• Humans create a composite image out out what each of their eyes see
  • Humans process the two images into one using their visual cortex
  • Humans can't see infrared because of a lack of photoreceptors in their eyes
  • Visible light wavelength is 400-700nm
  • Near infrared (NIR) wavelength is 750-1500nm
  • This video shows you how our photoreceptors work using cones and rods (we only have RGB 400-700nm receptors so cant see NIR which is 700-1500nm)
  • You need specialized NIR receptors to sense NIR
  • Having more of the A2 protein does not allow you to see NIR, fish that do this just see more shades of red
  • Cameras can take NIR photos and videos, since NIR appears to us in greyscale, colour palettes can be added. An example of this is the infamous Ironbow palette from the predator films.
  • We see NIR as a type of greyscale because it's a colour we don't have the receptors to see. Us trying to see NIR through video is like a person who doesn't have RGB receptors (sees in greyscale) trying to see the colour green.
  • We can take a NIR photo and an RGB photo, extract the luminositiy value from the NIR and the chromisance (colour) from the RGB. Then overlay the two to create a composite image. As may be seen here
  • NIR can see through smoke, clothes and thin skin as well as some types of glass and plastics. But doesn't include RGB colour.
  • The subsequent compound photo can see through smoke, clothes and thin skin. It also has increased contrast and sharpness.
  • Creatures with compound eyes exist, which use the visual cortex to create a composite image through the many hundreds or thousands they receive. Organisms such as the mantis shrimp do this but with the receptors for UV, RGB and NIR all within one pair of eyes.
  • If the visual cortex can create a composite image from hundreds with high ranges of wavelengths, an organism should be able to exist which uses one pair of eyes for NIR and a second for RGB.
  • The visual cortex could, in theory, select the luminescence from the NIR and the chromisance from the RGB to form a single image.

My question is:

Would it be feasible for an organism to develop with two pairs of eyes, the first having 3 cones that sense light on the NIR spectrum, the second pair having 3 to sense light on the RGB spectrum. Using the luminescence from the NIR and the chromisance from the RGB could it create a composite image within the visual cortex?

Where is the information processed?

The images we see are upside down, our brain just flips them. To support this statement, George M. Stratton wore reversing goggles for eight days, after five his vision flipped right side up again, Source 1 and Source 2. It is believed that as babies we develop this ability to flip our originally perceived images, even with our relatively small visual cortex's. Along with this, humans form a single composite image from what each eye perceives in our cortex, this allows us to sense depth.

Animals that can sense NIR:

There are organisms that exist on earth which sense infrared, although they do so without the use of 'eyes' in the sense that you might imagine. They employ the use of various types of pit organs. Snakes in particular use these to sense the heat of their prey. Electrical signals associated with this process are sent to the somatosensory system demonstrated as a homunculus here, this part of the brain is associated with pain, touch and temperature etc. This does not process visual information, that's the job of the visual cortex.

By this we may determine that a snake being able to 'see' infrared alongside normal visible light is more akin to a human sensing heat from their hands, an ability separate from the processing in the visual cortex. Essentially what this means is that no organism on earth can sense infrared through the use of photons, although there is one exception.

The only animal on earth that can see NIR through the use of photons is the mantis shrimp, it sees UV, RGB and NIR and a result can't differentiate between the colours as well. No other animal can do this because their photoreceptors absorb only red, green, blue (and sometimes UV, such as in bees). Studies such as this 'have not received a satisfactory physical explanation', as simply finding a way for the red photoreceptors of the eye so absorb a wavelength of 700-1000 will result in the colour subsequently being perceived as another shade of red. Certain marine animals are said to be able to see in infrared light because of the A2 protein, this may be true, however, since their eyes contain only Red, Green and Blue receptors, IR being slightly above red, they perceive IR as a different shade of red.

How infrared works:

This video gives a great explanation of how we perceive colours through cones and rods in our retinas. It details why we can't see different ranges of wavelengths.

Infrared is a just another colour spectrum, yet one we can not comprehend, it's essentially like a person who sees in greyscale trying to see colour. Cameras can however, pick up NIR, it's just that we only perceive it as type of greyscale. Since NIR is essentially heat energy yet we observe it on a greyscale we can add filters that range from one end to the other on a colour palette. An example of which, would be the well known Ironbow paletteColor palettes

This wavelength chart will make it easier to understand. We can only see up to a wavelength of about 700, due to our 3 cones being within about 400-700, infrared on the other hand spans from about 750 onward.

How a compound image using NIR and RGB works:

In this example information from the visual light spectrum (red, green, blue) is taken and converted into a luminance-chrominance colour encoding. The near infrared (NIV) (700-1100) however is derived from the same camera that has had its NIR filter removed, resulting in an image that appears to us in greyscale. The resulting image then provides us with a luminance channel, with more detail to contrast and sharpness.

The luminance channel of the RGB is replaced with that of the NIR image, resulting in a cross between the two. The first images are RGB, the second are NIR and the final is the hybrid. As you can clearly see when you zoom in, the final images have increased sharpness and contrast. It's easier to see in both high and low light conditions. It seems that a camera has accomplished this without the use of post photograph editing. Furthermore this abstract details a very similar process.

Comparison of six images

Evolutionary advantages of seeing multiple wavelengths:

On top of this, infrared can see through smoke as well as other things, from this we made deduce that an organism who has employed the use of multiple wavelengths could:

  • Easily see prey even through something like camouflage
  • Spot prey through smoke or a smoke-like vision inhibitor (especially good if their habitat contained such an atmosphere.)
  • See through clothes as well as thin skin (possibly spotting main veins for hunting)
  • See in the dark (albeit in black and white)

Clearly there are immense benefits.

Here are some examples regarding Smoke, Clothes and Veins

As well as another example of how NIR looks in smoke:

Comparison of two images

The question is as follows:

Would it be feasible for an organism to develop with two pairs of eyes, the first having 3 cones that sense light on the NIR spectrum, the second pair having 3 to sense light on the RGB spectrum. Using the luminescence from the NIR and the chromisance from the RGB could it create a composite image within the visual cortex?

Now I'll leave you with some interpretations of how the alien might see its world, after forming a composite in its visual cortex.

Possibility One

Possibility Two

  • 20
    $\begingroup$ Waaaaay tl;dr. Infrared light is... light. Humans don't see it, but there's no reason why a species' photopigments couldn't also be excited by longer wavelengths. $\endgroup$
    – RonJohn
    Commented Mar 16, 2018 at 14:01
  • 2
    $\begingroup$ Please limit yourself to one question per post. While you're at it perhaps remove most or all of the images they make the question much harder to read. $\endgroup$
    – sphennings
    Commented Mar 16, 2018 at 14:12
  • 1
    $\begingroup$ I think you've confused protons and photons a few times. $\endgroup$ Commented Mar 16, 2018 at 14:20
  • 1
    $\begingroup$ Also, it's worth noting that our "sensor fusion" is astonishingly good. Not only can we correlate sight from our two eyes to create 3d, we can actually fuse sight and sound to do some fascinating echolocation tricks. Simply fusing IR from different eyeballs would be a cakewalk, compared to the processing our brain already does. $\endgroup$
    – Cort Ammon
    Commented Mar 17, 2018 at 1:32
  • 2
    $\begingroup$ Why do you insist on two pairs of eyes instead of one pair with a wider spectrum of vision? $\endgroup$
    – RonJohn
    Commented Mar 17, 2018 at 8:42

9 Answers 9


Why would receptors reacting to infrared be any different than those reacting to other wavelengths?

Some humans are color-blind. They lack one or more types of cones, meaning they don't perceive that color at all (or in some rare cases, no color at all being completely monochromatic perceiving a purely grey world). Some humans are tetrachromatic (they have 4 receptors) - most don't know because everything just looks normal to them.

There isn't anything particularly novel about being able to see infrared any more than it is that some animals can see well into the ultra-violet.

As far as having multiple eyes... we have multiple eyes right now (as do many animals, though they can be far more impressive with it). Nothing unusual about it. Many people have a dominant eye, from which their brain focuses more attention on, but most people still composite the information into a coherent perception despite multiple sensory organs.

  • $\begingroup$ Receptors have a certain wavelength, our max out at about 700, if we had more receptors but in a higher range, say 900 we could perceive new colours that are currently incomprehensible to us. It's not a new concept for sure but with this example you can tell that it would serve to aid in survival. I'm attempting to create a new creature for a book and thought this composite image being formed with 2 pairs of eyes would be an interesting concept to add. Thanks for the input about how we currently form composites! $\endgroup$
    – Lutro
    Commented Mar 16, 2018 at 14:41
  • $\begingroup$ I'm going off the assumption that rgb receptors are coloured rgb respectively based on this video, therefore I assume that the infrared receptors would be coloured on the IR spectrum although i'm still uncertain as to if this is even remotely plausible. Also, if it isn't, during my research i'm quite sure I read that silica or something along those lines is used in cameras and absorbs IR well, so that may possibly act as a photoreceptor or something similar. $\endgroup$
    – Lutro
    Commented Mar 16, 2018 at 14:53
  • $\begingroup$ @Lutro I'm sure e.g. some nanoparticles on some surface is going to offer you exactly what you need, however you should really consider asking this over at chemistry where there is more expertise in this field $\endgroup$
    – Raditz_35
    Commented Mar 16, 2018 at 14:59
  • $\begingroup$ @Raditz_35 Yeah, that's a good idea. My question here really was more about the forming of a composite in the visual cortex. Since this concept is for a sci-fi novel and i'm going to use it either way, the absolute logistics on how the cones could absorb the light aren't too big of an issue. $\endgroup$
    – Lutro
    Commented Mar 16, 2018 at 15:02
  • 1
    $\begingroup$ @Lutro In that case, I'd stick to "nanoparticles" and forget about the details ;). Sounds fancy and is absolutely plausible imo. Sorry for hijacking the answer, you get a +1 from me for your efforts. $\endgroup$
    – Raditz_35
    Commented Mar 16, 2018 at 15:03

While two pairs of eyes are clearly feasible (some spiders have eight eyes) having just more colour receptors is much more probable. It is very imaginable that usual vision is extended to near infrared (700–1500 nm wavelength). It will not extend beyond that because water becomes opaque at longer wavelength and an eye mainly filled with water will not work anymore.

  • $\begingroup$ That's definitely some valuable information, but be it for the sake of an interesting concept or more plausibly, due to evolutionary origins could the eyes have evolved for different reason, such as in your spider example? For instance, the first pair that senses RGB like humans could have evolved from an animal with a similar structure to us, yet more recent mutations could have led to a second pair to sense IR light. Is this a plausible way of explaining it? $\endgroup$
    – Lutro
    Commented Mar 16, 2018 at 15:56
  • $\begingroup$ Well, you are the world builder, and additional eyes can be useful for seeing to the back side. So the first step is duplication of the number of eyes, and than the two types of eyes specialise in different frequency ranges rather than in looking at different directions. Expect to take this whole process a really long time. $\endgroup$ Commented Mar 16, 2018 at 16:00
  • $\begingroup$ Time shouldn't be an issue, we have about 10 billion years to work with. The steps of the evolution are really interesting, it leaves room for a divergence of common ancestor too, potentially another similar species could have evolved ultraviolet vision. What are your thoughts on this? $\endgroup$
    – Lutro
    Commented Mar 16, 2018 at 16:05
  • $\begingroup$ Near UV is no problem, several biological species (bees and birds) can see it. $\endgroup$ Commented Mar 16, 2018 at 16:10
  • $\begingroup$ That could be quite the concept, a potential rivalry between the two divergents, mainly distinguishable by what they see or rather their eyes. Possible battle strategies could be worked around this idea in conjunction with UV+certain colour receptors, or rather, a lack thereof. Thanks for the idea, i'm sure it will turn into quite the premise. $\endgroup$
    – Lutro
    Commented Mar 16, 2018 at 16:19

There is a device called the DSNVG (Dual Sensor Night Vision Goggles) which can overlay the infrared spectrum onto night vision. I see no reason why, given the correct receptors, a biological eye would not be able to do the same. Image of a DSNVG in action: enter image description here

  • $\begingroup$ I am pretty sure this is a NIR image overlayed with FIR $\endgroup$
    – bukwyrm
    Commented Oct 19, 2021 at 13:59

Yes. The size and shape of Neanderthal eyes implies they may have done this, and even regular humans see slightly into the infrared. We know of humans with four different photoreceptors (terachromats) and the animal kingdom has animals with everywhere from five to thirty five different receptors.

The chief problem is that there's always a trade-off. The number of photosensitive cells of a given size that you can pack onto a given area is bounded, and you need more complex processing which requires a larger visual cortex.

Neanderthals had, indeed, a substantially larger visual cortex, but reduced other parts of the brain to fit it in.

  • $\begingroup$ Neanderthals would not have done this unless they had a 4th receptor, as I stated in my question "No animal can sense infrared through the use of photons because their photoreceptors absorb only red, green, and blue. Studies such as this 'have not received a satisfactory physical explanation', as simply finding a way for the red photoreceptors of the eye so absorb a wavelength of 700-1000 will result in the colour subsequently being perceived as another shade of red.". $\endgroup$
    – Lutro
    Commented Mar 16, 2018 at 15:31
  • 1
    $\begingroup$ Strong claims about the Neanderthalers, can you give sources? $\endgroup$ Commented Mar 16, 2018 at 15:35
  • $\begingroup$ @Imipak I did come across the issue regarding the amount you can pack in, both regarding the receptors and the brain. Since this is a theoretical organism both eye size, number and brain size is dynamic. Assuming the cortex scaled proportionately to processing power needed while not inhibiting other brain functions, e.g. no trade-off. I see no faults with this, I asked the question to see if there is anything i've overlooked. Do you notice any issues with the scaling etc.? $\endgroup$
    – Lutro
    Commented Mar 16, 2018 at 15:36
  • 1
    $\begingroup$ Humans can have four colour receptors, as noted, although it's rare. I am using the following sources: rspb.royalsocietypublishing.org/content/280/1758/20130168 (please wait) $\endgroup$
    – Imipak
    Commented Mar 16, 2018 at 16:28
  • 2
    $\begingroup$ bbc.com/earth/story/20150805-neanderthals-strange-large-eyes has some useful insights. But it won't be until the Neanderthal Genome Project reports that we will know how many colour receptors Neanderthals had or what frequencies they worked at. There is currently no evidence on that score. $\endgroup$
    – Imipak
    Commented Mar 16, 2018 at 16:54

You merge several kinds of data now in your visual cortex. You have rods and cones. Rods respond just to luminance. Cones respond to different colours. You merge low res colour info with high res luminance info. You do the same thing in other senses too. Close your eyes. Touch your fingertips together behind your back. You integrated a whole bunch of positional data from the weight of your limbs.

Try this: Hold your hands about 2 inches apart. Have someone else put their hand between your hands, then you close your hands together. You have made a 'hand sandwich' It feels very odd because you are getting conflicting data. Kinaesthetic sense says your hands are separate. Touch says they are together.

The reason we don't see in NIR is that there isn't much to see. The earth's atmosphere is nearly opaque to it, so there isn't much illumination.

Animals such as snakes that detect warm targets are using very far infrared.

An object at 30 C (bit warmer than normal skin temp) has a peak wavelength at 10 microns. Compare to 700 nm for the red end of the visible. About 14 times the wavelength.

Making a detector has issues too. Snakes make it work because they are colder than what they are looking at. This is a general problem: it's hard to make a detector when all the components are glowing in the band it's trying to detect.

I'm less clear on why more animals don't see UV.


That last image with the firefighter is not near-infrared. It's from a thermal imaging camera which looks at longer wavelengths than NIR, ie, what's generally considered "heat".

The difference can be approximated by noting that NIR can be treated the same as visible light, just slightly below the range that the eye can detect. Most night-vision gear that's not light-amplification, such as that used by wildlife documentaries to film animals at night, is NIR. It uses a source of light that acts exactly like a normal spotlight, but at wavelengths the animals can't see, and the camera detects the reflected light.

A Thermal imaging camera, on the other hand, detects the radiation emitted by the object itself. A wildlife documentary wouldn't need a source of (invisible to the eye) light to illuminate the animals, you'd see the heat emitted by the animals, which would contrast with the temperature of the things around it.

To put it another way, if I looked at a normal interior wall with a NIR sensor, all I'd see is the wall looking not much different than what I'd see with normal light. If I looked at it with a decent TIC, I'd be able to see subtle heat variations: where the wall studs were touching the drywall, the slight waste heat generated by the electrical lines, where pipes were carrying cold or hot water, whether vents were heating or cooling the room. A NIR-capable camera cannot do that. But, and here's the trick, it has to be warmer than the sensor: if your camera's sensor is at 50 C, it wouldn't "see" anything 50 C or colder because the camera essentially blinds itself.

tl;dr: There's not particular reason why an organism can't be sensitive to NIR wavelengths. It wouldn't be different that how some organisms can sense ultraviolet. Seeing an image like that picture of a firefighter, though? Not happening.

  • $\begingroup$ An odd fact you have there, that infrared sensors would not be able to detect anything colder than themselves, and one that's rather hard to believe. Do you have a source for that claim? I have a hard time seeing thermal imaging being of much use outdoors (especially in warmer environments) if that is the case. Or for your interior example, for that matter: the camera would presumably be at room temperature and would be unable to identify the cold-water pipe separately from the rest of the wall. Did you mean to say instead that the object of interest must be warmer than its surroundings? $\endgroup$
    – Palarran
    Commented Mar 19, 2018 at 4:02
  • $\begingroup$ Using the example of the cold pipe, it's the relative temperature difference. The pipe will cool the wall adjacent to it slightly, and that difference will show up. The "cool" part is still warmer than the operating surface of the sensor, which cools itself by passive heat transfer. The microbolometers used in a TIC sensor are thermally isolated and held in vacuum, which makes it harder for them to get too hot to blind themselves, but it can happen. $\endgroup$ Commented Mar 19, 2018 at 5:02
  • $\begingroup$ Ah, I hadn't realized that the thermal sensors were insulated like that. Still, I'm pretty sure that animals with thermal sensors (snakes, for instance) can't manage vacuum-isolated sensors, and yet I doubt they blind themselves in that way, or else what would they be able to see when the sensors are body temperature? $\endgroup$
    – Palarran
    Commented Mar 19, 2018 at 14:56
  • $\begingroup$ Snakes are poikilothermic and use blood vessels to cool the heat-sensitive tissue in the pits as much as possible, which keeps them them colder than body temperature, which is already colder than the warm-blooded prey they're hunting. $\endgroup$ Commented Mar 21, 2018 at 4:10
  • $\begingroup$ I have an S60 smartphone with thermal camera, and i can assure you that it can discern well below 0degC, without cooling the device $\endgroup$
    – bukwyrm
    Commented Oct 19, 2021 at 14:02

Many animals see into the near infra red already. Some also into the near ultraviolet.

There is no hard dividing line between different parts of the EM spectrum, sensors can very well have an operational range that overlaps parts of different named areas, and that includes both artificial and biological sensors (of which eyes are one example, skin is another).


Of course. It's not about brain, it's about eye. They just need the right cells in their eyes to pick up both infrared and visible light.


Nature got there first

Some forms of fish such as goldfish, salmon, piranha and cichlid can see infrared light. Salmon and some other freshwater fish have an enzyme that switches their visual systems to activate infrared seeing, which helps them to navigate and hunt in murky waters.



You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .