Apparently, these illusions impact our observations of Saturn, and other planets!
No doubt. But if a feature can be shown to exist (optical depth), how is that an illusion? It's hard to separate illusion from reality, we have to be aware of them. I observe the Enke minimum every time I look at Saturn. It could be an illusion. However, if the central A ring is more transparent and provides enough contrast it very well could be an actual feature of the A ring. Since some anomaly (optical depth) can be show to exist, there should be no reason why we could not (and very often do) observe it.
Its important to keep the concept of detection, or seeing something, at least discriminated (since they are related) from the criteria of resolving or separating nearby structures of points or linear features. Cassini's division for example was discovered with a scope with lesser resolving power than what would be required to separate nearby lines of that width, which is approx 0.7". Isolated features may be seen well below the Rayleigh criterion, but not strictly resolved, as long as there is a sufficient contrast differential.
Glenn, yea, that's a great point to keep straight. I actually confuse them all the time, that being observed is being resolved. Not in terms of Raleigh, of course, but in differentiating one point from another in terms of contrast, as you say. Ganymede, and extended object, is a perfect example. It has essentially the diameter of a 150mm Airy disc, yet it is not a single Airy disc (as evidenced by it's washed out diffraction ring structure.) Yet, you can see brighter points on it's disc.
Another stunning example was a small craterlet 'g' on Plato's floor subtending just 0.7" arc (well below Dawes and the Airy disc diameter closer to lambda/D). During one night on several occasions in the most excellent seeing moments, there it was plain as day in full crater form. Was it resolved? Is Lambda/D closer to the real limit of resolution? One image of Plato also showed small craters in crater form (bright limb and dark floor) very near Lambda/D (for that aperture.)
If the Enke minimum reflects less light, it's conceivable it will not be as bright as the less transparent parts of the A ring and provide some level of contrast to the observer. In this case, it can be (and is often) seen as a separate feature, because it is a feature of the A ring. The dimming of the inner B ring might be similar in this respect, which we observe often enough, as well. No Mach band there, but there should be if the illusion is at work.