Choosing the right eye tracking sampling frequency

  by Mirta Mikac, Product Manager

Technicalities behind eye tracking

The first thing to understand when deciding on the eye tracking device is the operating unit of measure and sampling rate in eye tracking. Gaze points are standard units of measurement in eye tracking, where 1 gaze point equals 1 raw eye tracking sample. On the other hand, the sampling rate is the frequency with which the eye tracking device registers the position of the eyes per second measured in Hertz (Hz). Currently, on the market, there is a wide range of eye tracking devices available, starting with a sampling rate of 30 Hz to 2000 Hz. The 30 Hz, eye tracking device captures 1 data point every 33 ms, the 60 Hz device every 16 ms, the 120 Hz device every 8 ms, and the 250 Hz device every 4 ms. One may think that only eye tracking devices with a higher sampling frequency are reliable due to their higher temporal resolution, but a few crucial factors are often overlooked. Hence, this blog post aims to familiarize you with the relationship between eye movements and sampling frequencies to make choosing a suitable eye-tracking device smoother.

The relationship between eye movements and sampling frequencies

Since different types of eye movements have different duration, the sampling frequency defines what kind of eye movements can be accurately measured at which sampling rate. Saccades are rapid eye movements lasting approximately 10 to 40 ms, depending on their traveled distance. For detecting such eye movement, a higher sampling rate is needed. The same is with smooth pursuit. Smooth pursuit is another way of shifting the gaze where eyes are fixated on a moving target. Such eye movement is also rapid – 90 to 150 ms, but it is limited to the angular velocity of the moving target. On the other hand, fixations are longer (approx. 100-800 ms), allowing the gaze to be maintained longer in a particular location. A higher sampling rate does not provide better accuracy for identifying fixation location (sustained gaze on a specific area with the purpose of the visual system taking in detailed information about what is being looked at); however, with a higher sampling rate, it is possible to identify more precisely fixation onset and offset.

How to decide what I need?

While eye tracking devices with higher sampling frequencies provide better temporal resolution, they are also more expensive and are not a prerequisite for obtaining reliable data. The data’s reliability depends on the research question, research rigor, and the type of eye movements analyzed. For instance, in the simpler studies within the UX domain where attention levels are analyzed with fixation metrics and the main interest is the fixation location, our LogicOne 60 Hz or 120 Hz are more than enough to obtain reliable insights. These eye trackers also suffice pupil size research due to pupils’ slower response rate. For the more complex studies within the academic domain focusing on the analysis of shorter eye movements and their characteristics (e.g., onset and offset), an eye tracking device with a higher sampling frequency is required, e.g., LogicOne 250 Hz. However, a general rule of thumb is to use the same sampling frequency as other researchers within the research field.

Need more help with deciding which sampling frequency is the best for you? We’d be happy to assist you!

Andersson, R., Nyström, M., & Holmqvist, K. (2010). Sampling frequency and eye-tracking measures: how speed affects durations, latencies, and more. Journal of Eye Movement Research, 3 (3), 6-12.

Rayner, K. (1998). Eye Movements in Reading and Information Processing: 20 Years of Research. Psychological Bulletin, 124(3), 372-422.