Pseudo-Random Lists

[Robert Dörr]

Hi,

That's exactly what I need. I'll try this afternoon...

Thanks a lot
Robert



Tony Zuccolotto wrote:

>Robert,
>
>Check out the ExplicitOrder object and short example script in E-Basic Help.   This order can only be setup from script so that option is not currently present as an option in the graphical interface.   
>
>In short, if you can define and write the algorithm in E-Prime script to decide the order of the levels/rows that you want, then you can reassign the List.Order property to use ExplicitOrder and then when the list runs it will do so in the explicit order that you have defined.
>
>Here's the pseudo code
>
>'set the explicit sampling order to: level/row 2, level/row 1, level/row 3
>Dim strMyOrder As String
>
>strMyOrder = "2 1 3"
>
>Set List1.Order = ExplicitOrder( strMyOrder )
>List1.Reset
>
>
>Another possible approach is to first define a list in the experiment (it can just live in the Unreferenced branch of the tree) with all the exemplars you need and then use calls to List.GetAttrib( level, attribute ) to read the data off of that list and write out a tab delimited file in the order you want to use.   Then you have another list that you actual run which is setup to load itself from the file that you have written out.   This is particularly useful when you have a paradigm in which you want to perhaps change the stimulus presentation set or order of later blocks based on some subject specific performance criteria computed during earlier blocks.   
>
>In general I'd say the former approach is less work overall.
>
>Hope that gives you some ideas.
>
>Cheers,
>Tony
>
>Anthony P. Zuccolotto
>President and Chief Executive Officer
>Psychology Software Tools, Inc.
>2050 Ardmore Boulevard
>Suite 200
>Pittsburgh, PA 15221-4610
>Phone     412-271-5040
>FAX       412-271-7077
>Email     anthony.zuccolotto at pstnet.com
>Internet  http://www.pstnet.com
>
>
>  
>
>>-----Original Message-----
>>From: eprime at mail.talkbank.org [mailto:eprime at mail.talkbank.org] On Behalf
>>Of Robert Dörr
>>Sent: Wednesday, June 15, 2005 11:04 AM
>>To: eprime at mail.talkbank.org
>>Subject: Re: Pseudo-Random Lists
>>
>>Thats exactly the point. Generating an pseudo-random array of integers
>>which satisfies my conditions is no problem. But so far, I can't use
>>this array to control the selection of the ListObject.
>>
>>
>>
>>Katz, Lena B. wrote:
>>
>>    
>>
>>>Just script it, using random number generation(RAND, i think?), and check
>>>      
>>>
>>whether your restrictions require a "redraw"... I think this involves
>>removing the list object completely, as I'm not seeing any quick way to
>>choose which level you draw using the ListObject itself.
>>    
>>
>>>Hope this helps!
>>>
>>>Lena
>>>
>>>	-----Original Message-----
>>>	From: eprime at mail.talkbank.org on behalf of Robert Dörr
>>>	Sent: Wed 6/15/2005 7:56 AM
>>>	To: eprime at mail.talkbank.org
>>>	Cc:
>>>	Subject: Pseudo-Random Lists
>>>
>>>
>>>
>>>	Hi
>>>
>>>	Is there any way to influence the random list order on run time to
>>>      
>>>
>>get a
>>    
>>
>>>	random order with certain restrictions?
>>>	I. e., if I have "red" and "green" trials and I want to ensure, that
>>>	there is no sequence of more than three "red" trials.
>>>	So far I am using some external programming stuff to create text-
>>>      
>>>
>>files,
>>    
>>
>>>	but it would be better to do this in E-Basic on run time.
>>>
>>>	Thanks
>>>	Robert Dörr
>>>