Brief Stimulus Presentation Durations

I am running an experiment where I want to present picture stimuli for extremely short durations (as short as 1 refresh rate). In the event editor I have a 1 ms time limit for presentation so that the target appears for 1 refresh rate. I have found a discrepancy between black and white drawings and color photographs. The drawings seem to appear for longer on the screen than the color photographs. I have tried changing many things but I can not reverse this effect (I reduced the file size of the color images so that they would be similar to the drawings, I have tried out various forward and backward masks, put in ISIs of various lengths between the target and masks, moved each event into separate trials, turned on/off erasing the stimulus). Furthermore the drawing time Superlab claims for all of the images is less than a ms. Does anyone have any insight into why there would be such a difference? And how can I keep a consistent presentation time?

It would be interesting if this was a perceptual difference, but I have a feeling I am missing something with the technology.

When using 1ms for a refresh rate, what will happen is SuperLab will round up to the next multiple of the refresh rate. SuperLab does not erase anything by default. So if the next event does not erase the screen the stimulus will be longer then 1 refresh rate. We suggest using a blank text event to erase the screen.

Adding a blank screen only heightens the difference between the two types of stimuli.

Did you change the time limit to 1ms?