Randomization Issues

I'm trying to run a complex randomization in DirectRT (2010). The basic design:
1) Participants see 12 attributes about a candidate on one screen (drawn in a random order from a stim list)
2) They see information about four candidates in total and candidate presentation is randomized (using WGR)
3) After all four candidates have been presented, participants are told to think carefully about the merits of the candidates (a simple screen with just text).
4) Participants are presented with a screen containing ALL of the information about all of the previous candidates (in the same random order that they were previously seen. (using the ? function, for a total of 48 attributes on the screen plus the four candidate names)

I'm running into two issues. First (and the smaller issue) occurs at step 3. If I include the instruction and then follow it with the information about all the candidates, DirectRT will not display the data at all.

The second issue (and much more frustrating) is that DirectRt does not seem to be redisplaying the random data correctly (step 4). Even when I don't randomize the order of the candidates and take out the instruction (step 3), when I randomly recall information from a cell (say, ?b1t1s1) it seems to take the information from the trial AHEAD of it (e.g. ?b1t2s1). The result is that the final screen, which displays all the data (step 4) displays the random data from trial 2 when it should show trial 1, trial 3 when it should show 2, trial four when it should show 3, and most importantly, trial 2 again when it should show trial 4!


The result is that the information about one candidate is displayed twice, so one candidate is always not displayed. Furthermore, the order that all the attributes are displayed is NOT the same as the first presentation. Hmm.


I'm not sure if this is a bug in direct RT or my randomization is flawed. I've been grinding away at this problem all weekend and I'm actually at a bit of a time crunch. I'm running the experiment on 2/8, and if I can't get this last step to work I'm going to have to improvise.


I've attached the .csv, the stim files, and the style file in a zipped folder.

I've figured out a workaround, even though there doesn't appear to be a direct solution to my problem. If you're having issues with randomization, it would seem that you have to have a 0 block before your first block with random stimuli. If you try to recall those stimuli later using the ? function (without a block 0), it will not recall the correct stimuli otherwise.

It would be nice if this bug were fixed in an update.