Is there much difference between vision in daylight and in the dark?

Category: Biology      Published: June 16, 2026

By: Christopher S. Baird, author of The Top 50 Science Questions with Surprising Answers and Associate Professor of Physics at West Texas A&M University

Yes, there is a huge difference between how the human visual system functions in normal-light situations and in low-light situations. In fact, the human visual system employs two different types of photoreceptor cells for these two different situations—cone cells and rod cells—which function quite differently from each other.

Humans can see because light passes through the lens of each eye and, as a result, forms images of the outside world on the retina that sits on the inner back surface of each eye. An array of photoreceptor cells on the retina then converts the image into electrical impulses that are sent to the brain to be experienced as sight. The cone-shaped photoreceptor cells are primarily used in normal-light situations, such as being outdoors during the day, or being indoors during the day or night with the room's lights on. In contrast, the rod-shaped photoreceptor cells are primarily used in low-light situations, such as being outdoors during the night (far away from streetlights, car lights, campfires, and flashlight beams), or being indoors during the night when the room's lights are off, no candles are lit, no flashlights are on, the doors are shut, and the window shades are closed, or being indoors day or night in a room with no windows when the room's lights are off, the doors are shut, no candles are lit, and no flashlights are on. The ability to see in low-light conditions using the rod cells is called night vision or scotopic vision.

If the human retina only contained cone cells, we would be unable to see anything in low-light conditions. In low-light situations, the cone cells are mostly useless because they are not sensitive enough, so the rod cells must enable vision. On the other hand, if the human retina only contained rod cells, we would be unable to see anything in normal-light conditions. In normal-light situations, the rod cells are mostly useless because they are too sensitve and become overexposed. In mid-level lightning conditions, the rod cells and the cone cells both work together to enable vision.

The rod cell system uses several clever techniques to make it possible for humans to see something even if very small amounts of light are present, as summarized in the table below and explained after the table. However, these clever techniques come with a price.

No Color Differentiation

The cone cell visual system is able to see and differentiate colors, which is made possible by employing three different types of cone cells: L cells that detect the long-wavelength colors of the visible spectrum and therefore most strongly detect red colors, M cells that detect mid-wavelength colors of the visible spectrum and therefore most strongly detect green colors, and S cells that detect short-wavelength colors of the visible spectrum and therefore most strongly detect blue colors. The cone cell system is able to see millions of colors by comparing the relative strength of the signals coming from the L, M, and S cone cells that are at the same spot in the retina. For instance, there is no yellow-dominant cone cell, but yellow light triggers L (red-dominant) cone cells and M (green-dominant) cone cells at the same time with about equal strength, which the brain is able to interpret and experience as yellow.

In contrast, the rod cell visual system is not able to differentiate colors. There is just one type of rod-shaped photoreceptor cell in humans and it detects all colors in the visible spectrum (except for deep red colors). That is why, in true low-light situations, everything looks colorless, appearing only black and different shades of gray. The rod cell system's failure to differentiate colors may sound like a serious drawback, but it's actually a good thing! It helps make the rod cell system more sensitive to very low amounts of light. By accepting and detecting almost all colors of visible light inside a single rod cell, each rod cell can use a larger fraction of the available light.

Illustration of color vision (left) versus dark grayscale vision (right). Click the image to enlarge it. Public Domain Image, source: Christopher S. Baird.

For example, if a blue bit of light hits an L (red-dominant) cone cell, that blue bit of light will mostly not stimulate the cell and will be wasted. It will mostly be absorbed as heat without contributing to vision. This works fine for normal-light situations because there is so much light present that some can be wasted. However, in low-light conditions, wasting light would limit how sensitive your vision is. For this reason, each rod cell is designed to waste very little of the incoming light by detecting every color (except deep reds). The result is the ability to see at lower light levels than otherwise, but at the price of not being able to see color in low-light situations.

More Rod Cells

The total number of rod cells in a healthy human retina is far higher than that of cone cells. The human retina contains about 120 million rod cells but only about 7 million cone cells. The presence of more rod cells in the retina means that more light overall can be detected and used for vision for each integration time period. However, with so many rod cells to pack into the retina, distributing them uniformly across the retina would waste precious real estate in the fovea, which sees sharp central vision, where the cone cells need to be situated to enable high-resolution vision in regular-lighting conditions. To avoid this waste, the photoreceptor cells in the fovea are almost all cone cells and not rod cells. Therefore, the ultimate price to be paid for the boost in sensitivity that is gained by having many rod cells, without hindering sharp central vision in regular-light conditions, is that central vision is poor when using the rod cell system. But it's worth the price for this boost in sensitivity.

Illustration of good central vision (left) versus poor central vision (right). Click the image to enlarge it. Public Domain Image, source: Christopher S. Baird.

Note that the brain cleverly extrapolates the information that it receives from the rest of the retina so that your central vision does not usually look completely blank when seeing using rod cells in low-light conditions. Also, your eye jitters around and your brain overlays and combines several captured images to fill in the poor central vision. However, information gained from extrapolation and overlaying is not as accurate as directly detected information, which is why your central vision is poor in low-light conditions. You can even directly notice this problem. In low-light conditions, stare directly at a small island on a large map, for a minute, without moving your eyes at all. The island on the map will appear to become blurry or even seem to disappear.

Higher Synaptic Convergence, i.e. Larger Integration Areas

In the human retina, many rod cells connect to the same output signal. Up to thirty rod cells can connect to a single downstream pathway. As a result, the light-triggered activations of several different rod cells in the same spot in the retina are added together to form one, stronger signal. In other words, the rod cell system has high synaptic convergence. With more light being collected for a single output signal sent to the brain, lower light levels can be detected, making the rod cell system more sensitive to low light levels. However, the price that has to be paid for this boost in sensitivity is lower visual acuity, i.e. reduced spatial resolution.

Illustration of low synaptic convergence and therefore high spatial resolution (left) versus high synaptic convergence and therefore low spatial resolution (right). Click the image to enlarge it. Public Domain Image, source: Christopher S. Baird.

In simple terms, you can't see details as well when seeing using the rod cells than when using cone cells. This is because having a clump of rods cells wired to the same output signal makes that clump of rod cells act effectively as a single, large rod cell, in terms of visual acuity. In other words, the area on the retina that has its light detection events end up as a single output signal is larger for the rod cell system than for the cone cell system. For the cone cell system, typically a single cone cell is wired to a single output signal (i.e. a single retinal ganglion cell), whereas up to thirty rod cells can be wired to the same output signal. That's why it's difficult or impossible to read a book in the dark of night with all of the room's lights turned off and no candles lit and no flashlights on, despite the fact that you can indeed see the overall shape of the book (assuming that your eyes have had time to visually adapt to the darkness).

Use of Rhodopsin

Rod cells use the highly-sensitive chemical rhodopsin as the light-absorbing molecule that activates the cell, which is more sensitive than the light-sensing molecules used in the cone cells. Additionally, each rod cell is designed to make rhodopsin's response especially powerful. A single activated rhodopsin molecule can activate hundreds of transducin molecules, which is the next step in the chain of events that delivers the signal to the brain. As a result, each rod cell is about 100 to 1000 times more sensitive, once fully adapted, than a single cone cell.

Longer Integration Times

The rod cell system is also designed so that each rod cell absorbs light over a longer time period for each image acquisition. In other words, each rod cell employs a longer integration time period. The cone cell system in normal-light situations is able to get away with using a short integration time because there is so much light available, which means that more images can be acquired per second. Seeing more images per second, i.e. seeing at a higher temporal resolution, means that the cone cell system can see motion more accurately.

In contrast, rod cells detect light for a longer amount of time per single electrical output signal than cone cells, which typically means that more light overall can be received by each cone cell and summed together to become a stronger single output signal, thereby enabling vision in lower light levels than otherwise. The price paid for this boost in sensitivity is that the rod cell system sees fewer images per second than the cone cell system. This means that the rod cell system cannot see motion as accurately. As a result, when using the rod cell visual system in low light conditions, moving objects will appear blurrier and with a less clear sense of motion. Note that as the observer walks, rides a bike, or moves in other ways, his viewpoint is changing so that stationary objects visually appear to be moving or changing shape. This means that even stationary objects will appear blurrier when seeing using the rod cell system, if the observer is moving around.

Interestingly, the human visual system uses motion to help it see depth. This includes the depth perception cues of motion parallax, the kinetic depth effect, and optical expansion. Ultimately, this means that the rod cell visual system cannot see depth as well as the cone cell visual system. As a result, it's harder to discern the three-dimensional shapes and locations of objects when seeing in low-light conditions using the rod cell system. However, this is a price worth paying for the associated boost in sensitivity to very low light levels.

Summary

In summary, because of the way the rod cell visual system is designed, your vision in low-light situations is colorless, bad at perceiving motion, bad at perceiving depth, bad at seeing details, and bad at seeing in the central region of vision. But all of this is still better than seeing nothing at all!

The animation below of a rotating Dr. Baird bobblehead illustrates regular-light vision (the animation on the left) versus low-light vision (the animation on the right), when all the drawbacks are included. If you click on the animation below to enlarge it, these drawbacks will be more pronounced.

Simulation of regular-light vision (left) versus low-light vision (right) when all of the drawbacks are included. Low-light vision (right) sees only in grayscale, sees with lower spatial resolution—making objects appear blurrier, sees with lower temporal resolution—making moving objects appear even blurrier in the direction of motion, and sees with poor central vision. Click the animation to enlarge it.
Public Domain Image, source: Christopher S. Baird.

Topics: color, cones, dark, dark adaptation, eye, eyes, light, night vision, rods, vision