Skipping when trials are randomized?
Skipping when trials are randomized?
[edited from support email]
I'm using DirectRT. On each trial, subjects watch a video and indicate whether or not they detect a stimulus. If not, they press one button and this skips them on to the next trial. If they detect the stimulus, they press a different button and another question is asked. I realised that this wont work when I randomise the whole thing because I will not know in advance what the next trial is. Is there a way to skip them out of the trial and still make it all random?
Is it the video that needs to be randomized across trials? If so, you could leave the trials in a constant order (i.e., not random) but randomly select the video stimulus on each trial via a stimlist. e.g., if you have 10 trials, create a stimlist called "videos.txt" (or whatever) with the 10 video clips and simply refer to it as $videos for the stimulus value on each trial. Voila--randomized clips with trials in a constant order for the purpose of skipping and jumping and such. If that won't work for whatever reaosn, let me know and we can figure out something else. My memory is telling me that I wrote a function recently to allow skipping or jumping n trials rather than to a specific trial id--that would also do the trick. That may have just been a dream though.
Hi I am having the same problem.
I designed an experiment in which subjects are requested to answer yes or not to a question. If they answer no, they will skip to the follow question. I followed all the instructions for skipping and it worked, until I randomized the order of the trials. After running a few trials DirectRt just stops, as the experiment has been finished.
How can I skip to the following trial, in case of answer no, when the trials are randomized? Attached is the sheet of my experiment.
Thank you
I'm having a related issue to a previous poster. I was wondering if the "skipping or jumping n trials" function exists yet? Thanks!
Is it the video that needs to be randomized across trials? If so, you could leave the trials in a constant order (i.e., not random) but randomly select the video stimulus on each trial via a stimlist. e.g., if you have 10 trials, create a stimlist called "videos.txt" (or whatever) with the 10 video clips and simply refer to it as $videos for the stimulus value on each trial. Voila--randomized clips with trials in a constant order for the purpose of skipping and jumping and such. If that won't work for whatever reaosn, let me know and we can figure out something else. My memory is telling me that I wrote a function recently to allow skipping or jumping n trials rather than to a specific trial id--that would also do the trick. That may have just been a dream though.
Hello,
I am experiencing a similar problem. In my case it is not the video that needs to be randomized, but the question order. I have the same video on several trials, and those trials need to be randomized. However, the trials also need to be paired - 2 by 2. For example:
Trials 1a, 1b, 2a, 2b, 3a, 3b -> 1b should always follow 1a, 2b should always follow 2a, etc. But the order of 1, 2, and 3 should be randomized.
Is there a way of linking the a and b trials?
Under different circumstances I would use the BGR, but in this case I need that to group questions related to different videos - the overall structure is this: Several videos, each associated with a group of questions, and the questiones need to come in pairs. The video blocks need to be randomized, and the questions within the blocks need to be randomized too (keeping the pairing though).
The only solution that comes to mind is to create a separate .que file for each video, and then randomize the .que files within the .exp file...
Thanks in advance!
