[FieldTrip] ft_sourceplot on surface

[nwolpert at uni-osnabrueck.de]

Dear fieldtrip users,

I have an issue when visualizing beamformed data: When using
'ft_sourceplot' to show the interpolated data on slices in the three
planes (cfg.method = 'ortho'), reasonable results are produced. However,
when doing surface plots by calling cfg.method = 'surface', the power is
projected only to a subset of the brain, with very abrupt borders and zero
power everywhere else. It gives the impression that there is some kind of
scaling problem (i.e. the surface template used has the wrong scale), but
I can't figure out the cause for this problem.

I would be glad if anyone has a possible explanation. I attach the code
below that produced the figures.

Best,
Nicolai Wolpert

cfg              = [];
cfg.method       = 'dics';
cfg.grid         = grid;
cfg.headmodel    = headmodel;
cfg.dics.projectnoise = 'yes';
cfg.dics.lambda       = 0;
sourceAlpha_dics = ft_sourceanalysis(cfg, freq_alpha);

% remove center of head bias by using the neural activity index (NAI)
sourceAlpha_dics.avg.pow = sourceAlpha_dics.avg.pow ./
sourceAlpha_dics.avg.noise;

% reslice mri
mri_real_resliced = ft_volumereslice([], mri_real);

% Interpolation: Project computed sources on aligned template MRI brain
cfg            = [];
cfg.downsample = 2;
cfg.parameter = 'avg.pow';
sourceAlpha_dics_intp  = ft_sourceinterpolate(cfg, sourceAlpha_dics ,
mri_real_resliced);

% visualize the result
cfg              = [];
cfg.method       = 'surface';
cfg.funparameter = 'avg.pow';
ft_sourceplot(cfg, sourceAlpha_dics_intp);