varying image load time and RT/ ISI issues

I’m sorry to post a new thread here, but I’m still having issues with my change-detection experiment and I can’t seem to find the answers I need in order to get the study running sooner rather than later.

I’ve attached a sample of the experiment below. I need to present participants with a series of events which include the following trials: a 250 ms focal point (+), followed by a 250 ms .png image, followed by a 250 ms focal point (+), followed by a second 250 ms .png image. If participants have not responded during this time, the event should repeat (but ideally the timer should not reset). I was having issues with a “pause” on the last trial before the event restarted, but that seems to have been resolved in a previous post. Participants end the event when they detect the difference in the two images and respond

The problem that continues is that the presentation rate of each trial seems irregular, almost sporadic (each should be for 250ms). I attempted to remedy this (although i’m not clear on HOW it actually worked) by choosing the option to reset the timer on each trial. This seemed to make the presentation timing a bit more regular, but now I have to calculate the participants’ reaction time by counting in the output the number of trials they viewed (or iterations of the event) and multiplying this by 250 ms. However, as I believe it was mentioned in another thread, the cumulative time in the output is indicating that each trial is not being presented for exactly 250ms. Moreover, as the experiment progresses, the timing of the presentation seems to become more and more sporadic. How can I regulate this so that every trial is on the screen for exactly 250 ms. I can tell by running the study that this is not the case. It seems it would be due to the “stuff” going on in between trial presentation (loading the image), but either way, I need to regulate this so that my RTs are accurate. I tried to account for the load time of each image, but this varies as well - Each time I “preview” the same image, I get a different load time. Where is this time accounted for? It seems like it’s being taken out of the 250 ms for which the trial should appear on the screen, but the output seems to indicate that it is being tacked on elsewhere. How do I adjust the settings so that each trial is presented for exactly 250 ms and my output is accurately reflecting participants’ RTs?

I’m running the experiment on a 2GHz MacBook with built in 13 inch display. Some assistance is needed asap please!

Thank you.

sample.sl4 (161 KB)

The way you have your experiment set up is incompatible with some of your other requirements, such as your RT’s being relative to the 1st and your issues with timing.

Instead of restarting your trial over and over again (which is the cause for the time discrepancies), it would be best to pick a max. time limit for your participants to respond. It will include all the events within a trial. Instead of reduplicating everything, you could put in a copy of what you have already done in each trial. In the Feedback tab, everything stays the same expect you need to take out the repeat trial option.

So I need to copy the 4 events and paste them below multiple times (to go on as long as I can predict participants should need) and link them into the same trial? This is less than ideal, but if you think it will finally take care of the timing issues I’m having, then I guess that’s what I’ll do.

Unfortunately it seems this is the best way to do it.

Thank you Monika. I have followed your suggestions and made duplicates of the trials, but unfortunately it now takes over ten minutes to open a study and get it running! I just sit there and wait and wait and wait for it to get the experiment open and then wait forever more to get it running. I’ve tried different options (under the Experiment Options menu); Opening each stimuli before it needs to run and opening all the stimuli before running, but neither seems to make a difference. Are there settings that can be modified so that the experiment will open in a more efficient manner?

Can you post your experiment again? I want to see if there is anything that can be done regarding the long start up. A future release of SuperLab, 4.0.8, maybe able to help with this issue.

I am trying to load the experiment, but it says that the file size exceeds the limit (the file is 2.7 MB). Presumably this is the reason it takes so long to open as well. Is there something that can be done to reduce this? Presently each image event that must be loaded is 248 KB .png file.

What version of SuperLab are you using? Go to the Help menu, then choose About SuperLab 4.0. The current version is SuperLab 4.0.7b.

I’m currently running SuperLab 4.0.7b for Mac.

I will contact you by PM to discuss this issue.