|
vz29
|
|
|
Group: Awaiting Activation
Posts: 72,
Visits: 192
|
Hi,
I'm interesting in presenting a question (either likert or as a 'yes' or 'no' via the trial element) about a stimulus and then presenting a different question/trial about the same stimulus. I would then repeat the process for another stimulus and do this 258 times.
So the procedures are ask Ps to make judgments (a through c) on stimulus (n through 258) with all judgments about a particular stimulus presented together, but the stimulus selected randomly (without replacement).
What is the most straightforward way to implement this task?
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 13K,
Visits: 108K
|
Use <values> and a <list> element for item selection:
<values> / itemnumber = 1 </values> <block myblock> / trials = [1-4=sequence(q1,q2,q3)] </block> <list itemlist> / items = (1,2,3,4) / selectionmode = random </list> <text mytext> / items = ("A", "B", "C", "D") /select = values.itemnumber </text> <likert q1> / ontrialbegin = [values.itemnumber=list.itemlist.nextvalue] / stimulusframes = [1=mytext, q1text] / anchorwidth = 5% / position = (50%,75%) </likert> <text q1text> / items = ("Question 1") / position = (50%,25%) </text> <likert q2> / stimulusframes = [1=mytext, q2text] / anchorwidth = 5% / position = (50%,75%) </likert> <text q2text> / items = ("Question 2") / position = (50%,25%) </text> <likert q3> / stimulusframes = [1=mytext, q3text] / anchorwidth = 5% / position = (50%,75%) </likert> <text q3text> / items = ("Question 3") / position = (50%,25%) </text>
|
|
|
|
|
vz29
|
|
|
Group: Awaiting Activation
Posts: 72,
Visits: 192
|
Thanks!
It works, but I had to modify it because instead of presenting text as stimuli to-be-judged, I'm presenting images.
Moreover, the set of image stimuli to draw from is large (400+) and so it is not practical to list it in the way specified (e.g., "image1, image2, image3....image400").
So, I modified the script as follows:
<values> / itemnumber = 1 </values>
<list itemlist> / items = ("f_ad") / selectionmode = random </list>
<picture f_ad> / select = values.itemnumber / items = f_pics / size = (50%, 50%) / position = (50, 50) </picture>
<item f_pics> /1 = "image_1.jpg" /2 = "image_2.jpg" /3 = "image_3.jpg' </item>
Although the script runs, and for each stimulus, the three questions are presented in sequence, the same ad is selected every time the experiment is run, and the ad is not replaced in the subsequent trial.
If I remove the quotes from the list element so that it reads as follows:
<list itemlist> / items = (f_ad) / selectionmode = random </list>
In this case, no trials are presented (although the experiment does run). I'm able to see that there is an error (again, even though the experiment does run) that reads "Unable to initialize <picture f_ad> item number 0." V
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 13K,
Visits: 108K
|
> It works, but I had to modify it because instead of presenting text as stimuli to-be-judged, I'm presenting images.
Actually works exactly the same way. Your syntax is botched in various places, though, particularly your <list> element.
#1: / items = ("f_ad")
is invalid syntax and makes no sense (as a matter of fact, it would be circular).
#2: The list is expected to return a numerical value, i.e. the item number, to be stored in the respective value used for selection.
|
|
|
|
|
vz29
|
|
|
Group: Awaiting Activation
Posts: 72,
Visits: 192
|
your response wasn't really helpful...
|
|
|
|
|
vz29
|
|
|
Group: Awaiting Activation
Posts: 72,
Visits: 192
|
To be clear, the set of image stimuli I'm drawing from is large (400+). So it is not practical to use the following syntax you provided. Here's a variation of it:
<list itemlist>/ items = (image1, image2, image3 )/ selectionmode = random</list>
Is this the only way to specify the particular items? Seems cumbersome with 400+ images.
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 13K,
Visits: 108K
|
That <list> element is still broken (and does not correspond to the example I gave). But in any case, yes, you *must* list all external files somewhere in your script at least once. There are good reasons for this:
#1: Inquisit must make sure that all the given files actually exist -- otherwise your experiment might crash at some point at runtime.
#2: Inquisit will try to shove as much of the images into the system's RAM when it parses the script. Performance-wise this is far better than pulling files from disk one at a time at runtime. Disk access is slow, so you'll want to minimize it as far as possible.
|
|
|
|
|
vz29
|
|
|
Group: Awaiting Activation
Posts: 72,
Visits: 192
|
I'm able to display the images.
|
|
|
|
|
Dave
|
|
|
Group: Administrators
Posts: 13K,
Visits: 108K
|
> I'm able to display the images
?
|
|
|
|
|
vz29
|
|
|
Group: Awaiting Activation
Posts: 72,
Visits: 192
|
I got the script to work. Thanks!
|
|
|
|