Stimulus coding in various HSSM models

[hyang336]

Dear HSSM community,

What is the correct way to use stimulus coding for responses and models? If using DDM, I take that "-1" means "incorrect" and "1" means "correct" in the response column. Could I instead use those two values to represent two response options instead of accuracy? For something like race model with more than two response options, is the response by-default stimulus coded (i.e., each particle represents evidence accumulation of one response option)?

Thanks!

[digicosmos86]

Hi @hyang336,

We had a discussion today and we are going to implement stim coding in the next few months. Please stay tuned :)

[tigretigre]

Any updates on this one? Anything I can do to contribute to move it along more quickly?

[frankmj]

Hi there, In the interim you should be able to do stimulus coding by including a stim column that tells the model which response should have been made for that stimulus, and a response column that corresponds to which response was made (rather than correct and errors). Then if you coded stim as simply 1 and -1 then specify v ~ stim this should give positive coefficients for stim. Michael

…

On Fri, Sep 20, 2024 at 12:25 PM tigretigre ***@***.***> wrote: Any updates on this one? Anything I can do to contribute to move it along more quickly? — Reply to this email directly, view it on GitHub <#475 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAG7TFC3ZACZEJ4J4X3MQDLZXREA5AVCNFSM6AAAAABKBCACKKVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTANZQG4ZDGNI> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>