[open] Psi adaptive / Psi-Marginal

mark

Hi all,
I was wondering if anyone has played around with implementing the psi adaptive method (or it's update, the psi-marginal method) in OpenSesame via a code component, and if so, if they'd be willing to share their code. I want to put a calibration step in an experiment and this seems to be the best way, so I'd love to get a look at someone's implementation so that I can learn the ropes.
Thanks,
Mark

sebastiaan

Hi Mark,

I suspect that you will be the first to do this. I looked up the Psi marginal method and I assume you are referring to Prins (2013, J Vis), right?

That looks pretty complicated, at least at first glance. But assuming that you are familiar with this method, I'm sure we can figure out how to apply your knowledge to OpenSesame. Can you explain in more detail what this method does, what kind of paradigm you use, which parameters you estimate, and more generally what the Psi-marginal method does?

Cheers,
Sebastiaan

mark

Hi Sebastiaan,

Unfortunately I can't claim any expertise in the method. I went searching for an method that could give psychometric slope as well as boundary and the Psi-Adaptive technique appears to be the best current method. I was hoping to find a software implementation I could use - it is implemented in Palamedes, but that requires Matlab. Ivar Clemens and colleagues at the Donders Institute have created a function in Python for the Psi adaptive technique, but I'm afraid my background in both programming and psychometrics is insufficient to reliably implement this in OpenSesame.

If anyone out there has the skills and is interested, Ivar has kindly given permission for me to post their code. Here's the link:

https://github.com/wilberth/Rudolph/blob/master/root.py

Thanks,

Mark