Timing Test
Many labs running experiments underestimate timing problems, it is assumed that stimulus is presented on the screen/speaker when the trigger for beginning of stimulus is registered in the acquisition machine, however, this is not necessarily true for many reasons.
Depending on the sound/video card you are using, there might be some offset between the two, jitter due to vertical refresh for visual stimulus, and drift if the 2 clocks of presentation and acquisition are out of sync. There is additional delay for sound traveling from speaker to ears of participant at 344meter/second at room temperature or close to 3ms for each meter.
To correct for any misalignment, you can use photocells that detect stimulus presentation and save timing to compare with trigger time stamps. For auditory stimulus, you can use a microphone that feed signal back to acquisition machine.
It is recommended to monitor 3 measures:
Offset: Any delay between when stimulus is presented and when it is registered, this number should be cte, and can be easily accounted for during analysis.
Jitter: Due to refresh rate, low end video/sound card, difference could varies for each stimulus that could cause some smearing when averaging, best to keep this as small as few ms.
Drift: This is the worst of the three, and should be dealt with. When difference keep drifting, timing problem are then a serious one.
Many labs running experiments underestimate timing problems, it is assumed that stimulus is presented on the screen/speaker when the trigger for beginning of stimulus is registered in the acquisition machine, however, this is not necessarily true for many reasons.
Depending on the sound/video card you are using, there might be some offset between the two, jitter due to vertical refresh for visual stimulus, and drift if the 2 clocks of presentation and acquisition are out of sync. There is additional delay for sound traveling from speaker to ears of participant at 344meter/second at room temperature or close to 3ms for each meter.
To correct for any misalignment, you can use photocells that detect stimulus presentation and save timing to compare with trigger time stamps. For auditory stimulus, you can use a microphone that feed signal back to acquisition machine.
It is recommended to monitor 3 measures:
Offset: Any delay between when stimulus is presented and when it is registered, this number should be cte, and can be easily accounted for during analysis.
Jitter: Due to refresh rate, low end video/sound card, difference could varies for each stimulus that could cause some smearing when averaging, best to keep this as small as few ms.
Drift: This is the worst of the three, and should be dealt with. When difference keep drifting, timing problem are then a serious one.