Science AI Journal
Computer Science · Research topic

Open research questions in Gaze Tracking and Assistive Technology

38 unresolved questions extracted from the limitations and future-work sections of 711 Gaze Tracking and Assistive Technology papers in our library. Each links back to the study that raised it.

What the literature leaves open

  • This study showed that gaze and proxemics provide meaningful information about how learners detect, monitor, and respond to collaborative challenges. Physical distance sets the condition for regulation of learning, and gaze supports both the initiation and maintenance of these regulatory processes depending on spatial configuration. Adaptive AI scaffolding enhances monitoring and socioemotional responsiveness, while static prompts promote more procedural forms of regulation in collaborative learning. The multimodal methods used in this study offer new pathways for investigating SSRL as a temporally organized, embodied process and support the development of AI tools that are sensitive to the com- plex dynamics of collaborative learning. Going forward, it is important to note that individual roles shape how groups function (Isohätälä et al., 2020; Vuorenmaa et al., 2022), and tracing how these nonverbal behav- iors unfold would offer important insight into emerging regulatory patterns. For gaze, a valuable direction is to examine how gaze patterns relate to within-group dynamics, such as whether certain pairs share more mutual gaze or joint attention or how individual gaze sequences toward the task evolve during interaction. For proxemics, future studies could analyze how members differ in their pairwise spatial positioning and how these distances influence SSRL processes. The findings also carry practical implications for the design of AI agents that support collaborative learning. Systems that can detect nonverbal signals, such as shifts in mutual gaze or spatial reconfiguration, may be better positioned to deliver timely and contingent support than systems relying solely on discourse. In such a case, incorporating multimodal sensing capabilities that capture embodied indicators of regula- tory need would enable more responsive scaffolding that aligns with the natural temporal rhythms of group-level processes. Author Contribution R.W. wrote the main manuscript text - including the first draft, analyzed the data, developed the software adapted and created for the analysis, interpreted the findings and discussed the results. A.N. made substantial contributions to the conception or design of the work, contributed to the interpretation of findings, supervised on the methods, and edited the manuscript. S. J. Supervised the overall work, approved version to be published, made substantial contributions to the conceptualization of the work and editing of manuscript drafts. Funding Open Access funding provided by University of Oulu (including Oulu University Hospital). Data Availability The video data that support the findings of this study are not publicly available because they contain identifiable information about minor participants, and participant consent and ethics approvals restrict public sharing. Access to the data may be granted for research purposes, subject to approval by the relevant ethics body and data controller, and completion of an appropriate data use agreement. Requests should be directed to the corresponding author.

    How gaze and proxemics shape and signal shared regulatory processes during AI-supported collaborative learning · 2026 · DOI

  • , 2020; Taylor, 2014), future research should explore absolute or task-specific spatial benchmarks and consider how classroom design scaffolds or constrains SSRL processes.

    How gaze and proxemics shape and signal shared regulatory processes during AI-supported collaborative learning · 2026 · DOI

  • Although MOT performance is known to decline with age, and to depend on efficient oculomotor strategies, how these processes interact across the adult lifespan and under degraded visual input remains poorly understood.

    Effects of aging on multiple object tracking under normal and altered viewing conditions · 2026 · DOI

  • Limitations One limitation of the present study which is related to the multi-level factorial design of the statistical model is the size of the analyzed sample. Another limitation of the present study is that it was focused solely on gaze behavior and did not include other forms of the explainees’ feed- back behavior, such as head gestures and backchannel responses. We therefore assume that the same statistical analysis should be replicated on a larger sample.

    How are Gaze Aversions and Mutual Gaze Related to the Topical Development of Dyadic Explanatory Interactions? · 2026 · DOI

  • Additionally, our two-party protocol is not limited to the Needleman–Wunsch-based method but also supports the evaluation of MultiMatch and SubsMatch.

    Secure Storage and Privacy-Preserving Scanpath Comparison via Garbled Circuits in Eye Tracking ETRA008 · 2026 · DOI

  • My research will extend resource-rational modeling of visual attention toward increasingly adaptive and interactive settings. From a theoretical perspective, a key next step is to generalize computationally rational control from single-agent behavior to multi-agent and socially situated scenarios. Many real-world tasks, such as collaborative reading in classrooms or shared attention in AR/VR environments require coordination among multiple agents with overlapping but distinct goals. Modeling such settings will enable the study of how attention is allocated not only within individuals but also across groups, opening new avenues for understanding joint cognition and collaborative human–computer interaction. On the application side, I plan to leverage my models to inform adaptive interface design across a broader range of HCI scenarios. While reading provides a controlled and cognitively rich testbed, the same simulation framework can be extended to interactions with visualizations, images, and complex user interfaces, where users must strategically allocate attention under time and resource constraints. By simulating how users adjust attention during multitasking, time pressure, or high cognitive load, these models can support systematic analysis of design alternatives prior to extensive user studies, and guide the development of time-aware reading interfaces, anticipatory safety prompts in AR systems, and adaptive visual and educational interfaces that respond to predicted user states [2, 24]. Technically, I aim to integrate deep generative models into the simulation pipeline to enhance scalability and realism. Large language models can provide rich contextual predictions to support semantic inference and memory construction, while diffusion- A Resource-Rational Principle for Modeling Visual Attention Control CHI EA ’26, April 13–17, 2026, Barcelona, Spain Figure 3: Reading behavior under different time constraints (30 s, 60 s, 90 s). The model and human readers adapt their visual attention strategies as available reading time changes, reflecting resource-rational control of comprehension. (a) Heatmaps. With increasing time (from left to right), both humans and the simulation allocate attention more broadly across the text and invest more fixation time, whereas under tight deadlines attention is concentrated on high-coverage regions. (b) Recall. Free-recall responses exhibit the same trade-off, with longer reading time yielding richer gist-level recall and more complete and detailed memory. or transformer-based generative models can be used to synthesize realistic eye-movement patterns or controlled interaction corpora [5, 6, 9]. Incorporating these components will allow simulationbased analyses of nuanced comprehension and interaction processes and support interfaces that adapt dynamically to inferred user states.

    A Resource-Rational Principle for Modeling Visual Attention Control · 2026 · DOI

  • RESULTS: Study results were somewhat mixed but did demonstrate that with the tested CMS, participants took longer to pass a slower moving vehicle and maintained a greater resultant distance from the passed vehicle.

    Driver eye glance behavior and performance with camera-based visibility systems versus mirrors · 2023 · DOI

  • Font styles/features are also known to affect reading eye movements (EM); however, few studies have examined these effects - and none used high-resolution displays.

    The effect of font width on eye movements during reading · 2021 · DOI

  • Despite increased research focusing on the impact of digital video recorder (DVR) use, there continues to be a notable lack of data on viewers’ direct experience of time-shifted television ads.

    Biometric and eye-tracking insights into the efficiency of information processing of television advertising during fast-forward viewing · 2008 · DOI

  • Future research should examine the utility of the blink reflex according to the desirable properties of sensitivity, diagnosticity, validity, reliability, ease of use, unobtrusiveness, and operator acceptance.

    Spontaneous and reflexive eye activity measures of mental workload · 2002 · DOI

  • Is the presumed ability to detect an unseen gaze merely a superstition, a cultural myth without real substance, an overinflation of coincidental occurrences, or a response to subtle sensory cues? Or, alternatively, could the experience be a valid indicator of an exceptional yet poorly understood human capability? In 1898, Titchener published a short article in which he addressed the feeling of being stared at.

    Reactions to an unseen gaze (remote attention): a review, with new data on autonomic staring detection · 1993

  • ROBINSON AND O T H E R S August, 1972-323 method the precision of each measurement is limited by the consistency of the reaction time of the observer.

    Visual Search by Automobile Drivers · 1972 · DOI

  • Smooth pursuit eye movement abnormalities are well established in psychosis, but the specific components of pursuit performance that vary across clinical states and symptom profiles remain insufficiently characterized.

    Structured Occlusion Reveals State-Dependent Smooth Pursuit Deficits Across Acute and Chronic Psychosis · 2026 · DOI

  • The study thus offers an analytical framework that can serve as a starting point for further research on media literacy, regulation of sensitive content, and the design of empirical studies of short-form videos.

    Watching Tiktok: An Eye-Tracking Study of Visual Attention Paid to Short-form Videos · 2026 · DOI

  • We show that anchoring also occurs for visual-haptic targets but is limited to the initial phase of the reach, likely because vision is not required to confirm target attainment.

    Gaze-related functions driving gaze anchoring in reaching · 2026 · DOI

  • Although gaze following is an important socio-interactive process, little is known about how this behavior is affected when multiple gaze cues are encountered in groups.

    Flexible use of quorum and numerosity principles in evaluation of social and non-social cues in group contexts · 2026 · DOI

  • Wearable eye tracking in unconstrained settings is often compromised by slippage, yet the direct relationship between physical frame displacement and gaze error remains unexplored.

    Uncovering the effect of slippage on wearable eye trackers: a motion capture study · 2026 · DOI

  • A key open question is whether gaze awareness, an avatar’s ability to react to the gaze of a user, contributes to more natural and socially engaging interaction, particularly on conventional 2D screens.

    Investigating Mutual Gaze in Real-Time LLM-Driven Avatar Interaction on 2D Screens ETRA005 · 2026 · DOI

  • Perceptual load is a well-established determinant of attention and perception, however, the effects of perceptual load on a person's ability to respond to a takeover signal and the related gaze indicators are not yet known.

    Establishing gaze markers of perceptual load during multi-target visual search · 2023 · DOI

  • Measures of gaze behaviors indicative of visual monitoring strategies were most sensitive to task difficulty manipulations, and should be explored further for the pHRI domain where motor-control and internal-model formation will likely be strong contributors to workload.

    Eye-Tracking in Physical Human–Robot Interaction: Mental Workload and Performance Prediction · 2023 · DOI

  • The third part, first the description of the recently emerged concept of eye-tracking in virtual reality is given, followed by the current applications to education and studying, which has not been thoroughly described before.

    Eye-Tracking in Immersive Virtual Reality for Education: A Review of the Current Progress and Applications · 2022 · DOI

  • In the field of digital libraries, research on user experience is sparse, especially in terms of emotions, impressions, and stimulation evoked by interaction with the digital library.

    Exploring user experience in digital libraries through questionnaire and eye-tracking data · 2022 · DOI

  • On the one hand, studies have provided mixed evidence from eye measures about tasks: pupil size has mostly been reported to increase with increasing task demand while this expected change was not observed in some studies, and blink rate has exhibited different trends in different tasks.

    Pupillometric and blink measures of diverse task loads: Implications for working memory models · 2022 · DOI

  • CONCLUSIONS: Within the limitations of this study, the present findings suggest that in photographs of the mouth with non-esthetic restoration on either side, the non-esthetic restoration is first gazed before the opposite side.

    Analysis of gaze points for mouth images using an eye tracking system · 2017 · DOI

  • In addition, we examine the relationships between search performance and several search behaviors to allow further investigation of the differences engendered by the screens.

    Eye‐tracking analysis of user behavior and performance in web search on large and small screens · 2014 · DOI

Most-cited papers in Gaze Tracking and Assistive Technology

Most recent work

Find a gap in your own Gaze Tracking and Assistive Technology sub-topic

This page shows what the Gaze Tracking and Assistive Technology literature already flags as unresolved. To narrow it to your specific question, run the guided finder — it searches the gap library on demand and checks candidates against 250M+ OpenAlex works.

Open the Research Gap Finder →

Related topics in Computer Science

Online Learning and Analytics
363 open questions
AI in Service Interactions
278 open questions
Teaching and Learning Programming
239 open questions
Digital literacy in education
230 open questions
Mobile Learning in Education
183 open questions
Virtual Reality Applications and Impacts
178 open questions
Computational Drug Discovery Methods
126 open questions
Blockchain Technology Applications and Security
121 open questions
38 open questions have been extracted from the limitations and future-work passages of 711 Gaze Tracking and Assistive Technology papers in our library. Each one below links back to the study that raised it, so you can read the original claim in context.

Tools for your next paper

Gaps & Literature SearchFind open questions in any field — 120,000+ open gaps mined from our own library.Pre-CheckIs your paper ready? A Tier 1–5 acceptance probability in seconds.Journal FinderA ranked shortlist of target journals from a 17,500-venue index.Duplicate Publication CheckerCheck prior-publication & salami-slicing risk across 8 sources.Citation GeneratorFreeA DOI or title → APA, MLA, Vancouver, Chicago, BibTeX, RIS. No signup.Graphical Abstract MakerFreeTurn your findings into a colorblind-safe graphical abstract.

Compare the category — Honest roundups of the AI research tools, ours listed alongside the alternatives.

Command palette

Jump anywhere, run any action.