade C: 18%).

Conclusions:

• The BrainMap experimental-design taxonomy is a valuable tool for structuring and executing robust meta-analyses in functional brain mapping.
• The taxonomy demonstrates strong concordance with expert practices in data selection and filtering.
• The findings support the utility of the BrainMap taxonomy for ensuring consistency and rigor in neuroimaging meta-analysis.