Journal of Eye Movement Research
Not a member yet
517 research outputs found
Sort by
Detecting performance difficulty of learners in colonoscopy: Evidence from eye-tracking
No full textEye-tracking can help decode the intricate control mechanism in human performance. In healthcare, physicians-in-training requires extensive practice to improve their healthcare skills. When a trainee encounters any difficulty in the practice, they will need feedback from experts to improve their performance. The personal feedback is time-consuming and subjected to bias. In this study, we tracked the eye movements of trainees during their colonoscopic performance in simulation. We applied deep learning algorithms to detect the eye-tracking metrics on the moments of navigation lost (MNL), a signature sign for performance difficulty during colonoscopy. Basic human eye gaze and pupil characteristics were learned and verified by the deep convolutional generative adversarial networks (DCGANs); the generated data were fed to the Long Short-Term Memory (LSTM) networks with three different data feeding strategies to classify MNLs from the entire colonoscopic procedure. Outputs from deep learning were compared to the expert’s judgment on the MNLs based on colonoscopic videos. The best classification outcome was achieved when we fed human eye data with 1000 synthesized eye data, where accuracy (90%), sensitivity (90%), and specificity (88%) were optimized. This study built an important foundation for our work of developing a self-adaptive education system for training healthcare skills using simulation
I2DNet - Design and real-time evaluation of an appearance-based gaze estimation system
Full text linkGaze estimation problem can be addressed using either model-based or appearance-based approaches. Model-based approaches rely on features extracted from eye images to fit a 3D eye-ball model to obtain gaze point estimate while appearance-based methods attempt to directly map captured eye images to gaze point without any handcrafted features. Recently, availability of large datasets and novel deep learning techniques made appearance-based methods achieve superior accuracy than model-based approaches. However, many appearance-based gaze estimation systems perform well in within-dataset validation but fail to provide the same degree of accuracy in cross-dataset evaluation. Hence, it is still unclear how well the current state-of-the-art approaches perform in real-time in an interactive setting on unseen users. This paper proposes I2DNet, a novel architecture aimed to improve subject-independent gaze estimation accuracy that achieved a state-of-the-art 4.3 and 8.4 degree mean angle error on the MPIIGaze and RT-Gene datasets respectively. We have evaluated the proposed system as a gaze-controlled interface in real-time for a 9-block pointing and selection task and compared it with Webgazer.js and OpenFace 2.0. We have conducted a user study with 16 participants, and our proposed system reduces selection time and the number of missed selections statistically significantly compared to other two systems
Two hours in Hollywood: A manually annotated ground truth data set of eye movements during movie clip watching
Full text linkIn this short article we present our manual annotation of the eye movement events in a subset of the large-scale eye tracking data set Hollywood2. Our labels include fixations, saccades, and smooth pursuits, as well as a noise event type (the latter representing either blinks, loss of tracking, or physically implausible signals). In order to achieve more consistent annotations, the gaze samples were labelled by a novice rater based on rudimentary algorithmic suggestions, and subsequently corrected by an expert rater. Overall, we annotated eye movement events in the recordings corresponding to 50 randomly selected test set clips and 6 training set clips from Hollywood2, which were viewed by 16 observers and amount to a total of approximately 130 minutes of gaze data. In these labels, 62.4% of the samples were attributed to fixations, 9.1% – to saccades, and, notably, 24.2% – to pursuit (the remainder marked as noise). After evaluation of 15 published eye movement classification algorithms on our newly collected annotated data set, we found that the most recent algorithms perform very well on average, and even reach human-level labelling quality for fixations and saccades, but all have a much larger room for improvement when it comes to smooth pursuit classification. The data set is made available at https://gin.g- node.org/ioannis.agtzidis/hollywood2_em
Objective measurement of nine gaze-directions using an eye-tracking device
Full text linkPurpose: To investigate the usefulness and efficacy of a novel eye-tracking device that can objectively measure nine gaze-directions.
Methods: We measured each of the nine gaze-directions subjectively, using a conventional Hess screen test, and objectively, using the nine gaze-direction measuring device, and de-termined the correlation, addition error, and proportional error. We obtained two consecu-tive measurements of the nine gaze-directions using the newly developed device in healthy young people with exophoria and investigated the reproducibility of the measurements. We further measured the nine gaze-directions using a Hess screen test and the newly developed device in three subjects with cover test-based strabismus and compared the results.
Results: We observed that the objective measurements obtained with the newly developed gaze-direction measuring device had significant correlation and addition error compared to the conventional subjective method, and we found no proportional error. These measure-ments had good reproducibility.
Conclusion: The novel device can be used to observe delayed eye movement associated with limited eye movement in the affected eye, as well as the associated excessive movement of the healthy eye in patients with strabismus, similar to the Hess screen test. This is a useful device that can provide objective measurements of nine gaze-directions
Does pictorial composition guide the eye? Investigating four centuries of last supper pictures
Full text linkWithin art literature, there is a centuries-old assumption that the eye follows the lines set out by the composition of a painting. However, recent empirical findings suggest that this may not be true. This study investigates beholders’ saccadic eye movements while looking at fourteen paintings representing the scene of the Last Supper, and their perception of the compositions of those paintings. The experiment included three parts: 1) recording the eye movements of the participants looking at the paintings; 2) asking participants to draw the composition of the paintings; and 3) asking them to rate the amount of depth in the paintings. We developed a novel coefficient of similarity in order to quantify 1) the similarity between the saccades of different observers; 2) the similarity between the compositional drawings of different observers; and 3) the similarity between saccades and compositional drawings. For all of the tested paintings, we found a high, above-chance similarity between the saccades and between the compositional drawings. Additionally, for most of the paintings, we also found a high, above-chance similarity between compositional lines and saccades, both on a collective and on an individual level. Ultimately, our findings suggest that composition does influence visual perception. 
Cognitive strategies revealed by clustering eye movement transitions
Full text linkIn cognitive tasks, solvers can adopt different strategies to process information which may lead to different response behavior. These strategies might elicit different eye movement patterns which can thus provide substantial information about the strategy a person uses. However, these strategies are usually hidden and need to be inferred from the data. After an overview of existing techniques which use eye movement data for the identification of latent cognitive strategies, we present a relatively easy to apply unsupervised method to cluster eye movement recordings to detect groups of different solution processes that are applied in solving the task. We test the method\u27s performance using simulations and demonstrate its use on two examples of empirical data. Our analyses are in line with presence of different solving strategies in a Mastermind game, and suggest new insights to strategic patterns in solving Progressive matrices tasks
A two-step approach for interest estimation from gaze behavior in digital catalog browsing
Full text linkWhile eye gaze data contain promising clues for inferring the interests of viewers of digital catalog content, viewers often dynamically switch their focus of attention. As a result, a direct application of conventional behavior analysis techniques, such as topic models, tends to be affected by items or attributes of little or no interest to the viewer. To overcome this limitation, we need to identify “when” the user compares items and to detect “which attribute types/values” reflect the user’s interest. This paper proposes a novel two-step approach to addressing these needs. Specifically, we introduce a likelihood-based short-term analysis method as the first step of the approach to simultaneously determine comparison phases of browsing and detect the attributes on which the viewer focuses, even when the attributes cannot be directly obtained from gaze points. Using probabilistic latent semantic analysis, we show that this short-term analysis step greatly improves the results of the subsequent step. The effectiveness of the framework is demonstrated in terms of the capability to extract combinations of attributes relevant to the viewer’s interest, which we call aspects, and also to estimate the interest described by these aspects
Eye movements and mental imagery during reading of literary texts with different narrative styles
Full text linkBased on Kuzmičová’s (2014) phenomenological typology of narrative styles, we studied the specific contributions of mental imagery to literary reading experience and to reading behavior by combining questionnaires with eye-tracking methodology. Specifically, we focused on the two main categories in Kuzmičová’s (2014) typology, i.e., texts dominated by an “enactive” style, and texts dominated by a “descriptive” style. “Enactive” style texts render characters interacting with their environment, and “descriptive” style texts render environments dissociated from human action. The quantitative analyses of word category distributions of two dominantly enactive and two dominantly descriptive texts indicated significant differences especially in the number of verbs, with more verbs in enactment compared to descriptive texts. In a second study, participants read two texts (one theoretically cueing descriptive imagery, the other cueing enactment imagery) while their eye movements were recorded. After reading, participants completed questionnaires assessing aspects of the reading experience generally, as well as their text-elicited mental imagery specifically. Results show that readers experienced more difficulties conjuring up mental images during reading descriptive style texts and that longer fixation duration on words were associated with enactive style text. We propose that enactive style involves more imagery processes which can be reflected in eye movement behavior
Calibration-free gaze interfaces based on linear smooth pursuit
Full text linkSince smooth pursuit eye movements can be used without calibration in spontaneous gaze interaction, the intuitiveness of the gaze interface design has been a topic of great interest in the human-computer interaction field. However, since most related research focuses on curved smooth-pursuit trajectories, the design issues of linear trajectories are poorly understood. Hence, this study evaluated the user performance of gaze interfaces based on linear smooth pursuit eye movements. We conducted an experiment to investigate how the number of objects (6, 8, 10, 12, or 15) and object moving speed (7.73 ˚/s vs. 12.89 ˚/s) affect the user performance in a gaze-based interface.
Results show that the number and speed of the displayed objects influence users’ performance with the interface. The number of objects significantly affected the correct and false detection rates when selecting objects in the display. Participants’ performance was highest on interfaces containing 6 and 8 objects and decreased for interfaces with 10, 12, and 15 objects. Detection rates and orientation error were significantly influenced by the moving speed of displayed objects.
Faster moving speed (12.89 ˚/s) resulted in higher detection rates and smaller orientation error compared to slower moving speeds (7.73 ˚/s).
Our findings can help to enable a calibration-free accessible interaction with gaze interfaces
MAD saccade: statistically robust saccade threshold estimation via the median absolute deviation
Full text linkSaccade detection is a critical step in the analysis of gaze data. A common method for saccade detection is to use a simple threshold for velocity or acceleration values, which can be estimated from the data using the mean and standard deviation. However, this method has the downside of being influenced by the very signal it is trying to detect, the outlying velocities or accelerations that occur during saccades. We propose instead to use the median absolute deviation (MAD), a robust estimator of dispersion that is not influenced by outliers. We modify an algorithm proposed by Nyström and colleagues, and quantify saccade detection performance in both simulated and human data. Our modified algorithm shows a significant and marked improvement in saccade detection - showing both more true positives and less false negatives – especially under higher noise levels. We conclude that robust estimators can be widely adopted in other common, automatic gaze classification algorithms due to their ease of implementation
Data sources:
BORIS Portal Bern Open Repository and I...BAF-Online: Proceedings of the Berner Al...HASBonline – Hefte zur Archäologie des M...Informationswissenschaft: Theorie, Metho...Jahrbuch Diakonie Schweiz
Journal of Eye Movement ResearchJournal of Radiation Oncology Informatic...BOP SerialsVersants. Revista suiza de literaturas r...BORIS Theses
Journal of Eye Movement ResearchJournal of Radiation Oncology Informatic...BOP SerialsVersants. Revista suiza de literaturas r...BORIS ThesesAccess Repository Dashboard
Do you manage Journal of Eye Movement Research? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!