Research Digest

|

Psychology

Researchers 3D-Print an Add-On for Stereoscopic Eye Tracking with Neon

Custom Hardware

Hold a finger in front of your face and close one eye, then the other. Your finger seems to jump sideways. That is because your eyes see the world from slightly different positions.

Your brain uses differences between those two views to help judge depth: how far away things are and where they sit relative to one another. This is called stereoscopic vision.

Researchers can investigate this process by controlling exactly what each eye sees. Present two carefully offset views of an object, for example, and they can test how the brain turns those differences into a sense of depth.

Doing this while tracking eye movements, however, often requires expensive, elaborate equipment. Established approaches combine a stationary eye tracker with an arrangement of mirrors or with electronic shutter glasses and a specialised display. These setups can be difficult to take beyond the laboratory.

A Low-Cost 3D-Printed Add-On for Neon

Figure 1: The 3D-printable anaglyph add-on: (A) 3D-printable model, (B) assembled add-on with color filters, and (C) add-on mounted on the Neon “I can see clearly now” system. Spectrometer measurements confirmed that the filters effectively separated the color channels, with red signal passing almost exclusively through the red filter and blue signal through the cyan filter. Adapted from Grimm, S., Weirauch, A., & Einhäuser, W. (2026, June). An Anaglyph Add-on for the Pupil Labs Neon–Comparison to Stationary Eye-Tracking Setups for Stereoscopic Presentation using Binocular Rivalry. In Proceedings of the 2026 Symposium on Eye Tracking Research and Applications (pp. 1-7).

Researchers at Chemnitz University of Technology set out to build a simpler, more portable alternative. Their solution paired Pupil Labs’ Neon with a low-cost, 3D-printed attachment holding coloured filters.

The attachment uses the principle behind red-and-cyan 3D glasses, known as anaglyph glasses. Each filter lets through a different range of colours, so a standard RGB computer monitor can deliver a different image to each eye.

The team designed it to clip onto Neon’s “I can see clearly now” frame, which comes with magnetically swappable prescription lenses.

The authors estimate a printing cost of around US$1 for the attachment, with coloured filter sheets added separately. The printable design is available through the researchers’ OSF project.

Testing Against Established Laboratory Setups

How would Neon fitted with the low-cost anaglyph add-on compare with established laboratory equipment?

The team tested it against two setups: a mirror stereoscope and electronic shutter glasses, each paired with an EyeLink 1000 stationary eye tracker.

Binocular rivalry occurs when the two eyes receive competing images that the brain cannot easily combine. Your perception can alternate between them: one image dominates, then the other.

Figure 2: A participant wearing Neon fitted with the anaglyph add-on while completing the binocular rivalry task. Courtesy of Dr. Sabine Grimm.

The team used this effect to compare the three setups. They showed one eye stripes moving left and the other eye stripes moving right. The stripes kept moving in those directions, while participants’ perception could switch between leftward and rightward motion.

Eleven young adults tried all three setups. As they watched the moving stripes, they pressed buttons to report which direction they saw. At the same time, the eye trackers recorded a telltale movement: the eyes slowly followed the perceived motion, then flicked back and began following again. This repeating movement is called optokinetic nystagmus, or OKN.

The researchers excluded blinks and quick eye movements to isolate the slow tracking movements. They then compared their direction with the button presses to see how accurately each setup indicated what participants reported seeing.

The Low-Cost Add-On Performed Similarly to the Laboratory Setups

The main finding was that eye movements recorded with Neon and the anaglyph add-on revealed participants’ reported direction of motion with similar accuracy to the two established laboratory setups.

The direction of the slow eye movements tended to change before participants pressed a button to report seeing the new direction. To account for that delay, the researchers adjusted the timing of the eye-movement data relative to the button presses, finding the closest match for each participant.

After this adjustment, the accuracy score was 82% for Neon with the add-on, compared with 83% for the mirror stereoscope and 86% for shutter glasses. The measure, called “balanced accuracy”, gives equal weight to correctly identifying reports of leftward and rightward motion.

The study found no statistically significant difference in accuracy between the setups.

Toward More Accessible Stereoscopic Eye Tracking

A compact setup combining Neon, a printable attachment and a standard RGB monitor offers a very accessible way to combine stereoscopic presentation with eye tracking. This could make experiments easier to bring to children and to participants whose mobility or comfort makes conventional laboratory setups difficult to use. It could also support no-report studies, in which researchers infer what participants perceive from their eye movements without asking them to press buttons.

A Shared Design That Researchers Can Build On

The team achieved a useful measure of perception with a remarkably simple addition to a wearable eye tracker. By sharing the printable design, they have also made it easier for others to build on the idea. Pupil Labs’ reference CAD models provide a starting point for further Neon integrations. There is something appealing about that economy of design: a small, inexpensive part opens up new uses for equipment researchers already have.

Further Resources