Table 1.

List of some common terms used in this paper

Term	Definition
Pupil-Corneal Reflection (P–CR)	A video-based eye-tracking method. Gaze direction is calculated from the corneal reflection (CR) coordinate and the pupil (P) coordinate in the eye camera image, originally by subtracting CR from P. The P–CR method currently dominates the eye-tracking market (Section “Eye-tracking methods: Similarities and differences”).
Dual-Purkinje Imaging (DPI)	The DPI system is an analogue eye tracker that bases its estimation of gaze on the relative movement of an infrared reflection in the cornea (P1) vs. a reflection at the back of the crystalline lens (P4) (Section “Eye-tracking methods: Similarities and differences”).
Electrooculography (EOG)	EOG is a method used to estimate eye orientation from the difference in electrical potential between the front and back of the human eye (Section “Eye-tracking methods: Similarities and differences”).
Scleral search coils	Scleral search coil eye tracking involves attaching a silicon-enclosed copper wire onto the sclera of the eyeball. The participant is then placed in an oscillating magnetic field, and the amplitude of induction voltage can be taken to represent eye orientation with respect to the magnetic field (Section “Eye-tracking methods: Similarities and differences”).
Limbus tracking	Limbus trackers use the border between the iris and the sclera to track gaze and eye movements (Section “Eye-tracking methods: Similarities and differences”).
Retinal image-based tracking	Image processing of retinal features (such as blood vessels) are used to register very small eye movements (Section “Eye-tracking methods: Similarities and differences”).
Calibration	The process of mapping eye-tracker measurements to physical gaze direction or gaze position. This can be done e.g. using fitting procedures or by using physically and biologically plausible models or a combination of the two (Section “Calibration and accuracy”).
Gaze direction	The vectors from an eye, both eyes, or a cyclopean eye, which describe the line of sight.
Gaze position	The location of gaze in a measurement plane or space. Gaze is also termed point of regard.
Sampling frequency	For relevant eye trackers, the number of gaze direction or gaze position measurements made per second (Hz), determined by the eye camera or AD-convertor of the system. Sampling frequencies vary from around 10Hz in some web-camera eye trackers to 10000Hz in some recordings.
Accuracy	The difference between the true gaze position (assumed or instructed) and the gaze position reported by the eye tracker. Typically reported as error in degrees of visual angle where a larger error indicates poorer accuracy. Inaccuracy is also known as systematic error, while accuracy can be called trueness (Sections “Measuring data quality of eye-tracker signals” and “Calibration and accuracy”).
Precision	The reproducibility of a gaze position from one sample to the next, assuming a stable gaze position. Typically reported as RMS or STD (or both). Greater error indicates poorer precision. Imprecision is also known as random or variable error (Sections “Measuring data quality of eye-tracker signals” and “Signal properties and processing”).
Resolution	The just noticeable difference in a signal, which in eye tracking is measured as the smallest reliably detected eye movement (or rotation of an artificial eye) that an eye tracker can resolve (Holmqvist & Blignaut, 2020; Poletti & Rucci, 2016; Crane & Steele, 1985).
Data loss	The amount of eye-tracking data lost. It is the counterpart of what is commonly called tracking ratio or availability of eye-tracking data (Sections “Measuring data quality of eye-tracker signals” and “Signal properties and processing”).
System latency	The duration from when an actual eye movement is made, until the corresponding gaze sample is output by the eye tracker, made accessible for instance to affect change on a monitor in a gaze-contingent study (Sections “Measuring data quality of eye-tracker signals” and “Signal properties and processing”). Latency is sometimes referred to as the end-to-end delay or temporal accuracy (Reingold, 2014, p. 641).
Event detection, classification, and calculation	Usually refers to the segmentation of an eye-tracker signal into meaningful segments (or events) with a start, end, and duration, and calculation of event properties. Meaningful segments can be ‘fixation’, ‘saccade’, ‘smooth pursuit’, or other terms (Section “Fixation and saccade detection”). Terminology varies (Hessels et al., 2018).
Area of Interest (AOI)	A segment of a stimulus space (often defined by screen pixel boundaries in monitor-based eye tracking) that identifies a portion of the stimulus that is meaningful in the experimental design of a study (such as eyes and mouth areas in face perception, a pack shot in marketing research, or a target in a visual search experiment). In eye-tracking analysis AOIs allow for the calculation of commonly reported dependent measures; for instance, the number of times or amount of time gaze is within a specific AOI, or the number of transitions between AOIs (Section “Area-of-interest (AOI) measures”).