Approximately how much "bandwidth", in bits per second, can typical human visual perception process?

Consider "The Matrix", where we assume a near-perfect digital encoding that can reproduce any experience nearly perfectly, compressed with a digital intelligence beyond our own. Now they need to create a cable that plugs into our brainstem. How much bandwidth would that cable need to carry, in order to simulate full human visual experiences?

For auditory processing -- given all the effort in digital audio compression, it seems likely the answer is somewhere close to 256 kbps (256,000 bits per second), at least within an order of magnitude, give or take. But vision seems far more complex, and I can't find any study or near-approximation that seems plausible.

Presumably the answer is somewhere from 1 mbps (1,000,000 bits per second) to 1,000 times that amount. A naive approach would be what is the theoretical bandwidth to create any arbitrary video, to the fidelity where human perception could not distinguish differences. However, this is so grossly far from accurate that I suspect it is almost meaningless. I am relatively aware that our visual and auditory processing compress and transform the inputs we perceive, leading to optical illusions etc, which would suggest a much lower number than we might otherwise expect.

  • 3
    $\begingroup$ This seems relevant: The computer model of the brain $\endgroup$
    – Steven Jeuris
    Commented Mar 7, 2021 at 16:43
  • 2
    $\begingroup$ Amusingly, this indicates that most every other question of comparison between the brain and computers has already been asked except that of bandwidth. $\endgroup$
    – Abacus
    Commented Apr 8, 2021 at 19:24

2 Answers 2


From these recordings, the researchers calculated that a guinea pig retina transfers data at about 875 kilobits per second. Human retinas have about ten times as many ganglion cells, giving a “bandwidth” of 8.75 megabits per second.

Read more: https://www.newscientist.com/article/dn9633-calculating-the-speed-of-sight/#ixzz7Ig3MCMrR

Source Paper: (Open access)
Koch, K., McLean, J., Segev, R., Freed, M. A., Berry II, M. J., Balasubramanian, V., & Sterling, P. (2006). How much the eye tells the brain. Current Biology, 16(14), 1428-1434. https://doi.org/10.1016/j.cub.2006.05.056


Lets assume you cut the optic nerve and replace it by a computer. There are between 700,000 and 1,700,000 nerve fibers in the optic nerve (lets round it up to 1,000,000 for now). What the temporal resolution of your input should be is less clear. A typical sampling rate in physiology is 1ms so lets choose 1kHz. Then assuming that all you need to know if whether the fiber is carrying a spike or not (1 bit), that gives you a back-of-the-envelope estimate of 1gbps. Surprisingly small, a USB3 port could carry your signal. Furthermore you could theoretically compress that signal by quite a lot without noticeable difference considering the huge redundancies in the signal. So I would say a maximum of 2gbps and probably a lot less than that.



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.