FieldTrip is an Open Source project with contributions from various individual users and research groups.

Fieldtrip is primarily developed at the Donders Centre for Cognitive Neuroimaging (DCCN). Most people at the DCCN that use EEG or MEG in their research are using FieldTrip for (part of) their analysis. Some of these people also contribute by improving and extending the code.

The source code is maintained in Subversion/SVN, a version control system. A limited number of user has direct write access to the version control system. More details on the use of SVN for FieldTrip code management and release can be found here. All changes to the code are automatically available for the internal DCCN users, and every evening are also available in the latest version on the ftp server for the external users.

We have a seperate page where the known problems with the FieldTrip code and documentation are listed.

If you contribute to the FieldTrip code, either by improving existing functions, or by implementing new functions, please consider the code guidelines and try to adhere to the existing representation of the data in the data structures.

The remainder of this page contains documentation that pertains to the low-level development of FieldTrip. It is probably not of general interest to the typical end-users.

There are some low-level features of the FieldTrip code that are of interest to other Open Source software developers. To facilitate the re-use of those functions, we have split them off into seperate modules. You can download these modules as stand-alone code from our ftp server, and use them in your own projects. It would be nice if you attribute this to our work, e.g. by providing a link to our FieldTrip website.

• fileio import various EEG/MEG file formats into Matlab
• forwinv forward computation of the EEG/MEG field distribution of a dipole in a volume conduction model
• preproc preprocessing of EEG/MEG data, such as filtering and detrending
• realtime read data into Matlab in real-time for BCI/neurofeedback
• specest spectral estimation
• multivariate multivariate analysis of neuroimaging data
• plotting low-level functions for plotting and graphical user interfaces
• peer peer-to-peer distributed computing on an ad-hoc multi-core and/or multi-node Matlab cluster

• move internal fcdc documentation onto the wiki
• infrastructure for testing
• checkdata and make source-structure definition consistent with other datatypes. Implement consistent data structures for frequency and source data (e.g. add dimord to source data). Check which functions are potentially affected. This will facilitate merging of common functionality for source/freqdescriptives and cleaning up of sourceanalysis.
• ensure consistend trial definition
• read_neuroshare
• prefix all public functions with ft_
• CSP for classification
• check the documentation
• transform_grid
• add the spike functions from Martin
• Switch from SPM2 to SPM8
• provide an interface to the OpenMEEG software for BEM modelling
• ensure consistency of the documentation
• clean up inside_vol and similar functions
• Ensure consistent units throughout fieldtrip
• integration with SPM8
• Z-scores
• Implement online data processing and classification for BCI
• Implement support for a separation of data into a signal and noise subspace
• Check the correctness of the implementation of the algorithms
• Implement a framework for regression testing
• Document the deprecated functions and configuration options
• Implement a robust algorithm for constructing triangulated EEG-BEM head models
• timelockanalysis Reimplement the avg/cov/trial handling

• switch from CVS to SVN for the code version and release management (done)
• external_dependencies add stripped spm2 and other toolboxes as external dependencies (done)
• mvaranalysis Implement a function which computes an mvar-model based on the input data (done).
• couplinganalysis Implement a function which computes a variety of bivariate measures from the input data (done).
• handling of continuous data (done)
• Implement function that checks consistency of cfgs (done)
• Restructure the directory layout (done)
• Implement support for CTF synthetic gradiometers (done)
• integrate the new preproc module into fieldtrip (done)
• consistent flank detection for triggers (done)
• Implement trial selection option (done)
• Implement a graphical user interface as a "wizard" for certain analysis protocols (done)
• Redesign the interface to the read_fcdc_xxx functions (done)
• Check the consistency between the documentation and the implementations (done)
• Redesign and implement a common statistical backend for various data types (done)