Random selection in Go NoGo task with constraint on first trial-type followed by unconstrained
 
 
 
PST User Forum
Home       Members    Calendar    Who's On
Welcome Guest ( Login | Register )
        



Random selection in Go NoGo task with... Expand / Collapse
Author
Message
Posted 8/19/2008 6:11:38 PM
Forum Newbie

Forum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum Newbie

Group: Forum Members
Last Login: 9/12/2008 9:38:41 PM
Posts: 2, Visits: 16
Hi All,

Does anyone know how to constrain one type of trial condition (e.g., Go trials) for the first trial, while randomly select conditions (both Go and NoGo) for the remaining duration of a block of trials. For example, in a Go NoGo paradigm with 5 types of Go trials and 1 type of NoGo trials, I want the first trial to always be a randomly selected Go trial, while the remaining trials are randomly selected from both Go and NoGo trials. In addition, I need to make sure that the ratio of Go to NoGo trials is 5 to 1, despite having fixed the first trial type. Your suggestions and strategies are much appreciated.

Thank you,

Greg



Greg
Post #1972
Posted 8/21/2008 4:35:18 PM


Forum Guru

Forum GuruForum GuruForum GuruForum GuruForum GuruForum GuruForum GuruForum Guru

Group: Moderators
Last Login: 10/30/2008 3:23:17 PM
Posts: 127, Visits: 924
Hi Greg,

This type of constraint can typically be handled by nested List objects. Nested Lists allow E-Prime to sample conditions independently. In the case of your experiment, you could place all of your Go trials in one nested List and all of your No Go trials in another nested List, and then nest each as necessary (i.e., by entering the appropriate List name under the Nested column in the TrialList).

What this will allow you to do is create two TrialList objects for each block in your experiment. So, for example, say you have 30 total trials in your experiment (25 Go, 5 No Go). On your BlockProc, you would place a TrialList that runs only 1 trial. In the Nested column, you would enter the name of the List object holding your Go trials. This way, E-Prime will still randomly select a Go trial (i.e., rather than having to list a specific condition for the initial trial). Then, you would place another TrialList immediately following the first that will run 29 trials. This List will have 2 levels. One level will nest the Go List, while the other will nest the No Go List. The Weight for the Go level will be 24, while the Weight for the No Go level will be 5. This way, you will maintain your 5-to-1 ratio, but force a Go trial to start the block.

For more information on using nested List objects, please see section 2.5.1.2 of the E-Prime User's Guide.

I hope that this has been helpful. Please let me know if you have any further questions.


- Matt
PST Technical Consultant
http://pstnet.com
Post #1980
Posted 8/23/2008 3:07:49 PM
Forum Newbie

Forum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum Newbie

Group: Forum Members
Last Login: 9/12/2008 9:38:41 PM
Posts: 2, Visits: 16
Matt,
Thanks a lot. Your solution worked like charm.
Greg


Greg
Post #1982
Posted 8/26/2008 12:41:41 PM


Forum Guru

Forum GuruForum GuruForum GuruForum GuruForum GuruForum GuruForum GuruForum Guru

Group: Moderators
Last Login: 10/30/2008 3:23:17 PM
Posts: 127, Visits: 924
Greg,

I'm glad to hear that this solution helped. If you have any other questions, please let me know.


- Matt
PST Technical Consultant
http://pstnet.com
Post #1997
« Prev Topic | Next Topic »


Permissions Expand / Collapse

All times are GMT -5:00, Time now is 7:10pm

Powered By InstantForum.NET v4.1.4 © 2008
Execution: 0.063. 16 queries. Compression Disabled.