Hi, I've been spending some time working on creating some ego/resource depletion measures lately and I'm hoping you can help me with a question I have about programming the stroop task in DirectRT...

I've programmed a 90 trial stroop task that takes somewhere around 8 minutes to complete. It presents 10 blocks of 9 trials randomized within each block, asking participants to strike specific keys to indicate red, green, or blue font color regardless of what word is presented. Half the blocks present word-color congruent (e.g., the word "red" in red font color) trials to tap automatic processing and the other half present word-color incongruent (e.g., the word "red" in green font color) trials to tap effortful/regulated processing. These types of blocks are presented in alternating order.

I've read about two ways to program this in DirectRT, one producing a more comprehensive/helpful response record than the other:

- The first, simpler method is to have DRT present a forward mask (i.e., XXXXX) for 1000ms, then the target word for 2500ms, then a backward mask for another 1000ms. Using this method, DRT records participants' first keystroke as either correctly identifying the font color or not, as well as the RT for the response. Primary analysis of this data would be based on the number of errors in font color identification made during the trials as a measure of regulatory fatigue/depletion/distraction. The RT scores would also be recorded, and I suppose they could be analyzed, but a fair number of them would need to be dropped because they'd be RT's for incorrect responses.

- The second, more comprehensive method is very similar to the above. Like the above method, when participants respond to the target word DRT would record their first keystroke, thus noting whether they initially made an identification error. However, DRT also has a feature that will recognize incorrect responses and present an error signal (i.e., the words "try again") below the target until participants provide the correct response. Therefore, like the above method, this method would allow us to analyze error rates; but it would also offer the opportunity to also analyze how much time it took the participant to correctly identify the target font color.

My question is about programming DRT to require the correct responses without crashing if participants don't respond. We have DirectRT v2002; I've gone through the manual many times and have carefully followed the instructions for coding DRT to signal an error and wait for a correct response. And when I do a test run it works perfectly, as long as I actually provide a response. However, if, for whatever reason, I do not provide a response to a trial word before the end of the 2500ms time window (if I'm not quick enough on the trial or if I'm briefly distracted or refuse to respond to the trial) DRT just ends the whole session. Why is that happening? Have you guys encountered this before? How can I get it to reject incorrect responses but still move on to the next trial if participants are slow or distracted (or refuse to respond to a trial)?

Think you can help me here?