(no subject)

[Tony Zuccolotto]

Dear Dirk,

To vary the ITI you just need to add a Wait object at the start (or end)
of the Trial and set its Duration property to an attribute named "[ITI]"
(you probably tried/have this much of the solution already).   How you
proceed from here (i.e. to actually set the value of the ITI attribute)
depends on how controlled you need to be with respect to
choosing/assigning those values.   I will go through the two most common
approaches I have used in the past.

1) If you just want a random ITI between two points (say 50 to 250 msec)
then insert an Inline at the very beginning of the trial with the
following script.

	c.SetAttrib "ITI", Random( 50, 250 )

This will randomly choose a duration value and put it in the Context (so
the Wait object can resolve the attribute reference).

2) If you want more controlled randomization (e.g. randomly chosen
delays selecting one of only the values of 50, 100, 150, 200, 250 on
each trial) then you need to use "nesting".  

The basic approach is to type in the name of a new List object in the
Nested column of your trial list (e.g. "DelayList") and then hit Enter.
You will be prompted to create the new List - answer yes.   You should
then open the new DelayList object in the workspace and create a new
attribute named "Delay".   Create some new rows/levels and type in the
delay values you want to use.  Configure the DelayList to sample
randomly and set the weights appropriately for your needs.

You then go back up to your trial list and create a new attribute there
named "ITI".   For the values of ITI on each trial type "[Delay]".  

What happens on each trial is...
	* the trial list is sample and a specific row is selected
	* the Nested attribute is then examined and any other List that
is specified there is also then sampled.  
	* during sampling of the nested "DelayList" a row will be chosen
and a specific value will be assigned to the "Delay" attribute and it
will be put in the context.
	* after any/all Nested lists have been sampled, all other
attributes on the trial list are resolved and also put into the context.
Thus, during sampling the ITI attribute first resolves to "[Delay]".
Because this is also an attribute reference the system will see that and
then in turn resolve the [Delay] attribute to get the final value.
Ultimately the ITI attribute value will be assigned to one of the values
you specified on the DelayList.

Note that this solution requires no scripting at all and also allows you
to precisely control for equal occurrences of any single delay value.

Nesting is one of the most unique and powerful features of E-Prime and
using nesting effectively can be the key to eloquent solutions for many
experiment sampling interactions and needs.   It is often beneficial to
look closely at the NestingRT sample paradigm and read the nesting
related sections in the User's Guide (Page 35, and A-45 in the
Appendix).

Hope this helps.

Tony

*** DISCLAIMER: VIEWS EXPRESSED ARE MY OWN AND DO NOT NECESSARILY
REFLECT THOSE OF MY EMPLOYER *** 
Anthony P. Zuccolotto 
Vice President of Operations
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: Dirk Leube [mailto:dirk.leube at med.uni-tuebingen.de]
> Sent: Thursday, December 05, 2002 5:36 AM
> To: eprime at mail.talkbank.org
> Subject: (no subject)
> 
> Dear all,
> we would like to vary our interstimulus interval (ISI) randomly during
> the experiment. We can't see any simple solution to this problem
without
> the stimuli being coupled in a fixed manner to a certain ISI. Have you
> got any suggestions?
> Thank you very much for your support
> Dirk
> 
> --
> Dr. med. Dirk Leube
> Universitaetsklinikum
> Klinik fuer Psychiatrie und Psychotherapie
> Osianderstr. 24
> 72076 Tuebingen
> Germany
> 
> Fax 07071/29-4141
> 
>