Publications

Explore a collection of publications and projects, from diverse fields, that cite Pupil Labs and use Pupil Labs eye tracking hardware and software in their research.

Filters

Year

103
64
55
47
40
49
25
9

Product

324
44
16
8

Fields

80
54
44
39
35
33
27
27
25
24
23
20
19
17
13
11
11
10
10
9
6
6
6
6
6
5
5
5
5
5
5
5
4
4
4
4
4
3
3
3
3
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0-0 of 392 publications
PyPlr: A versatile, integrated system of hardware and software for researching the human pupillary light reflex
2021
Pupillometry, PLR, Neurology
Martin, Joel T. Pinto, Joana Bulte, Daniel Spitschan, Manuel
Behavior Research Methods
We introduce PyPlr —a versatile, integrated system of hardware and software to support a broad spectrum of research applications concerning the human pupillary light reflex (PLR). PyPlr is a custom Python library for integrating a research-grade video-based eye-tracker system with a light source and streamlining stimulus design, optimisation and delivery, device synchronisation, and extraction, cleaning, and analysis of pupil data. We additionally describe how full-field, homogenous stimulation of the retina can be realised with a low-cost integrating sphere that serves as an alternative to a more complex Maxwellian view setup. Users can integrate their own light source, but we provide full native software support for a high-end, commercial research-grade 10-primary light engine that offers advanced control over the temporal and spectral properties of light stimuli as well as spectral calibration utilities. Here, we describe the hardware and software in detail and demonstrate its capabilities with two example applications: (1) pupillometer-style measurement and parametrisation of the PLR to flashes of white light, and (2) comparing the post-illumination pupil response (PIPR) to flashes of long and short-wavelength light. The system holds promise for researchers who would favour a flexible approach to studying the PLR and the ability to employ a wide range of temporally and spectrally varying stimuli, including simple narrowband stimuli.
Spatial neglect midline diagnostics from virtual reality and eye tracking in a free-viewing environment
2021
Neuropsychology
Hougaard, B. I., Knoche, H., Jensen, J., & Evald, L
Frontiers in Psychology
Purpose: Virtual reality (VR) and eye tracking may provide detailed insights into spatial cognition. We hypothesized that virtual reality and eye tracking may be used to assess sub-types of spatial neglect in stroke patients not readily available from conventional assessments. Method: Eighteen stroke patients with spatial neglect and 16 age and gender matched healthy subjects wearing VR headsets were asked to look around freely in a symmetric 3D museum scene with three pictures. Asymmetry of performance was analyzed to reveal group-level differences and possible neglect sub-types on an individual level.Results: Four out of six VR and eye tracking measures revealed significant differences between patients and controls in this free-viewing task. Gaze-asymmetry between-pictures (including fixation time and count) and head orientation were most sensitive to spatial neglect behavior on a group level analysis. Gaze-asymmetry and head orientation each identified 10 out of 18 (56%), compared to 12 out of 18 (67%) for the best conventional test. Two neglect patients without deviant performance on conventional measures were captured by the VR and eyetracking measures. On the individual level, five stroke patients revealed deviant gaze-asymmetry within-pictures and six patients revealed deviant eye orientation in either direction that were not captured by the group-level analysis.Conclusion: This study is a first step in using VR in combination with eye tracking measures as individual differential neglect subtype diagnostics. This may pave the way for more sensitive and elaborate sub-type diagnostics of spatial neglect that may respond differently to various treatment approaches.
What Are You Looking At? Using Eye Tracking Glasses to Monitor Toddler Attention in Natural Learning Situations.
2021
Developmental Psychology
Nicole Altvater-Mackensen
Eye movements are taken to reflect cognitive processes. Because eye tracking is non-invasive and does not require an overt behavioural response, it is a suitable method to use with infants and toddlers. Most eye tracking studies to date use screen-based remote systems which allow high experimental control but limit the possibilities to test participants in more naturalistic, interactive settings. In this paper, we present a prototype of mobile eye tracking glasses suitable for toddlers and describe its use in a set-up that allows for automatic gaze coding. We present the components and set-up of the eye tracking glasses, as well as the basic principles and routines of data acquisition and analysis. Data from a pilot study testing 2- to 5-year-olds in a shared reading interaction and a preferential looking task is used to illustrate data quality and potential pitfalls in data collection.
An implicit representation of stimulus ambiguity in pupil size
2021
audio-visual, pupilometry, cognition
Graves, J. E., Egré, P., Pressnitzer, D., & de Gardelle, V.
Procedings of the national academy of sciences
To guide behavior, perceptual systems must operate on intrinsically ambiguous sensory input. Observers are usually able to acknowledge the uncertainty of their perception, but in some cases, they critically fail to do so. Here, we show that a physiological correlate of ambiguity can be found in pupil dilation even when the observer is not aware of such ambiguity. We used a well-known auditory ambiguous stimulus, known as the tritone paradox, which can induce the perception of an upward or downward pitch shift within the same individual. In two experiments, behavioral responses showed that listeners could not explicitly access the ambiguity in this stimulus, even though their responses varied from trial to trial. However, pupil dilation was larger for the more ambiguous cases. The ambiguity of the stimulus for each listener was indexed by the entropy of behavioral responses, and this entropy was also a significant predictor of pupil size. In particular, entropy explained additional variation in pupil size independent of the explicit judgment of confidence in the specific situation that we investigated, in which the two measures were decoupled. Our data thus suggest that stimulus ambiguity is implicitly represented in the brain even without explicit awareness of this ambiguity.
Social Virtual Reality: Implementing Non-verbal Cues in Remote Synchronous Communication
2021
VR
Kasapakis, V., Dzardanova, E., Nikolakopoulou, V., Vosinakis, S., Xenakis, I., & Gavalas, D.
International Conference on Virtual Reality and Mixed Reality
Social Virtual Reality (SVR) platforms allow remote, synchronous interaction and communication between individuals immersed in shared virtual worlds. Such platforms commonly implement full-body motion and real-time voice communication, but often lack complete non-verbal cues support. This work presents the development process and preliminary usability evaluation results of an SVR platform, incorporating non-verbal cues such as finger motion, gaze direction, and facial expressions, while allowing inter-communication between remotely located interlocutors.
Binocular vision and the control of foot placement during walking in natural terrain.
2021
Navigation, Walking, Motor Control
Bonnen, K., Matthis, J. S., Gibaldi, A., Banks, M. S., Levi, D. M., & Hayhoe, M.
Scientific Reports (Nature)
Coordination between visual and motor processes is critical for the selection of stable footholds when walking in uneven terrains. While recent work (Matthis et al. in Curr Biol 8(28):1224–1233, 2018) demonstrates a tight link between gaze (visual) and gait (motor), it remains unclear which aspects of visual information play a role in this visuomotor control loop, and how the loss of this information affects that relationship. Here we examine the role of binocular information in the visuomotor control of walking over complex terrain. We recorded eye and body movements while normally-sighted participants walked over terrains of varying difficulty, with intact vision or with vision in one eye blurred to disrupt binocular vision. Gaze strategy was highly sensitive to the complexity of the terrain, with more fixations dedicated to foothold selection as the terrain became more difficult. The primary effect of increased sensory uncertainty due to disrupted binocular vision was a small bias in gaze towards closer footholds, indicating greater pressure on the visuomotor control process. Participants with binocular vision losses due to developmental disorders (i.e., amblyopia, strabismus), who have had the opportunity to develop alternative strategies, also biased their gaze towards closer footholds. Across all participants, we observed a relationship between an individual’s typical level of binocular visual function and the degree to which gaze is shifted toward the body. Thus the gaze–gait relationship is sensitive to the level of sensory uncertainty, and deficits in binocular visual function (whether transient or long-standing) have systematic effects on gaze strategy in complex terrains. We conclude that binocular vision provides useful information for locating footholds during locomotion. Furthermore, we have demonstrated that combined eye/body tracking in natural environments can be used to provide a more detailed understanding of the impact of a type of vision loss on the visuomotor control process of walking, a vital everyday task.
Gaze-Adaptive Subtitles Considering the Balance among Vertical/Horizontal and Depth of Eye Movement.
2021
AR
Shimizu, Y., Ohnishi, A., Terada, T., & Tsukamoto, M.
IEEE International Symposium on Mixed and Augmented Reality Adjunct
Subtitles (captions displayed on the screen) are important in 3D content, such as virtual reality (VR) and 3D movies, to help users understand the content. However, an optimal displaying method and framework for subtitles have not been established for 3D content because 3D has a depth factor. To determine how to place text in 3D content, we propose four methods of moving subtitles dynamically considering the balance between the vertical/horizontal and depth of gaze shift. These methods are used to reduce the difference in depth or distance between the gaze position and subtitles. Additionally, we evaluate the readability of the text and participants’ fatigue. The results show that aligning the text horizontally and vertically to eye movements improves visibility and readability. It is also shown that the eyestrain is related to the distance between the object and subtitles. This evaluation provides basic knowledge for presenting text in 3D content.
Human gaze-aware attentive object detection for ambient intelligence.
2021
HCI, AR
Cho, D. Y., & Kang, M. K.
Engineering Applications of Artificial Intelligence
Understanding human behavior and the surrounding environment is essential for realizing ambient intelligence (AmI), for which eye gaze and object information are reliable cues. In this study, the authors propose a novel human gaze-aware attentive object detection framework as an elemental technology for AmI. The proposed framework detects users’ attentive objects and shows more precise and robust performance against object-scale variations. A novel Adaptive-3D-Region-of-Interest (Ada-3D-RoI) scheme is designed as a front-end module, and scalable detection network structures are proposed to maximize cost-efficiency. The experiments show that the detection rate is improved up to 97.6% on small objects (14.1% on average), and it is selectively tunable with a tradeoff between accuracy and computational complexity. In addition, the qualitative results demonstrate that the proposed framework detects a user’s single object-of-interest only, even when the target object is occluded or extremely small. Complementary matters for follow-up study are presented as suggestions to extend the results of the proposed framework to further practical AmI applications. This study will help develop advanced AmI applications that demand a higher-level understanding of scene context and human behavior such as human–robot symbiosis, remote-/autonomous control, and augmented/mixed reality.
Integrating Mobile Eye-Tracking and VSLAM for Recording Spatial Gaze in Works of Art and Architecture
2021
Eye Tracking, Attention, Visualization, Architecture
Eugene Han
Technology|Architecture + Design
This paper proposes a method for spatial eye-tracking for use with three-dimensional objects and environments. By integrating mobile eye-tracking (MET) with Visual Simultaneous Localization and Mapping (VSLAM) technologies, the study provides an unobtrusive technique for capturing an individual’s gaze across an open space and through an unprescribed viewing path. The included proof-of-concept is tested against three scales, from a large sculpture to a set of towering cement kilns and an exterior building passage. Demonstrations show that the integration of MET and VSLAM provides a useful tool for testing scenarios without predefined viewing conditions and allowing insight into how others view works of art and architecture.
CorrNet: Fine-Grained Emotion Recognition for Video Watching Using Wearable Physiological Sensors
2021
Psychology
Tianyi Zhang, Abdallah El Ali, Chen Wang, Alan Hanjalic and Pablo Cesar
MDPI: Sensors
Recognizing user emotions while they watch short-form videos anytime and anywhere is essential for facilitating video content customization and personalization. However, most works either classify a single emotion per video stimuli, or are restricted to static, desktop environments. To address this, we propose a correlation-based emotion recognition algorithm (CorrNet) to recognize the valence and arousal (V-A) of each instance (fine-grained segment of signals) using only wearable, physiological signals (e.g., electrodermal activity, heart rate). CorrNet takes advantage of features both inside each instance (intra-modality features) and between different instances for the same video stimuli (correlation-based features). We first test our approach on an indoor-desktop affect dataset (CASE), and thereafter on an outdoor-mobile affect dataset (MERCA) which we collected using a smart wristband and wearable eyetracker. Results show that for subject-independent binary classification (high-low), CorrNet yields promising recognition accuracies: 76.37% and 74.03% for V-A on CASE, and 70.29% and 68.15% for V-A on MERCA. Our findings show: (1) instance segment lengths between 1–4 s result in highest recognition accuracies (2) accuracies between laboratory-grade and wearable sensors are comparable, even under low sampling rates (≤64 Hz) (3) large amounts of neutral V-A labels, an artifact of continuous affect annotation, result in varied recognition performance.