Normalization of LFP power

#1

Hi,

I have seen in many papers that they normalize the time-frequency plot of each channel and then average it to have a single picture. In the paper’s it has been mentioned, " To control for impedance differences between tetrodes, LFP power was normalized, for each tetrode, to power in 5–10 Hz band, except for the analysis in Figure 1E. "
Does anyone have any idea how to implement this kind of normalization in Code?
In brief, I have 14 time-frequency matrices; what should I do before the averaging?

#2

Hi Faraz. If you have task-related activity, you can normalize to a pre-trial baseline. I guess what those authors did was normalize the spectrum to the average of theta power.

In general, spectral normalizations all involve the ratio activity/baseline. The question is how you define the baseline, whether it’s per-frequency from a time window or power from a specific frequency range (or the entire spectrum).