Knights et al. (2021). Brain decoding during real actions.

This repository contains code accompanying the timeseries machine-learning project that is published in the Journal of Neuroscience: https://www.jneurosci.org/content/41/24/5263

Prerequisites

• Python 3.0 (for dcm2bids)
• LibSVM (for machine learning classification; https://www.csie.ntu.edu.tw/~cjlin/libsvm/index.html)
• Searchmight matlab toolbox (for whole-brain searchlight machine-learning; http://www.franciscopereira.org/searchmight/)

1) Conduct the Experiment

Add minilab.m on to the matlab path before initiating the fMRI experiment: runExperiment/run_MRI_realAction.m

2) Convert dataset to standard BIDS format

Perform the .dcm conversion using python via: BIDSconversion/convertBIDS.m
Alternatively, use datalad to download the fMRI data from openneuro platform: https://openneuro.org/datasets/ds003342/versions/1.0.0

3) Perform machine-learning classification

Run the Leave-One-Run-Out cross validation MVPA or searchlight libsvm classifiers: https://github.com/fws252/ROI-based-and-searchlight-MVPA-decoding-Knights-et-al-2020-
Alternatively, download the derived MVPA summary dataset from the Open Science Framework platform: https://osf.io/wjnxk/

4) Hypothesis testing

Compare decoding accuracy using across classifiers using Bayesian t-tests & ANOVAs: stats/BayesTests.m

5) Generate plots

The resulting machine-learning distribution plots can be generated via: plots/run_violinPlots.m

How to Acknowledge

Please cite:
Knights, E., Mansfield, C., Tonin, D., Saada, J., Smith, F. W., & Rossit, S. (2021). Hand-selective visual regions represent how to grasp 3D tools: brain decoding during real actions. Journal of Neuroscience, 41(24), 5263-5273.