detect correct rt:text (open ended) responses?

Does anybody know how to do an anagram task that requires a correct answer to proceed? Or will put questions that are responded to incorrectly or omitted into the queue to be tried again? The require a correct response option only works with a single character. The anagram task shown as an example will let you put whatever you want in there and continue. I suppose what I am looking for is some sort of if-then logical structure like in most programming languages. Any help would be much appreciated.

[I][similar issue, added by moderator 9-23-08][/I]
[QUOTE]I am attempting to use DirectRT for a psychology experiment. What I need the program to do is to require a correct response to math problems, meaning that the participant needs to type in the correct answer before moving on to the next question. However, I cannot get DirectRT to require a response that is multi-keyed; I understand how to require one correct key before moving on, but even when I introduce multiple responses to the trial, I cannot get the program to require the correct response. I also understand how to allow for a free response with DirectRT, but as far as I can tell you can't require a free response to be "correct." Do you have any suggestions for me? Is MediaLab able to require such a correct response?[/QUOTE]

We upgraded the text response function in the 2006 version--you can now add skipping and jumping to trials based on open-ended response contingencies like this. It works just like it does for closed-ended responses. You can define correct and incorrect text responses and set up skips and jumps as if they were key codes (e.g., rt:text,cat<100,dog<101). There are a couple nice math test samples attached here to illustrate.