4
$\begingroup$

For our fMRI study we want to present videos of 3 seconds duration. The videos have different frame rates:

  • 48 videos were recorded with 50 fps

  • 16 videos were recorded with 25fps

  • 64 videos were recorded with 30 fps

I'm now faced with the problem that the videos are not presented for 3 seconds. The videos are either extended or shortened:

  • 48 videos (50 fps): extended (maximum extension between +6ms to +10ms)

  • 16 videos (25 fps): shortened (maximung shortening between -17ms and -20ms)

  • 64 videos (30 fps): shortened (maximung shortening between -17ms and -20ms)

For this reason I assumed that the problem is caused by the different frame rates of the videos. But then I discovered that the extent of the extended or shortened presentation is changing from presentation to presentation and independent of the videos. So for example: Video1 is presented for 3006ms instead of 3000ms during the first presentation. During the next presentation Video1 is presented for 3008ms instead of 3000ms and so on.

Another problem is the presentation of a blank screen between each stimulus. It is set to range between 3000ms and 5000ms in varying steps of 250ms. Here I noticed an extended presentation ranging from +16ms up to 30 ms.

Questions

  • Any ideas what might cause this timing deviation?
  • Does this have to do with the hardware of the stimulus presentation computer?

The refresh rate of the stimulus presentation computer is set to 60Hz. The computer is connected to a projector for the fMRI measurements (but the timing deviations do also occur when I present the stimuli full screen on the stimulus presentation laptop).

$\endgroup$
  • 1
    $\begingroup$ How do you know that it's your video projection rather than your camera which is timing-inaccurate? Are you using a high-speed camera? $\endgroup$ – Fizz Jun 6 '18 at 19:26
  • $\begingroup$ I feel this question might be better suited to video.stackexchange.com as video editors and engineers there may be able to have a better idea $\endgroup$ – Chris Rogers Jun 6 '18 at 23:10
  • $\begingroup$ @Fizz: I'm not sure if I understand your question correctly. No, we didn't use a high-speed camera. But if I open the videos in a versatile video editing program like GSpot it shows that the videos are in fact 3 seconds. That is why I assumed that it is the projection of the videos that is timing-inaccurate. $\endgroup$ – Jana Jun 7 '18 at 7:34
  • $\begingroup$ @ChrisRogers: You are right, thank you for your suggestion! $\endgroup$ – Jana Jun 7 '18 at 7:35
  • $\begingroup$ How do you know "Video1 is presented for 3006ms instead of 3000ms during the first presentation. During the next presentation Video1 is presented for 3008ms instead of 3000ms and so on."? $\endgroup$ – Fizz Jun 7 '18 at 23:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.