🔬 optics

Test-retest Reliability of Psychophysical Tasks Using Structured Light-Induced Entoptic Phenomena

This study demonstrates that a psychophysical task using structured light-induced entoptic phenomena is a reliable method for measuring polarization perception, showing high test-retest reliability (ICC up to 0.93) that could potentially aid in the early detection of macular diseases.

Original authors: Taranjit Singh, Mukhit Kulmaganbetov, Zhangting Wang, Dmitry Pushin, Benjamin Thompson, Andrew Silva, Melanie Mungalsingh, Iman Salehi, Davis Garrad, David Cory, Dusan Sarenac

Published 2026-02-10

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Taranjit Singh, Mukhit Kulmaganbetov, Zhangting Wang, Dmitry Pushin, Benjamin Thompson, Andrew Silva, Melanie Mungalsingh, Iman Salehi, Davis Garrad, David Cory, Dusan Sarenac

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The "Glitter in Your Eye" Test: A Simple Guide to This Research

Imagine you are looking at a piece of silk or a shimmering soap bubble. Sometimes, when you look at certain lights, you see tiny, ghostly patterns dancing in your vision—like little sparkles or faint "brushes" of light. These aren't actually in the room; they are happening inside your eye.

Scientists call these "entoptic phenomena." Think of them like the "glitch in the Matrix" of your own biology. Because these patterns are created by the way light hits the delicate layers of your retina (the "film" at the back of your eye), they can act like a secret messenger, telling us how healthy your eyes are.

The Problem: The "Flickering Candle" Challenge

For a long time, doctors have tried to use these tiny light patterns to check for eye diseases (like macular degeneration). But there’s a problem: these patterns are incredibly faint, shy, and hard to see. It’s like trying to measure the exact height of a flickering candle flame in a dark room—it’s so inconsistent that it’s hard to tell if a change in the flame is because the candle is melting or if you’re just blinking.

The Innovation: The "High-Tech Spotlight"

This research team decided to stop using "flickering candles" and started using Structured Light (SL).

Instead of a faint, natural glow, they used a specialized light projector to create a very specific, organized pattern of light (like a high-definition, rotating geometric shape). This is like replacing that tiny candle with a bright, rotating disco ball. It makes the "glitch" much easier to see and much more precise to measure.

The Experiment: The "Repeat Performance" Test

The big question the researchers asked was: "If we give a person this test today, and then give them the exact same test next week, will we get the same answer?"

In science, this is called "Test-Retest Reliability." If a bathroom scale tells you that you weigh 150 lbs, but then tells you that you weigh 180 lbs five minutes later, that scale is useless. It’s not "reliable."

Here is what they did:

They took 25 people with healthy eyes.
They gave them a task: "Watch this rotating pattern and tell us which way it’s spinning."
They measured exactly how much of the pattern the person could see (the "threshold").
They came back 1 to 14 days later and did it all over again.

The Results: A "Gold Standard" Score

The results were excellent. Using a statistical tool called an ICC (Intraclass Correlation Coefficient), they found that the scores were incredibly consistent.

The Analogy: Imagine you are throwing darts at a bullseye. If you throw a set of darts today, and then come back next week and throw another set, you hit almost the exact same spots both times. You didn't get "better" because you practiced (the "learning effect"), and you didn't get "worse" because you were tired. You were consistent.

The researchers found that the "score" for each eye stayed nearly the same between the first and second visits. This means the test isn't just a random guess—it’s a stable, repeatable measurement.

Why Does This Matter?

Because this test is so reliable, it can be used as a "Early Warning System."

If this test is consistent for healthy eyes, then the moment a person’s score starts to change, doctors will know it’s not just a "glitch" or a bad day—it’s a real sign that something is changing in the health of their retina. It’s like having a highly sensitive smoke detector that only goes off when there is an actual fire, rather than just reacting to a bit of steam from the shower.

In short: They have built a high-tech, highly reliable "eye-health thermometer" that could help catch vision-threatening diseases much earlier than ever before.

Technical Summary: Test-retest Reliability of Psychophysical Tasks Using Structured Light-Induced Entoptic Phenomena

Problem Statement

The early detection of macular diseases, such as age-related macular degeneration (AMD), requires sensitive tools to assess functional changes in the retina. While the perception of polarized light (Haidinger’s brushes) is a known indicator of macular health, traditional methods are clinically limited due to the faintness, transience, and low contrast of the percept. Existing psychophysical assessments, such as perimetry or heterochromatic flicker photometry (for MPOD), often suffer from high variability caused by cognitive load, attention, and environmental factors. There is a critical need for a more robust, detectable, and reliable method to quantify polarization perception for clinical screening.

Methodology

The study investigated the test-retest reliability of a novel psychophysical task utilizing Structured Light (SL) to induce entoptic patterns.

Stimuli: The researchers used SL-based stimuli to generate a rotating entoptic pattern consisting of 22 azimuthal brushes. This method provides a larger and more detectable percept compared to traditional polarization methods.
Task Design: Participants were required to perform a forced-choice discrimination task, identifying the direction of rotation (clockwise or counter-clockwise) of the pattern.
Measurement ( $R_T$ ): To quantify the perceivable area, the researchers used a spatial light modulator (SLM) to create a central obstruction (mask). The radius of this mask was adjusted using a 2-up/1-down staircase method to determine the retinal eccentricity threshold ( $R_T$ ), which represents the visual angle of the perceivable pattern.
Participants: 25 participants (50 eyes) with healthy vision (logMAR BCVA $\approx$ 0.00) aged 21–75 underwent two identical testing sessions separated by 1 to 14 days.
Statistical Framework: Reliability was assessed using:
- Intraclass Correlation Coefficient (ICC): A two-way mixed-effects, single-rater model (ICC(3,1)).
- Bland-Altman Analysis: To assess mean bias and limits of agreement between sessions.
- Pearson Correlation: To evaluate the strength of the association between the first and second sessions.

Key Contributions

Validation of SL-based Entoptic Tasks: The study provides empirical evidence that structured light can be used to create stable, quantifiable, and repeatable psychophysical stimuli for polarization perception.
Quantification of Perceptual Thresholds: By using a staircase method with an SLM, the study demonstrates a way to precisely map the visual angle of entoptic phenomena ( $R_T$ ).
Reliability Benchmarking: The research establishes a baseline for the repeatability of these measurements in healthy populations, which is a prerequisite for any future clinical diagnostic tool.

Results

The study reported high levels of consistency between the two testing sessions:

Reliability (ICC): The ICC values were 0.83 [95% CI: 0.62–0.93] for the right eye (RE) and 0.93 [95% CI: 0.84–0.97] for the left eye (LE), which classifies the method as having "good to excellent" reliability.
Agreement (Bland-Altman): The mean differences between sessions were minimal ( $-0.32^\circ$ for RE and $-0.12^\circ$ for LE), indicating negligible systematic bias. The limits of agreement were within a narrow range, confirming strong agreement.
Correlation: Pearson coefficients were high ( $r = 0.71$ for RE; $r = 0.87$ for LE) with significant p-values ( $p < 0.01$ ).
Learning Effects: The minimal mean difference ( $-0.22^\circ$ ) and the fact that 40% of participants actually showed a lower $R_T$ in the second session suggest that no significant "learning effect" influenced the results.

Significance

This research demonstrates that SL-based psychophysical tasks are a highly reliable method for assessing polarization perception. Because these tasks provide a quantitative measure of the retinal area capable of perceiving structured light, they offer a promising pathway for developing non-invasive screening tools for macular health. If these methods can be successfully applied to patients with early-stage macular disease, they could serve as a functional biomarker for detecting the onset of degenerative conditions.