Millisecond Forums

Hi, 

Before I get to my question, I'd like to point out that I believe there is an error in calculating the /ConflictEffect expression the current ANT script in the Task Library: 
/ConflictEffect = expressions.meanrt_correctcongruent - expressions.meanrt_correctincongruent
SHOULD BE : /ConflictEffect = expressions.meanrt_correctINcongruent - expressions.meanrt_correctcongruent
At least, this is my reading of Fan's original computes. 

My question: I'd like to adapt the current ANT script to use facial stimuli instead of "arrows". Instead of a "left pointing arrows" there would be smiling face, and instead of "right pointing arrow" there would be a frowning face. Sounds like a pretty easy fix, however the problem I get into is that the flankers are dynamically defined for each trial, which is easy when you only have 1 item representing each stimuli (e.g. left arrow). As is stands, flankers are defined in each trial by the select.flankertarget "value" which assigns a value of 3 or 4 in order to get the 3rd and 4th items listed in the <stimuli> element.

<item stimuli>
/1 = "+"
/2 = "dot"
/3 = "right arrow"
/4 = "left arrow"
/5 = " "
</item>

I'd like to make the 3rd and 4th items of the stimuli element each refer to a pool of images, for example, 
/3 = "pool of smiling faces"
/4 = "pool of frowning faces" 

I'm pretty sure this is not possible, however, I'm having trouble figuring out how to go about I could easily make the "left arrow" item a pool of pictures instead of a static image/text.  
In an effort to limit modifying the script, I'm looking for a way to make the 3rd and 4th items of the stimuli element refer to a "list/pool of something" instead of a static image/text. 

Thanks in advance,
Stéphane