[Eeglablist] newtimef, 'powbase' problem

[Christian Scharinger]

Dear eeglab users,

never running out of questions and problems, I have another one:
In order to get more familiar with the newtimef function and the 
possibility to feed in an external baseline ('powbase') I tried the 
following:

1) Calculate the ERSP for one experimental condition:
[ersp_data,ersp_itc, ersp_powbase,ersp_times,ersp_freqs] = 
newtimef(EEG.data(channel,:,:), EEG.pnts, [EEG.xmin EEG.xmax]*1000, 
EEG.srate, 'cycles', [0], 'itctype', 'phasecoher', 'detrend', 'off', 
'rmerp', 'off', 'padratio', [4], 'freqs', [2 22], 'nfreqs', [20], 
'freqscale', 'linear', 'verbose', 'on', 'baseline', [0], 'powbase', NaN, 
'plotitc', 'off');

This works well.

2) Do the same calculation but this time using the baseline calculated 
under 1) (ersp_powbase) as input for the 'powbase' argument.
[ersp_data2,ersp_itc2, ersp_powbase2,ersp_times2,ersp_freqs2] = 
newtimef(EEG.data(channel,:,:), EEG.pnts, [EEG.xmin EEG.xmax]*1000, 
EEG.srate, 'cycles', [0], 'itctype', 'phasecoher', 'detrend', 'off', 
'rmerp', 'off', 'padratio', [4], 'freqs', [2 22], 'nfreqs', [20], 
'freqscale', 'linear', 'verbose', 'on', 'plotitc', 'off', 'powbase', 
ersp_powbase);

I expected the data saved in ersp_powbase under 1) is exactly what is 
required as input for the 'powbase' argument. So using this data 
structure under 2) should give me exactly the same figure as in 1).
Unfortunately it does not. All what happens is a serious crash of my 
matlab environment when executing the code under 2).

Maybe someone has any suggestions what could be wrong with my code under 2)?

Best,
Christian


-- 
Christian Scharinger, M.A.

Knowledge Media Research Center (KMRC)
Schleichstraße 6 | 72076 Tuebingen

Phone: +49 (0)7071 979-360
Internet: http://www.iwm-kmrc.de/c.scharinger
E-Mail: c.scharinger at iwm-kmrc.de