Expected results for AUC and SF

[daniell-00]

I wold like to verify if I'm getting correct data. Could anyone tell me how are scaled the expected results using AUC and MP methods for pspm_sf using epochs about 300 s using with norm = 1. I'm getting results such as follows:

AUC: 0.55060383, 0.52754537, 1.58868552, 0.47168485, 1.58047146

SF: 0.35321559, 0.31989337, 0.34380557, 0.33831149, 0.33127456

Thanks in advance!

[dominikbach]

Happy to help!

1. In the current PsPM version 7.0, options.norm will not be used. If you are using a previous version then please send version number for me to check.
2. For AUC, units are the same as the original units. It is returning AUC/epoch duration, where AUC is computed after subtracting the baseline. In other words, what you get is mean(SCR-min(SCR)).
3. For SF, units are in Hz - i.e. number of above-threshold responses per second. In your example, there is about 1 subthreshold response every 3 seconds.
   Dominik

[daniell-00]

Hello

I'm using PsPM_v7.0.0, I've seen it is suggested to normalize data for within-subjects analysis, what options do I have to normalize data when using auc and mp methods?

[dominikbach]

Normalising SCR data could be beneficial for time-series analysis with GLM and DCM, due to large between-subject differences in SCR scaling.

SCR scaling is not an issue for the number of SF, so MP (and SF-DCM) analysis does not need normalisation.

SCR scaling might be an issue for AUC, but presently there is no way to normalise the data within PsPM for this analysis.