Context can affect bias, and on the Web -- if I can riff on Lessig -- code is context. So why not design media that accounts for the user's biases and helps him overcome them?
I have a good friend who writes for a very conservative political website, and whose views on politics I respect. Though I strongly disagree with him, he frequently challenges me to think harder about my own politics, so I try to read his writing as much as I can.
But there's a problem. When I visit the site to read his posts I'm hit with pop-up ads trying to sell me red-meat conservative titles like The Roots of Obama's Rage (Dinesh D'Souza) or Demonic: How the Liberal Mob is Endangering America (Ann Coulter). As a liberal, these ridiculous titles are enough to get under my skin, and so by the time I've bypassed the pop-up and made it to my friend's post, I'm all riled up and that much less likely to give his arguments a fair shake.
While I hate to admit that something as silly as a pop-up can alter my tolerance for opposing arguments, this case is instructive. It points to shortcomings in how we reason, and suggests how we might design media to help us reason better.
First off, this example is a reminder of just how hard it is to overcome our own political biases. As hard as we try, we are a far cry from our ideal of rationality, forming beliefs based on facts and figures. Instead, our rational faculties are frequently put to use finding reasons to confirm our existing prejudices. When we confront new evidence, our emotions play a central role in how we react. This tendency, known as "motivated reasoning," has been well documented by psychologists, and was recently explained in an excellent and accessible essay in Mother Jones by writer Chris Mooney. As Mooney aptly describes it:
The theory of motivated reasoning builds on a key insight of modern neuroscience  (PDF): Reasoning is actually suffused with emotion (or what researchers often call "affect"). Not only are the two inseparable, but our positive or negative feelings about people, things, and ideas arise much more rapidly than our conscious thoughts, in a matter of milliseconds -- fast enough to detect with an EEG device, but long before we're aware of it.
But my pop-up ad example illustrates another critical point: our susceptibility to bias is not static. We may be systematically biased, but we are also fickle. Our political behavior is surprisingly susceptible to context and framing. To some extent this should seem intuitive; our openness to new ideas can depend on our mood, on who is presenting the idea, etc. It has also been documented by numerous academic studies.