Is sanity rational?

[Michael B]

therefore, the more accurate mental representation means better survivability.

Except maybe not. For example, we may have an exaggerated fear response when faced with uncertainty (e.g. in the dark). To perceive a greater probability of a wild beastie being present behind a tree than actually exists in reality may enhance survival. The statistician David Spiegelhalter has all sorts of interesting anecdotes on how poor we are at perceiving likelihood of events. And this is something Alvin Plantinga has thought and written about: that evolution need not necessarily drive accurate perception of the world around us; there may be indeed be evolutionary advantages to a distorted sense of reality. And it goes back to that ancient philosophical problem: how can I be sure that I'm not hallucinating or dreaming?