Hi Vladimir and other SPM EEG experts, I am averaging my ERP data both in time and time/frequency domains and I am using SPM's default electrode locations for 64 (10-20) electrodes. In my time domain averages, the electrode locations look fine, while in my time-freq data, all the electrode (I am selecting 24 of the 64 electrodes for the TF decomposition and a frequency window of 3 - 30 Hz) seem to be located on the front of the head. The section of my script that does TF decomp is as follows: ... S = []; S.D = <filename>; S.tf.frequencies = [3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30]; S.tf.Mfactor = 5; S.tf.channels = [1 2 3 8 10 12 17 19 21 26 28 30 36 38 49 46 48 50 54 56 58 60 61 62]; S.tf.phase = 0; S.tf.pow - 1; D = spm_eeg_tf(S); ... However, when I convert these TF averages to images and do stats on them, I see results around the right (original) electrode sites, but when i extract the power for that electrode and do stats outside of SPM, the results don't correspond to the SPM results. Is it due to mis-registration of electrode during the TF decomp step? Any help will be much appreciated. Thanks -Muhammad