How to change the loss aversion in igt_vpp

[Yijun Chen]

Hi, everyone.

I am a beginner to study this package.

My task is IGT(Iowa Gambling Task). So I try to use the models related with igt.

And I saw the description that "It has the following parameters: A (learning rate), alpha (outcome sensitivity), cons (response consistency), lambda (loss aversion)."

I wanted to change these parameters. For example, remove loss aversion.

But in the usage part, I couldn't find how to change this.

Could you kindly help me with this?

Thanks in advance.

Best regards,

Yijun

[Lei Zhang]

Hi, 

you might need to change the stan file if you would like to modify the model. 

do save it as a new file, and then call it externally. 

hope it helps,

Lei

[Jeung-Hyun Lee]

Hi Yijun,

We apologize for the long delay!

Upon Lei’s post above, you can simply remove all lambdas referred in the model (including subject-level and group-level parameters) (link: https://github.com/CCS-Lab/hBayesDM/blob/develop/commons/stan_files/igt_vpp.stan). You might want to make sure that you have the indices of each list sentenced in the model (e.g., the total number of parameters becomes 7, not 8).

Let us know if you have further questions.

Best,
Jeung-Hyun

2023년 2월 7일 화요일 오전 3시 38분 25초 UTC+9에 Lei Zhang님이 작성: