Variable Precision Model without guessing parameter

[Lena Galeano]

Hi,

I would like to fit and compare the variable precision models (VP GaussianSD, VP GammaPrecision) to each subjects data but without estimating the guessing behavior (g).
To this end, I simply removed all guess rate related implentations in your MemToolbox scripts (see attachment).

When I fit this model version without guessing (Gaussian SD) to the data I have recognized that some values are very close to the upper limit of 100.
Because of this, I was wondering if this is really the right way to do? Or are there other steps to do to estimate only precision (mnSD, sdmode) and the variability of precision (stdSTD, sdPrecision) without the guess rate parameter?

I would be more than happy if you shared your ideas on this topic with me.

Thanks in advance,
Lena