I am currently running an experiment in which a fixation point is presented for 500ms, followed by a visual stimulus, followed by a mask. My stimulus files are drawn such that the fixation point always stays on the screen, so my stimulus files are as follows:
Fixation point alone, 500ms
Fixation + Visual stimulus, 200ms
Fixation + Mask, 200ms
When I present them in sequence, the fixation point, rather than appearing a constant, solid, point on the screen flashes or “flickers” with each new file presentation. I figure that this has to do with the fact that SuperLab has to draw each stimulus separately, but I have run experiments on older versions of SL that are nearly identical and I have not had this problem. I have also tried adjusting the memory so that SuperLab loads all files at the beginning of the experiment, and that doesnt seem to work either. Any ideas?
What happens if you have three “fixation point only” stimuli and present them sequentially? Do you still get flicker?
Also, does the background color setting you have chosen match the background of the stimulI (I don’t see why that should matter, but if the screen goes to the background color while the display is being written, maybe it would help if they matched).
Is it possible for you to have blank screen intervals between the stimuli? Maybe that would help. Or, you could implement the fixation-stim-stim+mask sequence as a movie–a kind of brief “slide show” but with more precise timing and no “fades” or other slick effects.
[Philosophical point: much as been written about how modern GUI systems are or are not realtime capable, but the fact is that while for many things they have a great deal of difficulty maintainings, say, 1 ms timing, for certain things, notably low-level device interfaces and high-level multi-media processes, they have been highly optimized to have excellent realtime characteristics. So, it might be time to consider routinely tapping into that capability for T-scope style visual stimulus presentation.]
I’ve had problems with flickers too. What seems to be happening is that SuperLab takes time to erase the screen. To avoid this, what I’ve done is use a 1024x768 PNG that is empty except for fixation cross in the middle (note that these specific dimensions will change depending on monitor size and resolution). Because of the PNG’s size, it will paint over anything that is on the screen whenever it is presented. So, there is no need to specify “erase screen” when presenting this event. Note, that from the participant’s point of view, the screen is erased except for the fixation.
So, the events are:
Fixation cross (full screen PNG with cross in the middle)
Stimulus (uncheck the “erase screen” option in the event settings; this will just put stuff on th screen without erasing the fixation cross)
3)Fixation cross (again, full screen, and uncheck the “erase screen” option)
Of course, in order for this to work, your stimulus (masking or other) has to be painted at a location that will not obscure the fixation cross.
An easy way to make a full screen fixation cross is to make a small SuperLab script that puts a “+” in the center of the monitor; of course set the background color to whatever you want. Once the fixation is presented, just take a screen shot (press "apple + shift +3 on a Mac). The file will be saved to your desktop, and you can simply move it to the appropriate folder for your experiment.
I have had flickering problems as well with pictures, the only way I found to make the picture seamless is to un-check this box under the picture “settings” and “presentation options” this should make any flicker go away.