There are a few important concepts here which can help UI designers create better UX. It's always good to be able to back good design up with proven science. Eg I always tell people that animations are more than just eye candy, they attract the brains attention and elevate the element's importance in the UI. Now I can append "prevents change blindness" to that argument.
Sure it is, for the perfectly ordinary notion of "truth" that most people use effectively and successfully every day.
Does it conform to some philosopher's essentially alchemical (that is: imaginary, unachievable, uninteresting and useless) idea of "truth"? Of course not, but then: why would it?
As a card-carrying Bayesian I'm somewhat sympathetic to Jaynes' position, but the fact is we can and do distinguish between relatively subjective phenomena ("broccoli tastes bad to me") and reasonably (but of course not perfectly) objective ones ("quantum uncertainty is not an artifact of our ignorance"). Both are a result of of the objective relationship that holds between the knowing subject and the "really" unknowable world, with different poles of the relationship being more or less dominant in different cases.
Reality is imperfectly knowable, and of course only knowable from the perspective of a knowing subject. We can only say a very small number of things about unknowable reality (it is probably non-local, for example) but all of our knowledge is still knowledge, and holding any of it up to the standard of "truth as in how reality would be known without a knowing subject" is not only wrong but incoherent.
That fallacy says that there are instances where people believe that something they imagine to be true is actually true, or that if they don't understand how something works then it is beyond understanding.
I don't see how that means that everything we observe at all times is false.
Facial recognition is also an interesting one that we tend to amuse ourselves with: https://twitter.com/faceinthings