Combining random subsets

OK, I read through the manual and searched the forum and did not find an answer to this, so here goes...

Sec. 5.17, p. 66 of my 2008.1 edition of the manual explains how to randomly present n of N items (as also explained in a thread on the DirectRT forum at [URL]http://www.empirisoft.com/Support/showthread.php?t=458[/URL]). That's all very good, but now suppose I want to combine a randomly selected subset of items with another set of items, all in random order? E.g., we want subjects to first study, say, 100 pictures in random order. Then we want to show a randomly selected 50 of the studied pictures, along with 50 unstudied pictures, in random order, and have subjects indicate which ones they remember from earlier. MediaLab works really well for presenting the trails here so I want to use it instead of my usual programming system, but I don't see how to do this sort of random recombination in MediaLab, and didn't find anything else searching the forum. Would we need DirectRT for this?

Thanks,
-- David McFarlane, Professional Faultfinder

It sounds like the hardest part of what you need done would be finding a way to get the 50 randomly selected images out of the 100 studied images. It would be much easier to specify these 50 selected images as constants, rather than have them randomly selected from the batch of 100 studied images.

If you made that small change to what you need completed, all of your other requirements would be possible using the WGR and BGR values. Both MediaLab and DirectRT would be able to successfully complete this task.

Thanks. I could ask the researchers, but that may not fit their experimental design. However, I took another look at DirectRT and got this to work by combining stimulus lists with rwg (or wgr, as it's labeled in the input file, I will ask in a later post about the inconsistent terminology).

-- David McFarlane, Professional Faultfinder

Ah - yes, I forgot about the usefulness of stimulus lists.
Using a stimulus list with the WGR and BGR features should be able to complete the task.

Let me know if you need further assistance with it.

[quote=newell;8115]It would be much easier to specify these 50 selected images as constants, rather than have them randomly selected from the batch of 100 studied images.[/quote]
I put this to the investigator, and she was very pleased with that solution, especially since then we could do the task plus questionnaires all in MediaLab without dropping into DirectRT. So I was working harder than I had to :)! Still, I like learning these things, and knowing this might come in handy later. I prefer to let research questions dictate the tools rather than the other way around.

-- David McFarlane, Professional Faultfinder