URL of experiment:
Description of the problem:
Two years ago I created an online experiment on visual and auditory sequence timing abilities. My concern now is on visual sequence timing. So participants saw three visual sequences containing multiple intervals in each trial and judged which one had a different pattern. sequences are something like [2,2,4,1,3,1] where they get multiplied by either 230 or 250 or 270 ms to present different trials with different tempos. You can also see they are regular sequences (i.e. beat-inducing where you can tap your foot along with it if it is auditory!). Each interval within the sequence appeared on one side of the screen and after resting there for the duration of interest appeared on the other side of the screen.
def durationn(intervall):
start5=[1.5,0,0,0,0,0]
start6=[1.5,0,0,0,0,0,0]
start7=[1.5,0,0,0,0,0,0,0]
This code controls the start time and “c” in the picture is a number and goes up one by one so it performs the next interval. three lists inside the function is because sometimes the three sequences in a trial have 6 intervals sometimes 7 sometimes 8.
ran=randint(1,3)
def stop(List,a,b,c,ran):
if ran==1:
return num1_to_sec(List[a][b][c])
elif ran==2:
return num2_to_sec(List[a][b][c])
elif ran==3:
return num3_to_sec(List[a][b][c])
Again a, b, and c are numbers that increase one by one to go to next interval or next sequence.
def num1_to_sec(n):
return float(0.230 * n)
The functions num2_to_sec and num3_to_sec are similar with the difference that they multiply the interval with 250 ms and 270 ms.
After submitting my paper to a journal, one of the main concerns of reviewers is the timing accuracy of my sequences. This is what they’ve written:
“The visual stimulus presentation time intervals will not fall within the time periods of standard monitor refresh rates of 30, 60, or 120Hz (230ms = no, 250ms = yes for 60 & 120Hz, no for 30Hz, 270ms = no). As a result, the length of time the intervals are presented will not be consistent. This is especially problematic for the visual beat sequences as the beat may not actually be preserved, and may be part of the reason beat sequence scores were lower than non-beat sequence scores in the visual data.”
s
So now I want to know whether it’s true that the timings of my stimuli are not precise and if not or if timing errors are minimal how can I explain it, Where can I see what amount of errors I have?
If the timing of the first interval in the sequence gets lagged, the second would be delayed too and in turn third and fourth and so on?
Also, considering this was an online exp how different computers with different frame rates can affect the timing of stimuli?
If more information or code is needed please let me know.
Any help is highly appreciated.