# How acute can human vision be enhanced to be while retaining the size and shape of the eyeball?

How much more acute could human eyes be biologically engineered to be while keeping the same shape? Animals like eagles have very large eyeballs which due to their size and shape cannot be moved, so they have to turn their whole head to look at something; I want the eye to still be able to move.

A starting point might be to add a pecten like birds have so fewer blood vessels obstruct the retina.

• Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking.
– Community Bot
Nov 21 at 19:03
• Why would you want to retain the size of the eyeball? if manga has shown us anything it's that oversized eyes (by extant real world standards), when done right, are cute 🤗 there's whole studies on attractiveness that have been done and conclude basically the same thing. Nov 21 at 19:16
• On what empirical measure of acuteness will you be evaluating responses? Nov 21 at 19:25
• Do you mean a really high resolution in the central part whilst low-res in the periphery much as birds of prey? The overall information the brain gets is same order of magnitude in humans, just organised differently. Can you clarify? Nov 21 at 19:29
• Eagle's eye-balls are actually of similar dimension as human eye-balls (at least for large eagles), but have far better visual accuity in bright light. This is because they have a much greater density of cones in their retina (relative to the rods). You could presumably engineer a human eye to have similar properties. The down-side would be that this would result in poorer night vision. Nov 21 at 22:19

The human pupil has a maximum diameter of about 8 mm.

Under ideal conditions, the Rayleigh criterion gives the maximum angular resolution $$\theta$$ of an ideal lens with an aperture $$d$$ as $$\theta = 1.22 \lambda / d$$ where $$\lambda$$ is the wavelength of the light.

Putting $$\lambda =$$ 555 nm (at the peak of the luminosity function) and $$d =$$ 8 mm we get $$\theta = 1.22 \times \frac{555 \times 10^{-9}}{8 \times 10^{-3}} \approx 0.00008\,\text{radians} \approx 0.3' = 18''$$ which is only about 3 times better than the angular resolution of a common or garden-variety human eye and a little more than 2 times better than the eyes of a person with exceptionally good vision.

Which is about as expected. The human eye comes quite close to the best possible resolving power for the size of the lens; moreover, this extraordinary feat of natural engineering was accomplished while still allowing the human eye to function over an enormous range of illuminations, from the 120,000 lux of a very bright summer noon to the paltry 0.002 lux of starlight of a clear moonless night. In exceptionally good conditions, a dark adapted human eye can even count individual photons! (Not really, more like one out of three, but still very impressive sensitivity.)

All in all, size matters. The size of the photographic sensor (or retina in a living photocamera) determines the size of the lens which determines the maximum resolving power and thus the maximum level of detail which can be recorded. A half-decent fifteen years old DSLR camera with an APS-C or full-frame sensor will outperform the camera of the most up-to-date very expensive smartphone by a wide margin, simply because its sensor is so much larger, and optics is merciless.

This is why astronomers used to lust over telescopes with larger and larger mirrors... Unfortunately, Earth's atmosphere is a non-isotropic, fluctuating, optically dispersive medium, which limits the angular resolution of an Earth-based telescope operating in the visible spectrum to about 0.4 arc-seconds. Nowadays, there are computational tools to combat the effects of astronomical seeing, but even with such wizardy the only real way forward in optical astronomy is to put the telescope outside the atmosphere.

• "No, biology hasn't already arrived at near-optimal balances of considerations! There's got to be some trick to do much better!" /s Nov 21 at 22:29
• The Rayleigh criterion assumes you are looking through the atmosphere all the way into outer space. Down here on Earth, while looking at shorter distances through much less atmosphere, a camera can get away with being MUCH smaller. This is how cellphones can get 12mp resolutions with a lens not much bigger than a pen head and how optical microscopes can give you a clear resolution on the micron scale. Nov 21 at 23:07
• @Nosajimiki: The lens of a smarphone camera is very much bigger than a pin head -- about 5 mm diameter on typical smartphones, 8 mm on more expensive smartphones, even larger on smartphones which are built with special attention to their use for photography. Effective resolutions top at about 6 to 8 megapixels on typical smartphones, in very good light. (And angular resolution has nothing to do with the distance between the camera and the object.) (I don't see why you bring in microscopes. They do get to see objects 1 µm across; you can compute how close the lens needs to be to the object.) Nov 22 at 0:06
• Sorry, it is the aperture that is so small, the lens behind it is technically bigger, but the aperture is the limiting feature. That said, I may have been mistaking Rayleigh scattering with the Rayleigh criterion. Nov 22 at 14:21
• @Nosajimiki: The physical size of the actual mechanical aperture a photographic lens is not relevant. (Because a photographic lens is almost always a compound lens, and in fixed-lens cameras the physical aperture is usually somewhere in between the elements, to minimize distorsion.) (The only single-element photo lens I have ever seen was on a very very cheap medium-format all-plastic Seagull sold as a toy.) The relevant diameter is the diameter of the exit pupil. For example, the main photo lens on my Note20 has a (real) focal length of 5.4 mm and a maximum aperture of f /1.8, or 3 mm. Nov 22 at 14:37

## Improve the neurology, not the geometry

AlexP already does a good job of discussing the physical limitations of optics. His conclusion that the human eye is about 1/3rd optimal seems correct, and I do not believe an eye made of living tissue and focused using muscles is really gonna get much better than this.

However, human vision is more than just a camera. What you see is actually determined by the neurological systems in your brain and retinas. Your brain is already responsible for flipping images, interpreting patterns, filtering out blind spots, and generally doing all sort of "post production" work on images. Heck, there are structures in your retinas that use a neuro-feedback mechanism to add the color yellow simulating 4-color vision even though we only have 3 kinds of cones. If our retinas and brains can do all of this, then they could also do deconvolution.

Deconvolution is the use of post-production algorithms to filter out the effects of "known distortions". Since the Rayleigh criterion is the result of a knowable distortion pattern, that means that a clever neurological system can filter it out to achieve better than optical acuity.

And that is exactly what the researchers at Princeton University and the University of Washington have developed. Neural nano-optical(NNO) systems are super small cameras that use built in feedback systems to filter out refraction patterns and give images as clear as cameras about 500,000 times thier volume.

If the human retina were modified with a similar feedback system, then human visual acuity could be increased exponentially assuming the brain could process that much information.