Introduction to DDTBOX
DDTBOX is a toolbox for multivariate pattern analysis (MVPA) of epoched EEG data. DDTBOX can be used as a standalone toolbox in MATLAB, however we also plan to add functionality as an EEGLAB plugin in a future version. For an overview of the functionality of DDTBOX, please see our publication describing the toolbox.
DDTBOX core scripts were originally written by Stefan Bode. The toolbox was written with contributions from: Daniel Bennett, Daniel Feuerriegel and Phillip Alday. The first version of the wiki was written by Daniel Feuerriegel.
The authors further acknowledge helpful conceptual input/work from: Jutta Stahl, Simon Lilburn, Philip L. Smith, Elaine Corbett, Carsten Murawski, Carsten Bogler, John-Dylan Haynes.
Copyright (c) 2013--2017 Stefan Bode and contributors.
Unless otherwise specified, code is distributed under the GNU Public License (GPL) version 2, and documentation under a Creative Commons Attribution-ShareAlike 4.0 International License.
We hope that you find the software and documentation useful. If you publish an analysis using the toolbox, we ask that you cite us.
A sample citation would be:
Bode, S., Feuerriegel, D., Bennett, D., & Alday, P.M. (2018). The Decision Decoding ToolBOX (DDTBOX) - A multivariate pattern analysis toolbox for event-related potentials. Neuroinformatics, 1-16. doi 10.1007/s12021-018-9375-z
The code in DDTBOX depends on the functionality supplied by LIBSVM for support vector machine classification and regression. We also offer support for LIBSVM's specialised and often faster cousin, LIBLINEAR. Backends for other classifiers may also be added to DDTBOX in the future.
If you publish analyses using LIBSVM or LIBLINEAR, please support the developers of these libraries by citing them.
Sample citation for LIBSVM:
Chih-Chung Chang and Chih-Jen Lin, LIBSVM : A library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2:27:1--27:27, 2011.
Sample citation for LIBLINEAR:
R.-E. Fan, K.-W. Chang, C.-J. Hsieh, X.-R. Wang, and C.-J. Lin. LIBLINEAR: A library for large linear classification. Journal of Machine Learning Research 9(2008), 1871--1874.
LIBSVM and LIBLINEAR come packaged with DDTBOX, however you may need to configure MATLAB to use these external dependencies. Please see their respective documentation or the Getting Started guide in this wiki for more information.
Some data transformation and statistical analysis methods rely on functions in the MATLAB statistics and machine learning toolbox, for example the ttest and zscore functions.
However, decoding and group-level statistical analysis options are available that do not rely on this MATLAB toolbox.