I've just finished reading McCauley's Why religion is natural and science is not (2011, OUP), a book I already bought at the eastern APA last year, but only now gotten round to reading. See this video for a 2-minute summary of the book (book trailer). McCauley makes an interesting case for the "naturalness" of religion and the "unnaturalness" of science. He uses "natural" (or, more specifically, "maturationally natural'') in a fairly restrictive sense, namely as those properties of human thought that arise early in development, without explicit instruction. Examples include speaking a natural language, walking, and chewing.
McCauley thinks science is not natural. Science provides us with concepts and world views that are contrary to our everyday knowledge and experience. For instance, the heliocentric model of the Earth orbiting Sun contradicts our visual observation of the Sun moving across the sky; particle physics is at odds with our intuitive understanding of solid objects. Scientific modes of explanation and ways of thinking are also contrary to our maturationally natural capacities. For example, the empirical sciences have, over time, appealed less and less to agents as causes, and more to impersonal causes to explain events. A recent paper by Shtulman and Valcarcel, in press in Cognition supports the unnaturalness of science thesis.
As the researchers point out, these findings are consistent with earlier studies that indicate that people exhibit more animistic thought and a tendency towards teleological thinking as they get cognitive impairments (such as Alzheimer's) or are put under increased cognitive load. There are a few studies with people who have more practice in science (such as people with a PhD), indicating these effects are somewhat attenuated, but nevertheless, they do not disappear completely in scientists either (e.g., an increased tendency to teleological judgments under time pressure such as saying that "The Sun shines to nurture life on Earth" is true). It seems -- regrettably -- that scientific reasoning doesn't become second nature to us.