after ICA, N100 disappear

Mailing List Archive (read-only)
1

Hi,

At first I used IGOR pro for EEG analysis, applying artefact rejection on
my data. And looking at the grand mean, I could see the visual N100
response in each condition(my experimental task was visual).

Then, in order to gain more trials I decided to analyse my data with your
package in Python for EEG analysis, by applying ICA. But now when I look at
the graphs of the grand mean but also at the individual level, the visual
N100 response has completely disappeared.

What happened? Is there an explanation? Many thanks

Cheers,
Emanuela Liaci
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20161020/c873fa71/attachment.html

2

to me your N100 can disappear for 3 reasons:
- you rejected a good ica components
- you reduced too much the number of components and did not put back
small PCA components
- you applied a stronger high pass filter

HTH
Alex