I have a quick and probably basic question. I just wanted to verify the units of the output data when using the normalization functioning in the non-linear models. Are these all z-scores?
Thank you!
Stephanie
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
if you choose "normalise" a GLM or non-linear model, the input data are normalised - not the output. So the output are not z-scores.
Staib et al. (Psychophysiology, 2015) have shown that it improves SNR if you normalise the output of the model - but this is of course only possible in within-subject designs.
In Staib et al. we proposed to z-score the model output. Biophysically and statistically it may also make sense to divide the model outputs by estimates from a "Baseline" condition, e. g. a CS- in a fear conditioning design.
Best
Dominik
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi Dominik,
I have a quick and probably basic question. I just wanted to verify the units of the output data when using the normalization functioning in the non-linear models. Are these all z-scores?
Thank you!
Stephanie
Hi Stephanie
if you choose "normalise" a GLM or non-linear model, the input data are normalised - not the output. So the output are not z-scores.
Staib et al. (Psychophysiology, 2015) have shown that it improves SNR if you normalise the output of the model - but this is of course only possible in within-subject designs.
In Staib et al. we proposed to z-score the model output. Biophysically and statistically it may also make sense to divide the model outputs by estimates from a "Baseline" condition, e. g. a CS- in a fear conditioning design.
Best
Dominik