How to implement ordered stimulus repetition in E-Prime

[David McFarlane]

>I think we are a bit in the dark on your exact needs still. If you'd
>write a script within e-prime to order the triallist each run of e-
>prime will give the exact same order, which isnt' what you want I
>think cause if I understand correctly you need 125 different 'orders'.
>Right? And these orders need to be known beforehand and can't be
>created 'on the fly', am I right to think so?

I didn't get what Valerie wanted either, liw, but your response rings 
a bell for me.  In our fMRI studies we often use a limited number of 
fixed pre-randomized orders so as to avoid the problem of 
multicollinearity during analysis of the functional brain images -- 
for correlation, deconvolution, or general linear modelling analysis 
to work we cannot allow any sequence to be a linear combination of 
any others, so we generally cannot leave this to chance by 
randomizing sequences "on the fly" (this is explained very nicely in 
the AFNI documentation from NIMH).

>In that case I think your best shot is to create 125 lists and store
>them in 125 textfiles that are loaded into the triallist by e-prime
>based on for instance subjectnumber.

You might also do this all directly in E-Studio with no external text 
files and no script.  You would have to make a master List with one 
row (level) for each possible sequence (in this case, perhaps 125 
rows).  You would then set its selection Order to Counterbalance, and 
Order  By to Subject (or perhaps to Session).  Counterbalance will 
select and run one *and only one* row from the master List, based on 
Subject (or Session).  (This is explained somewhere in the EP docs, 
or just search the EP Group with the term "Counterbalance" to see 
where I recently explained this.)  Back in your master List, you 
could have each row call a customized Procedure for one of the 125 
random orders, or have each row use the same Procedure but a 
different nested List for the random order, etc.

Don't know if this addresses Valerie's question, but it does address 
an fMRI issue that crops up from time to time, so I am adding this to my FAQ.

-- David McFarlane, Professional Faultfinder

--

You received this message because you are subscribed to the Google Groups "E-Prime" group.
To post to this group, send email to e-prime at googlegroups.com.
To unsubscribe from this group, send email to e-prime+unsubscribe at googlegroups.com.
For more options, visit this group at http://groups.google.com/group/e-prime?hl=en.