Reading Together, Apart: Aggregated Gaze Traces as an Asynchronous Social Cue

Master Thesis open

Project overview

Our recent VR dyad study examined shared attention between two people in the same place at the same time. Attention is also shared across time: when many people read the same document, their fixations leave a record of where attention went, and that record currently goes to researchers and system administrators rather than back to readers. This project renders aggregated gaze as a visible trace inside a reading interface and measures what it does to the next reader.

Project motivation

Traces of earlier use shape how people engage with a place, from worn paths to handled surfaces, and the same logic has been transferred to virtual environments [1, 2] and to public displays, where a self-checkout prototype made previous users’ touch traces visible and evaluated the result [3]. Reading interfaces were an early candidate for this idea, with read wear proposed as a way of making document histories visible [4], but implementations have relied on scroll position and dwell at the page level rather than on gaze. Work on shared gaze in collaboration has stayed synchronous, showing partners each other’s cursors or fixations in real time [5]. A trace is also ambiguous in what it records: a long dwell can mean the passage was difficult or that it was worth staying with, and those two readings call for different displays. Saliency models supply a further design option, since they can generate a plausible trace layer before the first reader arrives [6], which makes it possible to separate the visual form of a trace from the belief that other people produced it.

Project goals

  1. Trace rendering. Build a reading interface with eye tracking that accumulates fixations across readers and renders them with controllable intensity, spatial granularity and decay over time.
  2. Trace experiment. Compare human aggregated traces, saliency-generated traces and no traces on where the next reader looks, how long they read, what they recall, and whether they report a sense of other readers. Vary the stated provenance of the trace (passages others found difficult versus passages others stayed with) to test whether interpretation rather than visual form drives the effect.

You will

  • Perform a literature review on traces of use, read wear and shared gaze
  • Build the reading interface and the trace accumulation and rendering pipeline
  • Run a study with an accumulating participant pool, so that later participants see traces produced by earlier ones
  • Analyze the data with the existing Python/R mixed-model tooling
  • Summarize your findings in a thesis and present them to an audience
  • (Optional) co-write a research paper

You need

  • Strong communication skills in English
  • Good knowledge of web front-end development or Unity
  • Basic statistics (mixed models are a plus)

Co-supervised with Linda Hirsch, KTH Royal Institute of Technology.

References

  1. Hirsch, L., George, C., & Butz, A. (2022). Traces in Virtual Environments: A Framework and Exploration to Conceptualize the Design of Social Virtual Environments. IEEE Transactions on Visualization and Computer Graphics, 28(11), 3874–3884. https://doi.org/10.1109/TVCG.2022.3203092
  2. Hirsch, L., Haller, A., Butz, A., & George, C. (2022). What a Mess! Traces of Use to Increase Asynchronous Social Presence in Shared Virtual Environments. IEEE VRW 2022, 598–599. https://doi.org/10.1109/VRW55335.2022.00150
  3. Süsslin, L. E., Mäkelä, V., Alt, F., & Hirsch, L. (2025). Designing Safer Touch Displays: Digitally Distributing Physical Touch on a Public Display. Mensch und Computer 2025, 307–321. https://doi.org/10.1145/3743049.3743083
  4. Hill, W. C., Hollan, J. D., Wroblewski, D., & McCandless, T. (1992). Edit wear and read wear. CHI ’92, 3–9. https://doi.org/10.1145/142750.142751
  5. D’Angelo, S., & Gergle, D. (2016). Gazed and Confused: Understanding and Designing Shared Gaze for Remote Collaboration. CHI ’16, 2492–2496. https://doi.org/10.1145/2858036.2858499
  6. Itti, L., Koch, C., & Niebur, E. (1998). A model of saliency-based visual attention for rapid scene analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence, 20(11), 1254–1259. https://doi.org/10.1109/34.730558

Interested in this topic? Reach out through your university student email address via the contact form, with a short motivation, your transcript of records and, if available, a CV.

← All topics