Can bulls see red? How cattle see color

Jacobs, Deegan and Neitz (1998) measured the cones of cattle, goats and sheep with electroretinogram flicker photometry and found two cone mechanisms in each species. Across the three species the short-wave cone peaks at about 444 to 455 nm and the long-wave cone at about 552 to 555 nm. Each species has the basis for dichromatic color vision. Human eyes have three cone types, peaking at about 420, 534 and 563 nm (Bowmaker and Dartnall, 1980).

The abstract of the study gives these ranges for all three species together. The chart uses the middle of each range, 450 and 554 nm; we have not seen separate values for cattle alone.

A bull's long-wave cone responds to red light, but with one cone in that part of the spectrum red and green can give a similar color signal. The table runs a few pairs through the same model as the simulator and gives the color difference (ΔE 2000) for a person and for the cattle model: the smaller the number, the harder the pair is to tell apart. On the scale we use across this site, below 3 is a close match, 6 or more is a visibly different color, and 10 or more is clearly different. In the model red and green lose their difference in hue; what remains is mostly lightness, the red looking darker. Blue against green stays far apart in the model, yet calves in the test by Phillips and Lomas (2001) had limited ability to tell blue from green (see the questions below): the model and the behavior do not fully agree. The swatches are illustrative colors, not measured fabrics.

The cone studies say what a bull can tell apart, not what makes it charge. Why bulls react in a ring is a question about behavior that these sources do not answer.

The simulator is built from the cattle's own cone peaks (about 450 and 554 nm, the middle of the ranges). For each color it estimates how strongly it would excite each of the two cones, using the pigment template of Govardovskii and colleagues (2000) and daylight (CIE D65), and shows a color that excites them in the same way. The cone signals weigh the light by its energy; a cone counts photons, and weighting by photon count instead would shift some shown colors slightly, by up to about 16 of 255 in one color channel by our check. The shown colors lie on a yellow–blue axis plus lightness, the same plane that Viénot, Brettel and Mollon (1999) use for people with two cones, so colors that give the cattle the same pair of cone signals come out identical. It is an approximation: the eye's lens is left out, each color's spectrum is a guess from its screen color, and the picture is blurred to the animal's visual acuity, as explained below. Everything runs on your device: the camera picture, photos and videos are not uploaded.

Rehkämper and Görlach (1998) trained three adult dairy bulls to tell a black disc from a black ring with a white center, seen from at least 1.5 m. The bulls still noticed a white center 1 cm across, which the authors give as a visual angle of 23 minutes of arc. As stripes that is about 1.3 cycles per degree (our conversion: 30 ÷ 23). A review by Adamczyk and colleagues (2024) gives 1.3 to 2.5 cycles per degree for cattle, from this study and an earlier one with heifers.

The simulator blurs the picture to 1.3 cycles per degree, the most of the five animals. We read only the summary (abstract) of the Rehkämper and Görlach paper; the review we read in full.

How the blur is made: the simulator follows AcuityView, the method of Caves and Johnsen (2018). Fine detail is faded along a bell-shaped curve that leaves about 3% of the contrast at the animal's acuity limit, and coarse shapes and colors stay. A picture does not say how wide the scene was, so the simulator assumes that the long side of the picture covers about 70°, the view of a typical phone's main camera (the iPhone 16 main camera is 26 mm in 35 mm terms, about 69° across). If you zoomed in or cropped, the scene was narrower and the animal would see it blurrier than shown. The blur stands for the sharpest part of the eye only; the rest of the animal's view is not modelled.