Millisecond Forums

Stimulustimes: Stimulus doesn't disappear in some trials

https://forums.millisecond.com/Topic15723.aspx

By S&A15 - 3/10/2015

Hi,

at the moment I’m tryingto program the Stroop Task with the help of Inquisit. I used the Stroop Task scriptsin the Inquisit library as a basis, but I changed some elements. The problem isas follows:

I want to show the coloredwords only for 2000 ms. But the window center for participants to respondshouldn’t be restricted. Participants should have the possibility to answereven if the stimulus has disappeared. This is an example for a trial I programmed:

 

<trialredincongruentblue>

/ pretrialpause = 1500
/ stimulustimes = [0 =fixation; 500 = empty; 500 = redincongruentblue, redreminder, greenreminder,bluereminder, yellowreminder; 2500 = empty; 2500 = redreminder, greenreminder,bluereminder, yellowreminder]
/ correctresponse =("j")
/ validresponse =("d", "f", "j", "k")
/ responsetime = 500
</trial>


<text empty>
/items =("+")
/position = (50%, 50%)
/fontstyle =("Arial", 40%)
/txcolor = white
</text>


<textredincongruentblue>
/ items =("red")
/ color = blue
</text>


There is a fixationcross at the beginning of the trial, which is covered by a white cross after500ms. Simultaneously the stimulus and the reminders appear. After 2500 secondsthe stimulus is covered again by the white cross (disappears). The remindersremain.

The problem is thatthe stimulus disappears most of the time, but in some trials the stimulus doesn’tdisappear. I don’t know in which trials (the trials always change) and I don’tknow the reason for it. Can someone help me?

 

Thanks a lot!

Alexa
By Dave - 3/10/2015

If a participant responds prior to the display of the "empty" stimulus at 2500ms, the trial will terminate immediately (if you don't want that set its /responseinterrupt to frames). If some of the other stimuli are set to /erase=false, they will remain on-screen.

Since you did not provide the code beyond the excerpt, the above is only a guess, of course.
By S&A15 - 3/18/2015

Hello Dave,

thanks for your response!

It's okay that the trial is terminated directly, if a participant responds before the display of the "empty" stimulus at 2500ms. The participants should response as fast as possible, when they see the stimulus (colored word).

Unfortunately I still have the same problem: In some trials of the main experiment the stimulus doesn't disappear after 2500 ms, although it should. In the 20 pratice trials at the beginning of the experiment it works.

I added the complete code to this post. The stimuli and the instructions are written in German. Nevertheless I hope that you can understand it.

I would be pleased to hear from you.
Alexa
By Dave - 3/18/2015

Thanks for the script. What happens is the following:

If you respond in an instance of a given <trial> element *prior* to the stimulus "leer" being shown at 2500ms, the *next* instance of that same trial element will behave wrongly and *not* display "leer" at 2500ms. Basically, Inquisit fails to properly reset the trial's stimulus presentation sequence if that sequence was interrupted by an early response (i.e., prior to all scheduled stimuli having been displayed).

To see this for yourself, temporarily add

<block debug>
/ trials = [1-10 = greenincongruentred]
</block>

to the script and run just that block. Respond early in the 1st trial and observe that the 2nd trial will not display "leer" at 2500ms.

To avoid the issue, you can force a reset of the stimulus presentation sequence via

<trial greenincongruentred>
/ ontrialend = [trial.greenincongruentred.resetstimulusframes(); ]
/ pretrialpause = 1500
/ stimulustimes = [0=Fixationskreuz; 500 = leer, greenincongruentred, redreminder, greenreminder, bluereminder, yellowreminder; 2500 = leer, redreminder, greenreminder, bluereminder, yellowreminder]
/ correctresponse = ("d")
/ validresponse = ("d", "f", "j", "k")
/ beginresponsetime = 500
</trial>

Add the above statement to <trial greenincongruentred>, run the <block debug> element again as detailed above and observe the difference.
By S&A15 - 3/29/2015

Hello Dave,

thanks your very much for your answer and for your help!!
Now everything works fine!!

Alexa