Two Levels of Improbability

On the Freakonomics blog, Steve Levitt is asking for possible ideas for terrorist attacks we should be prepared for. He also points out that an attack will probably not affect a specific individual can also create terror:

Hearing about these rules got me thinking about what I would do to maximize terror if I were a terrorist with limited resources. I'd start by thinking about what really inspires fear. One thing that scares people is the thought that they could be a victim of an attack. With that in mind, I'd want to do something that everybody thinks might be directed at them, even if the individual probability of harm is very low. Humans tend to overestimate small probabilities, so the fear generated by an act of terrorism is greatly disproportionate to the actual risk.

They can compound the probabilities by picking an attack that causes a 0.1% probability of affecting 0.1% of the American public. A “dirty-bomb” attack may be ideal. One advantage of a “dirty-bomb” attack is that it might provoke regulations that hamper the replacement of imported oil with nuclear energy.

This also shows that letting innumerate left-wing Luddites set the public agenda could make a terrorist attack more effective and therefore more likely. Those ex-hippies aren't as harmless as they look.