[Eeglabnews] EEGLAB Version 6.00b

[arno]

A major new release of the EEGLAB environment, version 6.00-beta, is
now available at http://sccn.ucsd.edu/eeglab/. We are now also making
available a weekly release of EEGLAB on the same download page. We
particularly wish to thank Andreas Romeyke and Ronny Lindner of the
Max-Planck-Institute for Human Cognitive and Brain Sciences, Leipzig,
Germany, who have implemented a large set of testcase functions,
allowing us to increase the stability of EEGLAB functions. New
features in v6 include:

All *STUDY functions* have been updated to compute statistics across
subjects using either a parametric or a non-parametric method. Also
operations such as filtering, resampling, or computing all dataset
ICA decomposition are now available at the STUDY level. The STUDY
structure for cluster had to be modified slightly (see updated
tutorial below) but all functions remain backward compatible.

STUDY functions have been updated or added to process scalp channel
data as well as independent component data. For all new STUDY
features, check screen captures at

http://sccn.ucsd.edu/eeglab/clusttut/clustertut.html

Other notable changes in EEGLAB v6 include:

*time-frequency decompositions*: newtimef() and newcrossf() now perform
standard time-frequency decompositions instead of timef() and crossf().
Both are based on the new timefreq() function, and are mostly backward
compatible with their predecessors although they have been completely
reprogrammed. The new functions are the fruit of 3 years of development
and testing.  They fix minor bugs (e.g., the data mean should be removed
before computing wavelets). The new function have a new range of options
including log-frequency plotting, individual frequency selection,
individual cycle selection, true Morlet wavelets, single-frequency
plotting, etc... The old timef() and crossf() functions remain available
in the distributions, though the EEGLAB menu now call newtimef() and
newcrossf().

*Mutual information plugin*: this plugin, contributed by Nima Bigdeley
with Jason Palmer, allows clustering ICA components within a single
dataset. See its tutorial.

*Function organization*: EEGLAB functions are now better organized
in subfolders.

*ERP enveloppe plotting and scalp maps*: clicking on the ERP now shows
the ERP scalp map at the specified latency (timtopo() function).

*DIPFIT*: support for localization of MEG components for CTF system data,
thanks to Nicolas Robitaille.

*Channel interpolation*: eeg_interp() and std_interp() now implement
channel interpolation for datasets and STUDY respectivelly.

For a more detailed list of changes in EEGLAB v6, see
http://sccn.ucsd.edu/eeglab/revisions.html

To report a bug, or to suggest a new feature, please use Bugzilla for
EEGLAB at http://sccn.ucsd.edu/eeglab/bugzilla