Good points all.

Step one: add logic and critical thinking as required curriculum from elementary school on.

I enthusiastic concur, and within/accompanying that subject, teaching that combats "innumeracy". Although I wouldn't want to underestimate the power of believing what one wants to believe, perhaps if there were less innumeracy there would be somewhat fewer folks on the right, for example, nonsensically insisiting that rising revenues in years following tax cuts "proves" causation when in reality it doesn't even establish a positive correlation*.

I also agree that media other than the Internet play a very large role in the trend -- I'd say in terms of numbers probably even a larger role than the Internet, considering the size of the audiences for ideologically hyperpartisan talk radio as well as the opinion TV shows on MSNBC and Fox News -- and certainly the proliferation and ideological fragmentation of those media that you mention has been the broader, sweeping force driving the trend over the past decade or two.

I do think, though, that there is something uniquely powerful in that echo chamber, interactive participation that you note; my sense is that active participation tends to become almost akin to a pledge of allegiance to one ideological "side" as well a public staking of one's positions on particular issues, all of which I think have a uniquely reinforcing effect, creating a stronger association with one's identity and sense of worth (supposed righteousness,  supposed insightfulness) and making it harder to climb down from a position or modify one's general ideological orientation in light of new information. Additionally, there is the social, community aspect of those echo chamber blogs, which tends to reinforce a feeling of "our (good and wise) side" vs. "their (ignorant, stupid, and/or evil) side" as well as create a sort of peer pressure. Just as one small example of a manifestation of that dynamic, often when I dispute the facts or reasoning of some echo chamber regular who is presenting some argument supporting a policy that his "side" is seeking, his facts or reasoning is so obviously invalid that I have to think that some other participants know it, yet rarely does anyone point out his error, because it's all about who is on their side and who isn't (and by the way, they assume that anyone who challenges some of their "facts" or reasoning [without even expressing a policy preference] must be a hyperpartisan on the "other side", which is why the Redstate folks branded me a "lefty troll" and baselessly attributed to me very liberal policy preferences, and the regulars on echo chamber of the left call me a right winger and baselessly attributed to me very conservative/libertarian policy preferences).

I'll admit that I tend to watch and read those perspectives that are at least similar to my own

There is nothing bad about (and some good from) skewing such consumption in that way to some degree, as long as it isn't skewed excessively. It may be optimal for one to spend somewhat more time with such sources in order to go deeper into some issue, policy, etc. rather than spending fully equal time  with opposing voices who disagree with the more basic components. This may not be the best example, but if someone wants to learn more about (and perhaps discuss/debate) the optimal strategy for the war in Afghanistan, he may reasonably end up spending more time with sources that believe we should continue fighting that war than with those that advocate withdrawing ASAP (which isn't to say he should cut out the latter completely). Or if someone wants to learn just how bad the climate change problem is, the causes and potential solutions, he may spend more time with sources that believe climate change is occurring than with those that contend it is not, or is not affect by human activity, or is happening but will bring no significant net harm worth worrying about.

And to some extent we all do this in an obvious way, as is rational, most notably with regard to what we regard as extreme sources with what we consider to be whacky views -- I'm not going to spend much time on 9/11 truther blogs searching for a better understanding of the events and issues surrounding 9/11. Ditto with some neo-Nazi website/blog regarding societal problems in America. Etc.

* The fact is, GDP is more often growing than declining, and as a result revenues tend to grow over some period of several years whether tax rates are raised, lowered, or left alone, so increasing revenues in years following a tax cut doesn't even provide a hint that revenues are higher than they would have been if tax rates had not been cut. The problem with their "analysis" (observing rising revenue after the Kennedy, Reagan and Bush tax cuts -- even leaving aside the fact that Reagan subsequently raised taxes) is thus that they ignore what has happened to revenues in years before which there was NOT a tax cut. In other words, they fail correlation analysis 101. One analogy of mine: I could observe that every time I live for 5 years in New York City, I look older at the end of that period than I did at the beginning. Applying their "logic" I'd conclude that New York City very strongly, positively correlated with my looking older, yet for all they know the same thing happens to me (perhaps even to a greater degree) when I'm NOT living in New York City. [edit: Oh, and when I point them to my compilation of economists, including many of the most prominent conservative economists, stating disagreement with their "tax cuts always/generally increase revenues" mantra (swordscrossed.org/node/1671 ), they respond with "That's just an appeal to authority. I can see for myself that revenues went up. Isn't it obvious??"]

I think you'd find very interesting a book I often recommend, www.amazon.com/True-Enough-Learning-Post-Fact-Society/dp/0470050101 . It's all about this stuff.