Experiment design question

[Todd Thompson]

Paul and George - both solutions help; thanks so very much. I hadn't
yet played with permutation selection, and that seems to make things
quite simple.

Thanks again!
Todd



On Tue, 14 Dec 2004 22:19:38 +0100, Groot PFC <PFC.Groot at psy.vu.nl> wrote:
> Hi Todd,
>
> I'm not sure the following is (part of) the solution you're actually looking
> for, but it is probably the easiest way to let EPrime run one of all
> possible block permuations:
>
> - Make sure that the 4 blocks are part of a top-level block
>   list containing 4 levels (A..D)
> - Set the 'order' property of the block list to 'permuation'
>   and the OrderBy property to 'Group'.
> - Enable the group parameter in the startup info list and set
>   the minimum value to 1 and the maximum to 24.
>
> This allows you to choose any one of the 24 possible orders when starting
> the script (see attach.)
>
> If you would predefine this selection depending on subject number, then you
> would probably have to disable the 'group' startup parameter and insert some
> inline script at the beginning of the exp. which sets the group-attribute
> according to subject number and perform some block list initialization.
>
> best,
> Paul Groot
> Vrije Universiteit
> Amsterdam
> The Netherlands
>
> -----Original Message-----
> From: Todd Thompson [mailto:twthompson at gmail.com]
> Sent: dinsdag 14 december 2004 21:39
> To: eprime at mail.talkbank.org
> Subject: Experiment design question
>
> Hi, all. I've got a question that I hope will be very simple, but I haven't
> yet found a convenient answer. Any help will, well... help.
>
> I'm building an experiment that has four blocks of different trial types
> (A,B,C,D). I'd like to run these blocks in one script, one after the other.
> So far, no problem. I've got several lists: 1 that specifies the block
> orders, and four more that contain the trials for each block.
>
> However, I'd like to be able to run the blocks in a specified order,
> depending on the subject. For subject 1, I might want to run DCAB. For
> subject 2, I might want BCDA. And so on.
>
> Needless to say, I don't want to build and maintain one script for each of
> the 24 permutations of my block order. Can you think of a clever solution?
> Is there a way to manually specify the order of lists at the beginning of
> the experiment (via parsing a text input, if necessary, but preferably
> choosing from a drop-down or radio button)?
>
> Thanks so much!
> Todd
>
> ----
> Todd Thompson
> Helen Wills Neuroscience Institute
> University of California, Berkeley
>
>
>