EEGLAB bugs
Table of contents
Dealing with EEGLAB bugs and suggestions
EEGLAB bugs are managed under Github EEGLAB Issues. The old EEGLAB Bugzilla interface has been deprecated, although it is still being used to track potential improvements.
Since EEGLAB has been developed under MATLAB, there are little risks that using EEGLAB will crash your machine or erase files inadvertently, unless MATLAB itself crashes. This is one advantage of using MATLAB.
About 1000 test cases run daily on EEGLAB code to test its integrity and that its functions are stable.
EEGLAB is an open-source project. To understand in more detail how any signal processing is performed, you may study the function source file. To adjust its performance, you may edit it yourself. Note: If you do this successfully, please consider issuing a pull request on Github. See more information on contributing to EEGLAB.
How do I report a Bug?
If you encounter a bug:
- Please first read the MATLAB command-line and any error window text carefully to determine whether you may be able to avoid the problem directly.
- Next, test whether the error occurs using the current release of EEGLAB. If so, check the Github EEGLAB Issues to see if your issue has already been reported:
- If your issue has been reported, you may comment on the bug.
- If your issue has not been reported, you may submit a new bug. Once you press the “New issue” button, you will be guided on how on the type of information needed to report your bug.
Mex files errors in EEGLAB
EEGLAB itself does not include any precompiled functions (also called mex functions). If you get an error “could not locate mex file” or any related mex file error, do not blame EEGLAB. However, some external modules use mex files for reading binary data files or performing source localization. Usually, an error indicates the precompiled file is not available for your platform (it would thus need to be recompiled, something you can sometimes do yourself - for more information, see below). There are mainly four modules in EEGLAB that use mex files.
-
Fieldtrip functions: If you get an error that some functions in the Fieldtrip folder cannot be found, refer to the Fieldtrip documentation for how to recompile such functions.
-
BIOSIG: Some release of BIOSIG contains some updated mex files so you might want to check if you have the latest version of BIOSIG. Sometimes BIOSIG does not preserve backward compatibility so you may experience problem when reading data after updating BIOSIG.
-
ANT plugin: The ANT plugin was made by the ANT company. Contact ANT for an updated version of the compiled binaries.
-
ERPSS plugin: Simply recompile decompresserpss.c (type “mex decompresserpss.c”)
Some known EEGLAB bugs and/or missing features:
-
Epoch selection using pop_eegplot.m: Epochs selected for rejection using eegplot data scrolling are not saved in EEGLAB history until the epochs are rejected. This means that they will not be reproduced automatically in a new EEGLAB session. However, the labeled epochs are identified in the field EEG.reject.manualrej that is saved along with the dataset. Also, and more importantly, when the labeled epochs are rejected, this operation is saved in EEGLAB history.
-
Zooming using pop_eegplot.m: When zooming and selecting epochs, only the data (no the background epoch markings) may be zoomed.
-
Spectral analysis (with no MATLAB Signal Processing Toolbox): The spec.m function emulates the function psd.m but not the function pwelch.m. As a result, the scaling of the spectrum (by the spectopo.m function only) may differ. Also, for unknown reasons, the spec.m function cannot handle frequencies that have been filtered out and may return inaccurately high power at these frequencies.
-
MATLAB versions and OS: MATLAB versions have different bugs under different OS and these bugs - usually graphical bugs - may affect EEGLAB. The latest one we know of is the fact that MATLAB version 2018a (all OS) requires a patch for EEGLAB to work.