The Decoding Toolbox (TDT)

TDT is a MATLAB toolbox for multivariate analysis of functional and structural MRI data. It supports searchlight, region-of-interest, and whole-brain analyses, together with classification, regression, representational similarity analysis, feature selection, and parameter selection.

The current stable version is 3.999K. Download the ready-to-use ZIP from the latest GitHub release. The default main branch tracks stable releases; ongoing development is kept on the develop branch.

Requirements

• MATLAB 7.3 or later. Version 3.999K was release-tested with MATLAB R2021b.
• SPM or AFNI for reading and writing brain images. TDT 3.999K adds tested SPM25 support. It also includes forward-compatible SPM26 dispatch adapters.
• A working MEX build of LIBSVM or LIBLINEAR when using those classifiers. Precompiled builds are included for several platforms; see README.txt if recompilation is necessary.

Installation

1. Download and extract the latest release ZIP.
2. Add the extracted decoding_toolbox directory to the MATLAB path.
3. Add SPM or AFNI to the MATLAB path if you will work with brain-image files.

addpath('/path/to/tdt_3.999K/decoding_toolbox')
cfg = decoding_defaults;

Calling decoding_defaults adds the TDT subdirectories automatically.

Getting started

For a standard SPM analysis, start with:

help decoding_example

AFNI users can start with help decoding_example_afni. For an editable walkthrough, open decoding_tutorial.m; reusable scripts are in templates, and complete examples are in demos.

The project website provides background, datasets, and additional documentation: TDT – The Decoding Toolbox.

Citation

If you use TDT, please cite:

Hebart, M. N., Görgen, K., & Haynes, J.-D. (2015). The Decoding Toolbox (TDT): A versatile software package for multivariate analyses of functional imaging data. Frontiers in Neuroinformatics, 8, 88. https://doi.org/10.3389/fninf.2014.00088

Support and bug reports

• Ask usage questions on Neurostars with the tdt tag.
• Report reproducible bugs through GitHub Issues.
• Review release changes in LOG.txt.

Please include the TDT version, MATLAB version, image software/version, relevant configuration, and the complete error message in bug reports.

License

TDT is distributed under the GNU General Public License; see LICENSE.txt and GPLv2.0.txt. Bundled third-party components retain their own licenses, documented in LICENSE.txt and their source directories.