I have an experiment that displays a drifting grating continuously for 60s. The gratings are imported into psychopy as images since they need to be specific colors every grating is presented for 50ms. At the start of the block I reset the clock using kb.clock.reset() then during every image presentation, I check for any keypresses and save those key presses and the reaction times out using key.name and key.rt. I had someone piloting this experiment recently. They completed 20 blocks on this where the name and rt were recorded into a csv file normally however on block 12 the reaction time was out of order. Meaning on the 3rd key press the rt was 11s then on the 4th key press, the rt was 9s then 22s then 30s then 19s etc. I am not sure what could be causing this kind of behavior. I haven’t been able to replicate it. Any ideas on why this happened and how I can avoid it from happening in the future?