Number of SCR peaks

[Eichhof]

Hi,

The EDA stuff works but unfortunately I’m getting only peaks for 50% of my stimulus (10s pictures), i.e. in 50% of these 10s windows (picture duration) there is no peak. I have to evaluate a bit more to see the reason behind it (it seems a bit strange to me). Do you have any idea why this happens?

[DominiqueMakowski]

It's indeed normal (and almost reassuring) to not have peaks for all of your stimuli, as the algorithm that finds the peaks is a bit more complex than only "taking the max value of a particular window". To be considered a true SCR peak, a peak must followed a certain pattern (see the eda_scr function for more details).

Following that, you have several options: 1) Do not bothering with the peaks and taking the max value of your EDA signal in a window 2) Using the magnitude of the peak, ie input 0 when there is no peak 3) Using the amplitude of the peak, ie the value of "true" (detected) peaks only 4) Using the number of peaks rather than their magnitude (= 80% of my emotional stimuli were followed by a peak, while only 20% of the neutral stimuli elicited a peak) 5) Decreasing the value of the scr_treshold parameter in eda_process. This will lower the treshold of peak detection and, thus, detect more "peaks"

[Eichhof]

Thank you very much for your answer.

Regarding the first option, would you recommend taking the max of the raw EDA signal or the max of the phasic component?

[DominiqueMakowski]

@Eichhof I would use the phasic component, especially if you're interested in emotional processing. However, it mostly depends on your needs as always 😸

[Eichhof]

Thank you for the fast answer.