Working Memory Task

[LRL CSU]

I am interested in finding or developing a measure of working memory.

We want it so that participants hear and repeat words they hear via headphone, and to be able to increase the number of words if they get all of them correct (e.g. an experimenter presses a button to increase number of words, etc.).

I am wondering if there is anyone that already has a similar experiment that they would be willing to share.

Thanks

[Indra]

maybe you can try with a task called N-Back.
It's explained in a lot of published papers and it's really easy to develop it with superlab.
It was made for visual presentation of words, but you can develop an auditive task.
It's a good mesure of working memory.

[Patrick]

I am working on a similar task, but non-audio. The task I am trying to design is based on research by Shah and Miyake, in which the participant is asked to identify whether or not a rotated letter is normal or mirror-imaged. In addition to the recognition task, participants are asked to remember the orientation of the arrow and the order in which it appeared in a series of 2-6.

I am just starting this and am at the point of figuring out how to make the buttons work to identify whether a letter is normal or mirror-imaged.

-Patrick

[Hank]

If you haven't seen our letter rotation experiment, check it out here: http://community.cedrus.com/showthread.php?t=46

[lobserve]

Here is a working memory experiment I created. It uses text not sound, provides feedback but does not branch depending on correct or error response (I can't make that work yet :confused: ). Only the experiment file is needed. If anyone can add conditional branching to it please post a revision for me to look at. Enjoy :)

[Cigdem Oenal]

Hi!
What does 3\8\5\7 in stimulus list mean? Could you please explain the logic behind it?
thanks!

[robinhancock]

maybe you can try with a task called N-Back.
It's explained in a lot of published papers and it's really easy to develop it with superlab.
It was made for visual presentation of words, but you can develop an auditive task.
It's a good mesure of working memory.
Indra-I see that the n-back paradigm is easy to implement in Superlab, but is it easily to score from the output Superlab gives? I am still using the demo version (waiting for the full version to arrive!).

[Indra]

Yes, it was easy to score. You must take as scores RT, comision errors and omision errors. All of this it's present on the output file.

[robinhancock]

Thanks,

Robin

[AlexC]

I was looking though the Help contents today, and I believe that the relative jump function could help with making a working memory task, such as lobserve put out, I haven't tried it yet, but I believe that it could also use an absolute jump to take a participant out of the experiment if they got more that two incorrect. Just thought I would let you all know.

Alex

[CSU_LRL]

Thank you