Dear people,
I have recorded EEG responses to 3 different 1-minute songs. I want to compare the amplitude of one specific frequency (the frequency of the tempo of the songs, aka the BPM) in the EEG spectrum across songs. However, the amplitude of this same frequency in the spectrum of the envelope of the songs is different for each stimuli. I think these differences are caused by the musical characteristics of the beats of the songs.
I have not found literature on this topic yet, but I suspect that the amplitude of the frequency of interest in the spectrum of the songs is affecting the amplitude of the frequency of interest in the spectrum of the EEG responses. Therefore, I thought to normalize or correct the amplitude of the EEG response based on the amplitude of the songs, but I am lost on what is better for this analysis.
I am a newbie in EEG signal processing, so any help is appreciated.
Best Regards,
Raquel Aparicio.
Hi
This article may help you.
The Effect of Music on Human Brain; Frequency Domain and Time Series Analysis Using Electroencephalogram
https://ieeexplore.ieee.org/abstract/document/8416667
Amplitude of a tone will affect the amplitude of certain spectral characteristics, you can see an example of this when running an auditory augment-reducing paradigm.
It might help in your case to normalise your spectra for the EEG. There are a number of ways to do this depending on your method (FFT, Wavelet; frequency, time-frequency). A simple way is to divide each element in the array by the sum of the array. At least this way you now have your spectra on a comparable scale.
You might also want to check your spectra for each stimuli relative to a silence condition.
- Badges
- Science topic
- Similar topics
- Computer Science
- Program