Search

In this paper, we present a review of how the various aspects of any study using an eye tracker (such as the instrument, methodology, environment, participant, etc.) affect the quality of the recorded eye-tracking data and the obtained eye-movement and gaze measures. We take this review to represent the empirical foundation for reporting guidelines of any study involving an eye tracker. We compare this empirical foundation to five existing reporting guidelines and to a database of 207 published eye-tracking studies. We find that reporting guidelines vary substantially and do not match with actual reporting practices. We end by deriving a minimal, flexible reporting guideline based on empirical research (Section “An empirically based minimal reporting guideline”).

Zemblys et al. (Behavior Research Methods, 51(2), 840-864, 2019) reported on a method for the classification of eye-movements ("gazeNet"). I have found three errors and two problems with that paper that are explained herein. Error 1: The gazeNet classification method was built assuming that a hand-scored dataset from Lund University was all collected at 500 Hz, but in fact, six of the 34 recording files were actually collected at 200 Hz. Of the six datasets that were used as the training set for the gazeNet algorithm, two were actually collected at 200 Hz. Problem 1 has to do with the fact that even among the 500 Hz data, the inter-timestamp intervals varied widely. Problem 2 is that there are many unusual discontinuities in the saccade trajectories from the Lund University dataset that make it a very poor choice for the construction of an automatic classification method. Error 2 The gazeNet algorithm was trained on the Lund dataset, and then compared to other methods, not trained on this dataset, in terms of performance on this dataset. This is an inherently unfair comparison, and yet nowhere in the gazeNet paper is this unfairness mentioned. Error 3 arises out of the novel event-related agreement analysis employed by the gazeNet authors. Although the authors intended to classify unmatched events as either false positives or false negatives, many are actually being classified as true negatives. True negatives are not errors, and any unmatched event misclassified as a true negative is actually driving kappa higher, whereas unmatched events should be driving kappa lower.

Eye trackers are sometimes used to study the miniature eye movements such as drift that occur while observers fixate a static location on a screen. Specifically, analysis of such eye-tracking data can be performed by examining the temporal spectrum composition of the recorded gaze position signal, allowing to assess its color. However, not only rotations of the eyeball but also filters in the eye tracker may affect the signal’s spectral color. Here, we therefore ask whether colored, as opposed to white, signal dynamics in eye-tracking recordings reflect fixational eye movements, or whether they are instead largely due to filters. We recorded gaze position data with five eye trackers from four pairs of human eyes performing fixation sequences, and also from artificial eyes. We examined the spectral color of the gaze position signals produced by the eye trackers, both with their filters switched on, and for unfiltered data. We found that while filtered data recorded from both human and artificial eyes were colored for all eye trackers, for most eye trackers the signal was white when examining both unfiltered human and unfiltered artificial eye data. These results suggest that color in the eye-movement recordings was due to filters for all eye trackers except the most precise eye tracker where it may partly reflect fixational eye movements. As such, researchers studying fixational eye movements should be careful to examine the properties of the filters in their eye tracker to ensure they are studying eyeball rotation and not filter properties.

Detecting eye movements in raw eye tracking data is a well-established research area by itself, as well as a common pre-processing step before any subsequent analysis. As in any field, however, progress and successful collaboration can only be achieved provided a shared understanding of the pursued goal. This is often formalised via defining metrics that express the quality of an approach to solving the posed problem. Both the big-picture intuition behind the evaluation strategies and seemingly small implementation details influence the resulting measures, making even studies with outwardly similar procedures essentially incomparable, impeding a common understanding. In this review, we systematically describe and analyse evaluation methods and measures employed in the eye movement event detection field to date. While recently developed evaluation strategies tend to quantify the detector's mistakes at the level of whole eye movement events rather than individual gaze samples, they typically do not separate establishing correspondences between true and predicted events from the quantification of the discovered errors. In our analysis we separate these two steps where possible, enabling their almost arbitrary combinations in an evaluation pipeline. We also present the first large-scale empirical analysis of event matching strategies in the literature, examining these various combinations both in practice and theoretically. We examine the particular benefits and downsides of the evaluation methods, providing recommendations towards more intuitive and informative assessment. We implemented the evaluation strategies on which this work focuses in a single publicly available library: https://github.com/r-zemblys/EM-event-detection-evaluation .

We empirically investigate the role of small, almost imperceptible balance and breathing movements of the head on the level and colour of noise in data from five commercial video-based P-CR eye trackers. By comparing noise from recordings with completely static artificial eyes to noise from recordings where the artificial eyes are worn by humans, we show that very small head movements increase levels and colouring of the noise in data recorded from all five eye trackers in this study. This increase of noise levels is seen not only in the gaze signal, but also in the P and CR signals of the eye trackers that provide these camera image features. The P and CR signals of the SMI eye trackers correlate strongly during small head movements, but less so or not at all when the head is completely still, indicating that head movements are registered by the P and CR images in the eye camera. By recording with artificial eyes, we can also show that the pupil size artefact has no major role in increasing and colouring noise. Our findings add to and replicate the observation by Niehorster et al., (2021) that lowpass filters in video-based P–CR eye trackers colour the data. Irrespective of source, filters or head movements, coloured noise can be confused for oculomotor drift. We also find that usage of the default head restriction in the EyeLink 1000+, the EyeLink II and the HiSpeed240 result in noisier data compared to less head restriction. Researchers investigating data quality in eye trackers should consider not using the Gen 2 artificial eye from SR Research / EyeLink. Data recorded with this artificial eye are much noisier than data recorded with other artificial eyes, on average 2.2–14.5 times worse for the five eye trackers.

Event detection is a challenging stage in eye movement data analysis. A major drawback of current event detection methods is that parameters have to be adjusted based on eye movement data quality. Here we show that a fully automated classification of raw gaze samples as belonging to fixations, saccades, or other oculomotor events can be achieved using a machine-learning approach. Any already manually or algorithmically detected events can be used to train a classifier to produce similar classification of other data without the need for a user to set parameters. In this study, we explore the application of random forest machine-learning technique for the detection of fixations, saccades, and post-saccadic oscillations (PSOs). In an effort to show practical utility of the proposed method to the applications that employ eye movement classification algorithms, we provide an example where the method is employed in an eye movement-driven biometric application. We conclude that machine-learning techniques lead to superior detection compared to current state-of-the-art event detection algorithms and can reach the performance of manual coding.

The objective of this study was to compare the development of sleepiness during manual driving versus level 2 partially automated driving, when driving on a motorway in Sweden. The hypothesis was that partially auto-mated driving will lead to higher levels of fatigue due to underload. Eighty-nine drivers were included in the study using a 2 ? 2 design with the conditions manual versus partially automated driving and daytime (full sleep) versus night-time (sleep deprived). The results showed that night-time driving led to markedly increased levels of sleepiness in terms of subjective sleepiness ratings, blink durations, PERCLOS, pupil diameter and heart rate. Partially automated driving led to slightly higher subjective sleepiness ratings, longer blink durations, decreased pupil diameter, slower heart rate, and higher EEG alpha and theta activity. However, elevated levels of sleepiness mainly arose from the night-time drives when the sleep pressure was high. During daytime, when the drivers were alert, partially automated driving had little or no detrimental effects on driver fatigue. Whether the negative effects of increased sleepiness during partially automated driving can be compensated by the positive effects of lateral and longitudinal driving support needs to be investigated in further studies. Funding Agencies|ADASME [688900]; Mediator projects - European Unions Horizon 2020 research and innovation programme [814735]; Autoliv Development; SmartEye AB

Tracking users' gaze in virtual reality headsets allows natural and intuitive interaction with virtual avatars and virtual objects. Moreover, a technique known as foveated rendering can help save computational resources and enable hi-resolution but lightweight virtual reality technologies. Predominantly, eye-tracking hardware in modern VR headsets consist of infrared camera(s) and LEDs. Such hardware, together with image processing software consumes a substantial amount of energy, and, provided that hi-speed gaze detection is needed, might be very expensive. A promising technique to overcome these issues is photo-sensor oculography (PS-OG), which allows eye-tracking with high sampling rate and low power consumption. However, the main limitation of the previous PS-OG systems is their inability to compensate for the equipment shifts. In this study, we employ a simple multi-layer perceptron neural network to map raw sensor data to gaze locations and report its performance for shift compensation. Modeling and evaluation is done via a simulation.

Virtual reality (VR) is employed in a variety of different applications. It is our belief that eye-tracking is going to be a part of the majority of VR devices that will reduce computational burden via a technique called foveated rendering and will increase the immersion of the VR environment. A promising technique to achieve low energy, fast, and accurate eye-tracking is photo-sensor oculography (PS-OG). PS-OG technology enables tracking a user's gaze location at very fast rates - 1000Hz or more, and is expected to consume several orders of magnitude less power compared to a traditional video-oculography approach. In this demo we present a prototype of a PS-OG system that we started to develop. The long-term aim of our project is to develop a PS-OG system that is robust to sensor shifts. As a first step we have built a prototype that allows us to test different sensors and their configurations, as well as record and analyze eye-movement data.

It has come to our attention that the section "Post-processing: Labeling final events" on page 167 of "Using Machine Learning to Detect Events in Eye-Tracking Data" (Zemblys, Niehorster, Komogortsev, & Holmqvist, 2018) contains an erroneous description of the process by which post-processing was performed.

Prediction of object motion allows overcoming of object occlusions in its trajectory. In this study pseudorandom target trajectory with occlusions of 500 ms was used. It was found that for a period up to 200 ms after target occlusion, oculomotor system was driven by a short-term memory of the pre-occlusion target motion trajectory. We conclude, that when visual system is no longer able to predict location of occluded target, there is a tendency to shift gaze away from previous target location, similarly, away from the edges of the screen. We suggest that this is due to probability, accumulated in long-term memory, which supports expectation that the target should reappear near the centre of the screen. This behaviour supports basic principles of Bayesian decision theory. DOI: http://dx.doi.org/10.5755/j01.eee.18.9.2807

Knowledge on human eye-hand coordination can be used for human-like system design and medical diagnosis. This document analyses and briefly presents the parameters of the coordination while executing different eye-hand related tasks. Existing quantitative model of manuo-ocular coordination, capable of simulating the human performance in target tracking, is redesigned for a capability to simulate the performance of object guiding through a visible predefined path. Qualitative model, based on quantitative model, is proposed and explained.

Catch-up saccades and smooth pursuit are two components, quick and slow, of tracking eye movements. Their interaction has usually been studied during tracking a target moving only in the horizontal direction. In these experiments catch-up saccades were invoked by changing target velocity or introducing target jumps and occlusions. Our research was focused on ocular pursuit of two-dimensional non-predictive target trajectories without target jumps. Target trajectories presented to the subjects had the same shape but moved with slow, medium and high average speeds, and each trial had the duration of 60, 30 and 15 s, respectively. During the long tracking experiments, the sequences of successive catch-up saccades were obtained and their parameters were computed. Due to the contribution of catch-up saccades, elicited during ocular pursuit, the position errors and lagging times were reduced from 30 to 50 % and larger values were obtained for higher target velocities. Analyzing the sum of tracking distances which were overcome during slow phases and during catch-up saccades, we proposed new method to evaluate subject's pursuit gain.

The goal of the present investigation was to evaluate the effect of visual Muller-Lyer (M-L) illusion on the control of saccadic and smooth pursuit eye movements and compare it with the results obtained during perceptual judgment of the length of the shaft of M-L illusion. Experimental investigation revealed that the primary saccades elicited in the reflexive mode were mostly affected by the M-L illusion. The position errors of the primary saccades elicited in the reflexive mode were affected by 4% for wings-in illusion and by 3.6% for wings-out illusion comparing with the 0.25% and 0.1% for the saccades elicited in the voluntary mode. The position errors of complete saccades (0.14% and 0.02%) and tracking errors obtained during the smooth pursuit (0.11% and 0.05%) were negligibly small. Nevertheless, experimental results obtained during perceptual judgment of M-L illusion were substantially larger - 14% and 10% respectively. The obtained results have demonstrated that vision for perception and vision for action were executed by different neural networks and supported the two-visual-system hypothesis. DOI: http://dx.doi.org/10.5755/j01.itc.41.1.763

Optic stimulus used for human oculomotor system or eye-hand coordination researches is often represented on a computer screen. They must be synchronised with the eye tracker to achieve the accurate results of the conducted experiments. Relation between stimulus presence and eye image capturing times can be defined only with reference to knowledge of all system delays. The suggested research evaluates our eye tracking system delay and propose a practical method for measuring and evaluating delays to the computer graphic subsystem. Experimental research shows that LC Technologies eye tracker reports gaze angle with 28 ms lag and the average stimulus lag of the computer graphic subsystem is 9,5 ms. Santrauka Atliekant žmogaus okulomotorinės sistemos ar akies ir rankos koordinacijos tyrimus, naudojami optiniai stimulai, kurie atvaizduojami kompiuterio vaizduoklyje. Norint gauti tikslius tyrimų rezultatus, reikalinga šių stimulų sinchronizacija su žvilgsnio krypties registravimo įranga. Stimulo pasirodymo ir akies vaizdo užfiksavimo laikus sutapdinti galima tik įvertinus visas sistemos vėlinimo dedamąsias. Darbe apskaičiuotas naudojamos akių judesių registravimo įrangos vėlinimas ir pasiūlytas praktinis metodas kompiuterio grafinės posistemės vėlinimams išmatuoti. Atlikus eksperimentinius naudojamos sistemos tyrimus išmatuotos visos vėlinimo dedamosios ir apskaičiuotas bendras žvilgsnio sekimo įrangos vėlinimas – 28 ms ir vidutinis bendras kompiuterio grafinės posistemės ir vaizduoklio vėlinimas – 9,5 ms. Reikšminiai žodžiai: stimulo vėlinimas, kompiuterio vaizdo posistemė, kompiuterio vaizdo vėlinimas, stimulo atvaizdavimas vaizduoklyje, videookulografinis metodas, stimulo sinchronizavimas.

The purpose of this research was to investigate quantitatively the catch-up saccades occurring during smooth pursuit. In the first experiment, to evoke catch-up saccades we used high velocity predictable two-dimensional time-continuous target trajectories. In the second experiment, catch-up saccades were evoked using target jump paradigm during sustained two-dimensional pursuit. Target jumps in the different directions were presented at the unexpected moments and positions of the interrupted time-continuous target trajectory. From the experimental results we made a comparison of the main sequences (relationship between peak velocity and amplitude) of the catch-up and refixation saccades and found that they are different. Also we can conclude that the peak velocity of catch-up saccades is strongly correlated with the velocity of the smooth pursuit target component. We found that both position error and retinal slip are taken into account in catch-up saccades programming to predict the future trajectory of the moving target. Ill. 5, bibl. 5 (in English, summaries in English, Russian and Lithuanian).