As part of my research I have recently collected eye-tracking data in (semi)dynamic contexts during simultaneous interpreting. For this I used Tobii Glasses 3 (50 Hz). This meant that participants were seated and they engaged in their tasks, while having the freedom to inspect several types of information available in their booths: notes, slides, speakers, slides, laptop, etc.. They also gazed towards the room in front of them and the seated participants.

Participants were seated on chairs that permitted rotation, they were able to move their head and body, as well as their hands.

I am particularly interested in the resulting saccadic measures computed for certain intervals. So far the maximum peak velocity of saccades raise some red flags, as these maximum peak velocity values range between 478 and 2650 degrees/second. 2650 degrees per second seems high, but I have not been able to find any literature that focuses on maximum peak velocity or that states some filtering procedure for this metric. I am wondering whether anybody has any experience with cutoff values of peak velocity and amplitude of saccades in (semi)dynamic experimental settings. So far I have not been able to identify this and I would like to avoid including bogus data into the analysis.

I have seen that 700-1000 deg/sec seems to be cut-off for static ET studies, but would this hold for dynamic studies?

Note: I am interest in saccadic metrics due to wanting to explore the cognitive load throughout the interpreting task. There is evidence suggesting that peak velocity decreases with increasing cognitive load in real-world, highly dynamic work tasks (e.g., Di Stasi et al., 2011).

Your help would be very much appreciated!

Thank you!

I have conducted extensive literature research but failed to find anything regarding my issue at hand.