I'm designing creatures with naturally occurring white or pale grey irises and pupils so that it looks like their eyes are all white. What effect could this have their vision (colorblindness, light perception, etc)? Please note I'm not talking about leukocoria or eyeshine.
If the pupil is covered with material that is like Muslin - a "sieve" with holes small enough not to be apparent - it would appear white from outside, but pass enough of light to still create a visible image; Muslin allows enough light through from the "bright side" so that a woman wearing a veil over her face can still see, but her face is obscured - similarly to one-way mirror except diffusing the reflected light instead of keeping it structured. Seeing in darkness would be severely limited but it would be far more immune to glare.
That is not to say the being would need to be blind in darkness. The "veil" needs to be only opaque to human eyesight, that is, in visible spectrum. It could, for example, be completely transparent to ultraviolet and infrared, and related sensors in the eye could see with perfect sharpness - especially at night infrared would work quite well.
So, yes, it's possible at cost of reducing ability to see colors visible to humans - but not blocking it completely, just "making it worse than human".
A side effect to the "sieve" cornea might be that with the holes small enough, light would diffract. A very specialized eye (...inner structure entirely different from ours) plus a very complex neural "backend" could resolve the diffraction patterns into holographic image of the visible objects. That would not only result in superior depth perception, but also unlimit the sight range]("What do your elf eyes see" - what are the limits of sharp eyes?) making it possible to see objects at an arbitrary distance in smallest detail, only providing they are bright enough.
The pupil appears dark (not red-orange as the back of the eye which can be seen with careful inspection techniques) because light goes in but doesn't come out; it is the same as an absorbing surface. The eye, like the velvet lining of a telescope tube, doesn't want un-absorbed light bouncing around causing glare and fog.
Our cornea blocks UV, which is damaging to the innards. It probably doesn't absorb UV (which is damaging!) but reflects it. So our pupils would appear as a white lenscap to sensors that used only those frequencies we don't.
We don't have a name for color perception beyond the ones we see, but Yperiodēs or ypio for short (as primary colors are simple words) might be the perception of a distinct color in the 300-350nm range (see wikipedia and primary reference 31 Ödeen & Håstad "The phylogenetic distribution of ultraviolet sensitivity in birds"), and a secondary color where (in the UVS arrangements) where the shortest cone overlaps the short up to the blue primary. That would be different from our purple which causes a closing of the color wheel.
The color ypio is important for seeing mouse trails on the ground, spotting ripe fruit, and the patterns of plumage on the opposite sex. It's an eye-catching, important color; one that's worth considerable engineering difficulty to maintain as a sense.
Our protective lenscap reflects only ypio and absorbs everything else, so that's the color it appears to them. An intelligent Psittaciforme would notice that human children's eyes may be dark but definitely turn vivid ypio as we age. It's probably an indicator of sexual maturity (he supposes), like the iridescent ypio trim on his own beautiful ladies (but we see them as only blank gray, btw). My Tyndall blue iris is also a iridescence-based (not a pigment) color used by bird feathers, so the total effect involves colors used in body markings they are instinctively tuned to.
Any earthly species will use roughly the same visible spectrum, so they overlap a lot even if not identical. Earthlings will never see our pupils as white, but as a tint that reflects (pun intended) the difference in our visions, on the short-wave end.
IR-seeing animals would see all our body glow, not just the pupils. A richer FLIR sense might notice different emission coefficients on different materials, but I'll leave it to someone else to analyse that, or look at a FLIR instrument to see if there are useful distinctions.
White implies a high level of diffuse reflectance. Mundane pupils are dark because the light which passes into the eyeball through them gets absorbed, and no (or very little) light is passed back out through the pupils.
So a being with uniformly white eyeballs would be blind (at least in the visible spectrum). Unless, of course, this is a magical being, in which case all bets are off.
The pupil is black not because it's coloured, but because there's no light coming back out to colour the image you see.
A pupil any other colour than black implies light coming back out, which in turn implies that the eye is not working at full efficiency. Light coming out means that not all the light that has gone in has been processed, so there may be missing parts of the image the organism sees or the image may just be dim.
It is, however, possible. The human eye has many layers or retinal receptor cells (the rods and cones that let you see light and colour). If an eye has just one or two layers, not every available millimetre of surface will be covered, so not every bit of light will be absorbed. This would seem to imply that the image seen would be incomplete, but since most of it would be there, the brain can fabricate the rest.
You would probably find that this animal moves its head a lot more to make up for not having complete vision. This ensures it sees all that's going on, including its potential prey.