X3 120 eye tracker

Participants were invited to sit in an adjustable chair to make themselves comfortable at the experimental desk, at which there was a standard 1920 × 1080 p computer monitor. For optimal tracking, the eye tracker was placed in such a way so that the gaze angle (α) would not exceed 36° when the participant is located about 65 cm (26″) from the eye tracker. The allowable operating distance (from eye tracker to participant’s eyes) for the Tobii Pro X3-120 Eye Tracker is 50–90 cm (19.6–35.4″). In this study, as Figure 1 illustrates, the distance from the participant’s eyes to the eye tracker and monitor was approximately 60 cm, and the gaze angle (α) was around 35°.
Participant head movements would be performed unconsciously during the experiment, which could have an impact on data quality. In order to make sure that pupillary response data could be tracked properly, each participant must have at least one eye within the trackable area at all times. In order to avoid environmental auditory interruption, participants were equipped with standard headphones, which provided a suitable sound level with a consistent sound stimulus throughout the experimental procedure. The indoor temperature was controlled to 27 °C.

A Tobii Pro X3-120 Eye Tracker was used for data collection. The eye tracker was attached to a 22-inch Dell monitor that presented video stimuli. It has an average gaze position error of 0.5° and a spatial resolution of 0.2°. Eye movements were recorded binocularly at a sampling rate of 120 Hz. A table covered by black cloth supported the monitor at the infants’ eye level (approximately 55 cm from the floor). A Logitech HD webcam was placed below the monitor to allow the experimenter to observe the infants’ looking behavior. The webcam was concealed from the infants by black cloth with a small hole for the camera lens.

While children completed the psychomotor vigilance task, their gaze position and pupil diameter were recorded from both eyes via a Tobii X3-120 eye tracker mounted to the bottom of the monitor. We used the pupil diameter of the right eye for our analyses consistent with prior research (Unsworth and Robison 2016; left and right pupil sizes were highly correlated, r = 0.82, p = < .001; the results were qualitatively the same when the left eye was used). Only valid pupil data were included in the analyses (invalid pupil data due to missing data or blinks were excluded). From the pupil data, we extracted two measures. First, we measured the baseline pupil diameter by averaging the pupil values over the first second of the wait screen. We picked the first second of the wait screen given that the length of the wait screen varied randomly between 1 and 8 s. Second, we computed a task-evoked pupillary response by examining changes in pupil diameter in response to the appearance of the mole. To do so, for each trial, we calculated the pupil size change by subtracting the average pupil size over the 2000 ms time window post-target onset from the average pupil size over the 500 ms time window before target onset, consistent with prior research (Unsworth and Robison 2016 (link)).

Infant visual attention was measured in a lab set up for this purpose. Participants sat on the parent/caregiver’s lap while eye movements were recorded by a Tobii X3-120 eyetracker (Tobii Technology, Stockholm, Sweden) sampling at 120 Hz. Stimuli were presented on a screen placed ~60 cm from the participant using Tobii Pro Studio software. Before the start of each block of trials, participants completed a 5-point calibration.