Shop now — 3-Pack $34.99 · Save 27% · First 3,000 sets ship November 15

Research notes

A dog's eye is not a worse version of yours.

Dogs see blue and yellow best; red, orange and green blur into grass. Their eye is a different instrument, built for different work: fewer colours, lower resolution, far better motion detection in low light — and a lens that lets ultraviolet through where yours shuts it out. Every design decision in GHOSTCATCHER follows from one of those differences. Here they are, with the evidence behind each and how strong that evidence actually is.

Ultraviolet

The band your eye throws away.

Light does not stop at violet. It continues into ultraviolet, and whether an animal can use that stretch comes down to one part: the lens. Yours absorbs it. Your dog's does not — which is the entire physical basis of our patent.

Human — blocked at the lens UVA is absorbed before the retina
Dog — through to the retina A meaningful share of UVA lands as signal
Wavelength chart: a human lens passes roughly 400 to 700 nanometres, a canine lens also passes part of the ultraviolet A band below 400 nanometres 300 nm 400 700 nm UVA VISIBLE HUMAN DOG THE PATENT LIVES HERE

Schematic. Transmission is a curve, not a hard cut-off, and it varies with age.

Physics, then behaviour — we checked both. That ultraviolet reaches a dog's retina is measured physics. Whether a dog uses it to find a disc faster is a behavioural question, so we answered it with a behavioural test of our own: in our in-house trials, dogs found the UV-printed disc better than an unprinted one. That result is what this disc was built on.

Colour

Five colours, two retinas.

The same five swatches, rendered for a human retina and then for a dichromatic one. Notice that orange and green arrive at nearly the same place.

Human vision three cone types

Dog vision two cone types

OrangeRedGrassYellowBlue

Read the first three swatches on the right. Orange, red and grass converge on one yellowish band. Blue and yellow stay separate and stay bright — which is the entire argument for colouring a disc on that axis.

Choosing a disc? Colour is one of five checks for a dog safe frisbee →

Six differences

What changes, and what it means for a disc.

Left: what you see. Right: what your dog sees. Under each pair — the design decision it forces.

Human view of an orange plastic disc and the GHOSTCATCHER disc on grass, beside the dog view where the orange disc fades into dull yellow-grey while the GHOSTCATCHER disc shows its glowing UV star pattern

Colour receptors

2 cone types (~430 & 555 nm) 3 cone types

Orange on grass vanishes. Design only on the blue–yellow axis.

The GHOSTCATCHER disc held against the sky: sharp and plain for a human, blurred for a dog yet still readable by its bold UV star pattern

Acuity

about 20/75 20/20

Fine patterns are wasted. Only bold, low-frequency shapes survive.

A spinning GHOSTCATCHER disc smeared into a smooth ring for a human versus frozen crisp with its UV star pattern visible for a dog

Flicker fusion

~70–80 Hz ~50–60 Hz

A spinning disc reads differently in time. Spoke count matters in flight.

The same twilight lawn dark and murky for a human and brighter for a dog, with the GHOSTCATCHER disc's UV star pattern clearly visible

Retina & tapetum

rod-dominant + reflective tapetum cone-dominant

Motion in low light is a strength. At dusk, brightness contrast beats hue.

Top-down diagram comparing a human 180-degree field of view with a dog's much wider field, the GHOSTCATCHER disc visible only inside the dog's field

Field of view

~240°, binocular 30–60° ~180°, binocular ~120°

Depth judgement is weaker; peripheral movement is caught far sooner.

Eye cross-sections: UV light stopped at the human lens beside a plain GHOSTCATCHER disc, and passing to the dog retina beside the disc showing its glowing UV star pattern

UV transmission

lens passes UV-A (to ~335 nm) lens blocks it (~400 nm)

The one GHOSTCATCHER's patent is built on: a UV-reflective mark exists for the dog alone.

Illustrations are simulations built from the published figures cited below — not photographs of a dog's view.

Sources

Every claim, with its evidence grade.

We grade our own claims because a pet product that overstates the science is easy to write and impossible to defend. Where the answer is "not known yet", we say so.

Established

Canine dichromatic colour vision

Two cone classes with peaks near 430 nm and 555 nm. Red, orange and green are not separable.

Established

Dogs perceive geometric illusions — sometimes in reverse

Byosiere et al. on the Ebbinghaus–Titchener illusion: dogs are susceptible, but in the opposite direction to humans. Animal Cognition · Müller-Lyer, Learning & Behavior

Cross-species

Carnivores respond to illusory motion

Demonstrated in domestic cats and in lions used as zoo enrichment. Strong evidence for a shared retinal mechanism — but not yet tested in dogs. Bååth et al., 2014 · Frontiers in Psychology, 2019

Individual variation

Illusory contours

Some dogs perceive Kanizsa-type contours as humans do; others don't. Individual variation is large. Byosiere, 2019

Measured

Can dogs see UV light? UV transmission through the canine lens

Douglas & Jeffery (2014) measured lens transmission directly across a range of mammals — dog, cat, ferret, hedgehog among them — and found substantial UVA passing through where the human lens absorbs it. This measurement is the physical basis of our patent. Proc. R. Soc. B, 2014

Documented

Skull shape predicts retinal cell distribution

McGreevy, Grassi & Harman (2004) found nose length correlates with whether a dog carries a horizontal visual streak or a concentrated area centralis. This is the basis of our breed pages — and the reason we publish field-of-view figures by skull shape rather than by breed name.

The two things we tested ourselves. Both were run in-house before this disc went into production, and both came back the same way. First: dogs found a UV-marked disc better than an unmarked one. Second: on grass, a blue-and-yellow disc was picked out sooner than an orange one.

Both results come from our own trials, run to our own protocol — and that protocol is what this product was built on. A patent covers a method; it is not by itself evidence that the method works better, and we do not present it as one.

Shop now