The Elephant in the Brain: Hidden Motives in Everyday Life
The best synthesis of the study of human nature (cognitive psychology, interactionist sociology, primatology, and economics) I've ever seen. Freud done right ("although the explanations in this book may seem Freudian at times, we follow mainstream cognitive psychology in rejecting most of Freud's methods and many of his conclusions"). It's introductory, laced with illustrative anecdotes but with much deeper scholarship underneath.
The 'elephant in the brain' is our unwitting selfishness. We compete without knowing or admitting it, for we are social animals seeking power or status, and thereby sex.
Modeling the world accurately isn't the be-all and end-all of the human brain. Brains evolved to help our bodies, and ultimately our genes, get along and get ahead in the world—a world that includes not just rocks and squirrels and hurricanes, but also other human beings. And if we spend a significant fraction of our lives interacting with others (which we do), trying to convince them of certain things (which we do), why shouldn't our brains adopt socially useful beliefs as first-class citizens, alongside world-modeling beliefs? Wear a mask long enough and it becomes your face. Play a role long enough and it becomes who you are. Spend enough time pretending something is true and you might as well believe it.
Incidentally, this is why politicians make a great case study for self-deception. The social pressure on their beliefs is enormous. Psychologically, then, politicians don't so much 'lie' as regurgitate their own self-deceptions. Both are ways of misleading others, but self-deceptions are a lot harder to catch and prosecute.
Simler undertook the book in lieu of a PhD, and his work is a welcome modification of Hanson's usual relentlessly lucid style: he is more concrete, chattier, more personable.
Information is sensitive in part because it can threaten our self-image and therefore our social image. So the rest of the brain conspires—whispers—to keep such information from becoming too prominent, especially in consciousness. In this sense, the Freuds were right: the conscious ego needs to be protected. But not because we are fragile, but rather to keep damaging information from leaking out of our brain and into the minds of our associates.
You can probably skip this if you're familiar with Overcoming Bias / LessWrong / Econlog - but even then it's a pleasant read. I'm going to give this to every teenager I know. Armour and key.