Google 'Peltzman effect' for more information.
Hmm, I don't think the Peltzman effect covers it... rather than being an observed phenomena, it's all about what 'may' happen in a hypothetical scenario. Applying the Peltzman effect is also reliant on the assumption that people wearing cycle helmets believe that the helmet offers unlimited protection against any kind of collision.
And I don't believe that you can fully apply the same idea of risk compensation to people ON a vehicle as to those IN a vehicle. Motorists are inside a 'protective' environment, cyclists are completely exposed so a safety device for a car isn't going to have the same effect as one on a cycle.
Of course there are people who are going to act against what is logical or the norm, but do you make a generalisation about cycle helmet wearers based on stupid/illogical behaviour?
As an example, take 2 groups of cyclists you get in London, the cycle couriers and the bike hire cyclists. Typically, the majority of people in these 2 groups don't tend to wear helmets: the couriers because they seemingly don't want to and the hire cyclists because (unless they are in the minority of users that commute on them) they don't have access to them. Couriers (in my observation) are the most reckless group of cyclists around, the idea seems to not be 'if' they get hit, but 'when'. Most of the hire cyclists in Central London are tourists unfamiliar with the UK highway code so their cycling is inherently more risky.
As I say, I think it's an individual's choice whether they want to wear a helmet, the only person the decision affects is themselves but I don't buy into this psychobabble second-guessing about how people act or react to helmets. I personally wear one because I don't want to get hit, wearing the helmet is one element to that, riding safely is another. The idea of putting on a helmet and going 'woohoo, eff you society, I'm invicible!' is completely alien to me.