Code for project comparing Neurosynth and BrainMap functional decoding
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.



All code necessary to reproduce the analyses run for a poster presented at INCF's Neuroinformatics 2018 meeting, entitled Quantitative comparison of functional decoding approaches across meta-analytic frameworks, comparing lexicons and functional decoding results from BrainMap.org1,2 and Neurosynth.org3.


No installation necessary, Python code was written in 2.7.13 because I'm stubborn and didn't want to update to Python 3. Requires numpy, pandas, scipy, seaborn, and neurosynth to recreate python analyses and figures. Regrettably, recreating BrainMap-style functional decoding depends on MATLAB.


All files comparing terms from each database are located in terms, scripts required for performing BrainMap functional decoding are located in /brainmap, and scripts required for performing Neurosynth functional decoding are located in /neurosynth. Decoding input for both methods is a nifti file of a binary region/network mask and output is a series of files detailing feature weights for said region/network.

Neurosynth Decoding

  1. ns-get-dataset.ipynb
  2. ns-decoding.ipynb

BrainMap Decoding

  1. BMA_ImageSearch.m
  2. BMA_ForRevInf.m

Run decoding_comparison.ipynb to compare results of decoding. Due to the format of BrainMap's decoding output, significant CogPO terms must be typed in manually.


  1. Fox PT, Lancaster JL. (2002) Mapping context and content: The BrainMap model. Nature Rev Neurosci 3, 319-321.
  2. Laird AR, Lancaster JL, Fox PT. (2005). BrainMap: The social evolution of a functional neuroimaging database. Neuroinformatics 3, 65-78. pdf
  3. Yarkoni T, Poldrack RA, Nichols TE, van Essen DC, Wager TD. (2011). Large-scale automated synthesis of human functional neuroimaging data. Nature Methods 8, 665–670. link