# Clay & Cortex: results so far (interim, 2026-09-12)

Machine-readable twin of https://clayandcortex.com/results. The same numbers as https://clayandcortex.com/results.json. Regenerated from the study database by site/scripts/build-results.py, using the studies' own report code, so the page, this file and the investigator's report cannot disagree.

## Caveats, which travel with every number

- Interim data from small samples: one session per person in most cases, experience self-reported, no peer review. Not a finding.
- Each threshold is a 75% correct point on a psychometric function fitted by maximum likelihood; a threshold is counted only when the fit converged and lies inside the range of values actually presented.
- A threshold below the smallest value the task can present is reported as "at the floor", never as a number.
- Sessions on a display slower than the protocol requires are excluded and counted; the studies now refuse such a display before the first trial.
- Trials rejected for timing (frames delivered late) are excluded from every threshold.
- No performance claim is made: nothing here shows that any lens, tool or training improves scores.
- Sessions are S01, S02, ... in order of upload. No participant identifier, account or email is present in any file.

## Trap occlusion study

A clay leaves the trench at a small angle left or right of straightaway, is visible for a fixed few tens of milliseconds, then is hidden; the observer calls left or right. An adaptive procedure (ZEST) finds, at each viewing time, the smallest launch angle called correctly 75% of the time. That angle is the threshold; smaller is better.

- Sessions: 8 shown of 10 uploaded (2 excluded by the display rule), from 8 participants, 2026-08-05 to 2026-08-29.
- Participants by self-reported experience: Elite / international 4, Club / developing 2, Novice / non-shooter 1, Not stated 1.
- Trials: 2,544 main trials, 171 rejected for timing; 144 lapse probes, 0 missed.
- Thresholds: 26 usable of 34 fitted; 8 at the task's floor (below about 0.1 degrees).
- Smallest launch angle called so far: 0.094 degrees at 67 ms of visible flight.

### Median threshold by viewing time (all shown sessions)

| Visible for | Median threshold | n | Distance at occlusion | Sideways drift |
|---|---|---|---|---|
| 33 ms | 0.426 deg | 8 | 16.9 m | 12.5 cm |
| 50 ms | 0.227 deg | 7 | 17.3 m | 6.9 cm |
| 67 ms | 0.137 deg | 5 | 17.8 m | 4.3 cm |
| 100 ms | 0.258 deg | 4 | 18.6 m | 8.4 cm |

### Median threshold by viewing time and experience (cells with two or more thresholds)

| Visible for | Group | Median threshold | n | At floor |
|---|---|---|---|---|
| 33 ms | Club / developing | 0.602 deg | 2 | 0 |
| 33 ms | Elite / international | 0.283 deg | 4 | 0 |
| 50 ms | Club / developing | 0.353 deg | 2 | 0 |
| 50 ms | Elite / international | 0.227 deg | 3 | 1 |
| 67 ms | Club / developing | 0.292 deg | 2 | 0 |
| 67 ms | Elite / international | 0.11 deg | 2 | 2 |
| 100 ms | Elite / international | 0.133 deg | 2 | 2 |

### Every session shown

| Session | Group | Protocol | Visible for | Threshold | 95% CI | Trials |
|---|---|---|---|---|---|---|
| S01 | Club / developing | 2.0.0 | 33 ms | 0.751 deg | 0.213 to 2.056 | 30 |
| S01 | Club / developing | 2.0.0 | 50 ms | 0.548 deg | 0.188 to 1.409 | 32 |
| S01 | Club / developing | 2.0.0 | 67 ms | 0.622 deg | 0.188 to 2.483 | 25 |
| S01 | Club / developing | 2.0.0 | 100 ms | 0.426 deg | 0.156 to 2.331 | 33 |
| S01 | Club / developing | 2.0.0 | 141 ms | 0.311 deg | 0.069 to 0.8 | 31 |
| S02 | Elite / international | 2.0.0 | 33 ms | 0.376 deg | 0.177 to 0.852 | 60 |
| S02 | Elite / international | 2.0.0 | 50 ms | 0.242 deg | 0.121 to 0.454 | 60 |
| S02 | Elite / international | 2.0.0 | 67 ms | 0.094 deg | 0.05 to 0.188 | 60 |
| S02 | Elite / international | 2.0.0 | 100 ms | 0.114 deg | 0.053 to 0.227 | 59 |
| S02 | Elite / international | 2.0.0 | 141 ms | 0.1 deg | 0.05 to 0.2 | 59 |
| S03 | Novice / non-shooter | 3.0.0 | 33 ms | 2.816 deg | 1.409 to 5.628 | 74 |
| S03 | Novice / non-shooter | 3.0.0 | 50 ms | 1.409 deg | 0.751 to 2.644 | 73 |
| S03 | Novice / non-shooter | 3.0.0 | 67 ms | 1.095 deg | 0.584 to 2.056 | 70 |
| S03 | Novice / non-shooter | 3.0.0 | 100 ms | 0.622 deg | 0.331 to 1.242 | 72 |
| S04 | Elite / international | 3.0.0 | 33 ms | 0.213 deg | 0.107 to 0.4 | 75 |
| S04 | Elite / international | 3.0.0 | 50 ms | 0.227 deg | 0.114 to 0.454 | 75 |
| S04 | Elite / international | 3.0.0 | 67 ms | 0.129 deg | 0.061 to 0.227 | 75 |
| S04 | Elite / international | 3.0.0 | 100 ms | at floor, below 0.11 deg | - | 75 |
| S05 | Club / developing | 3.0.0 | 33 ms | 0.483 deg | 0.258 to 0.966 | 75 |
| S05 | Club / developing | 3.0.0 | 50 ms | 0.227 deg | 0.121 to 0.454 | 75 |
| S05 | Club / developing | 3.0.0 | 67 ms | 0.137 deg | 0.069 to 0.258 | 75 |
| S05 | Club / developing | 3.0.0 | 100 ms | at floor, below 0.119 deg | - | 75 |
| S06 | Elite / international | 3.0.0 | 33 ms | 0.548 deg | 0.258 to 1.095 | 75 |
| S06 | Elite / international | 3.0.0 | 50 ms | at floor, below 0.118 deg | - | 75 |
| S06 | Elite / international | 3.0.0 | 67 ms | at floor, below 0.113 deg | - | 75 |
| S06 | Elite / international | 3.0.0 | 100 ms | 0.156 deg | 0.083 to 0.292 | 75 |
| S07 | Not stated | 3.0.0 | 33 ms | 0.311 deg | 0.166 to 0.548 | 75 |
| S07 | Not stated | 3.0.0 | 50 ms | 0.188 deg | 0.094 to 0.353 | 75 |
| S07 | Not stated | 3.0.0 | 67 ms | at floor, below 0.119 deg | - | 75 |
| S07 | Not stated | 3.0.0 | 100 ms | at floor, below 0.114 deg | - | 74 |
| S10 | Elite / international | 3.0.0 | 33 ms | 0.177 deg | 0.094 to 0.331 | 75 |
| S10 | Elite / international | 3.0.0 | 50 ms | 0.114 deg | 0.057 to 0.213 | 74 |
| S10 | Elite / international | 3.0.0 | 67 ms | at floor, below 0.114 deg | - | 75 |
| S10 | Elite / international | 3.0.0 | 100 ms | at floor, below 0.1 deg | - | 75 |

## Lens tint and clay contrast study

Contrast thresholds on three cone-contrast axes and on a clay-sized moving disc, the red-green isoluminant point by heterochromatic flicker photometry, then, for six lens classes presented blind as cone-contrast vectors, the fraction of each lens's real clay-background contrast at which the observer can just tell the direction of motion. The physical model predicts that fraction from the observer's own thresholds; the comparison is the study's point.

- Sessions: 2 shown of 2 uploaded; 1 under the current protocol and pooled, 1 under an earlier protocol, shown but not pooled; 0 excluded by the display rule. 2 participants, 2026-09-11 to 2026-09-12.
- Trials: 834 main trials, 7 rejected for timing.
- Thresholds: 18 usable of 18 fitted; 0 at floor.
- Mean rank agreement between the model's lens order and the observer's (Kendall tau, pooled sessions): 0.333; the study's gate for a personal recommendation is 0.6.
- Median ratio of the measured threshold multiplier to the model's prediction, pooled sessions: 3.32 (1.0 would be a perfect model; above 1 means the moving clay-sized target was harder than the static thresholds imply).

### Lens classes: measured against predicted threshold multiplier (pooled sessions; lower is better)

| Lens | Measured (median) | Predicted (median) | n |
|---|---|---|---|
| clear | 0.163 | 0.054 | 1 |
| light-orange | 0.264 | 0.051 | 1 |
| orange | 0.182 | 0.047 | 1 |
| vermilion | 0.073 | 0.047 | 1 |
| purple | 0.1 | 0.029 | 1 |
| bronze | 0.163 | 0.05 | 1 |

### Every session shown

| Session | Group | Protocol | Pooled | Kendall tau | Measure or lens | Threshold | Predicted | Trials |
|---|---|---|---|---|---|---|---|---|
| S01 | Clay shooter | 0.1.0 | no | 0.429 | B:lum | 0.0583 | - | 36 |
| S01 | Clay shooter | 0.1.0 | no | 0.429 | lens: vermilion | x0.077 | x0.069 | 31 |
| S01 | Clay shooter | 0.1.0 | no | 0.429 | lens: clear | x0.21 | x0.174 | 32 |
| S01 | Clay shooter | 0.1.0 | no | 0.429 | lens: yellow | x0.236 | x0.187 | 32 |
| S01 | Clay shooter | 0.1.0 | no | 0.429 | lens: purple | x0.139 | x0.151 | 32 |
| S01 | Clay shooter | 0.1.0 | no | 0.429 | lens: orange | x0.247 | x0.203 | 32 |
| S01 | Clay shooter | 0.1.0 | no | 0.429 | lens: bronze | x0.232 | x0.226 | 32 |
| S01 | Clay shooter | 0.1.0 | no | 0.429 | lens: light-orange | x0.205 | x0.258 | 32 |
| S02 | Clay shooter | 1.0.0 | yes | 0.333 | A:lum | 0.0219 | - | 35 |
| S02 | Clay shooter | 1.0.0 | yes | 0.333 | A:lm | 0.0067 | - | 36 |
| S02 | Clay shooter | 1.0.0 | yes | 0.333 | A:s | 0.0742 | - | 35 |
| S02 | Clay shooter | 1.0.0 | yes | 0.333 | B:lum | 0.0776 | - | 48 |
| S02 | Clay shooter | 1.0.0 | yes | 0.333 | lens: vermilion | x0.073 | x0.047 | 40 |
| S02 | Clay shooter | 1.0.0 | yes | 0.333 | lens: clear | x0.163 | x0.054 | 40 |
| S02 | Clay shooter | 1.0.0 | yes | 0.333 | lens: purple | x0.1 | x0.029 | 40 |
| S02 | Clay shooter | 1.0.0 | yes | 0.333 | lens: orange | x0.182 | x0.047 | 40 |
| S02 | Clay shooter | 1.0.0 | yes | 0.333 | lens: bronze | x0.163 | x0.05 | 40 |
| S02 | Clay shooter | 1.0.0 | yes | 0.333 | lens: light-orange | x0.264 | x0.051 | 40 |

## Citing

Kumar, S. (2026). Clay & Cortex: open psychophysics studies of vision in clay shooting. https://clayandcortex.com/ (results page: https://clayandcortex.com/results, retrieved on the date of access).

Run by Dr. Satwant Kumar, MBBS, PhD, vision and cognitive neuroscientist. Contact: satwant@asperity.ai. A guide for agents: https://clayandcortex.com/llms.txt