Surprise Me, Please

Of the dilemmas that bedevil me, the tension between the reliability of predictive behaviors in people and the perils of stereotyping is one that creates the most consistent cognitive dissonance. In other words, how do I best avoid stereotyping - and its uglier cousin, demonizing - when people frequently are so damn predictable? Although I'm confident not many people will admit they struggle with this as I do, I'm even more confident I'm not alone here on the bell curve.   

As for that vapid piece of conventional wisdom - "the exception proves the rule" - I don't know where to begin. What the hell does that even mean? Predictive behavior, ipso facto, is not an absolute; there are always exceptions. That's not the source of my cognitive dissonance. Let's use me as an example to help minimize defensiveness. Ready?

I'm a vegetarian. OK, respond to the following using that one piece of information: 
* Characterize my second amendment views. Be as inflammatory as you like, promise I won't object.
* If I have a religious affiliation, what is it?
* Predict one brand of car I'm likely to pick and one I'm unlikely to pick.

Since mass marketers have always made money using predictive behaviors - and now search engines are engineered to reinforce those same predictive tendencies - I know this genie is never going back where it belongs. But as someone who fights stereotyping, this is crazy making. If nothing else, I'd welcome surprising myself with my own behaviors more frequently than I do. At least then I'd be more receptive to being surprised by others.