Passenger profiling, game theory and terror prevention Commenting on <a href="http://www.econ.upenn.edu/%7Epersico/research/Papers/airportaea11.
Commenting on Passenger Profiling, Imperfect Screening, and Airport Security. Via Bruce Schneier.
The paper by Persico and Todd describes a game where
- v is the value of committing the crime
- l is the loss when detected and stopped
- zero is the value of not committing the crime
That seems a reasonable model of me taking an extra bottle of single malt across the border. Or of a professional cocaine courier.
But I think the game theory math breaks down when
- v is marthyrdom (and 72 raisins)
- l is hero status and a long life in prison
- survival and your normal life is the value of not committing the crime
In this case, I don't think the difference between v and l will deter anyone. The detection rate may still shift attacks to targets with lower detection rates, though.
The game difference between terror and other crime: Detecting one criminal will deter another one. Detecting a terrorist will (we hope) stop that one terrorist, but not the other ones.
In the 70's, hi-jacking was something of a gentleman's sport. The terrorists threatened with murder, but knew they could survive if they kept their murdering at a minimum. Deep down, the terrorist's focus was on his own survival, not on the 72 virgins. And often both terrorists and hostages came out alive.
It would be very useful if we could move a terrorist's focus away from Paradise, and back to his own survival. But I have no idea of how to achieve that.
tags: passenger profiling game theory airport security terror prevention terror deterrence marthyrdom Nicola Persico Petra E. Todd game theory CAPPS Computer Assisted Passenger Pre-Screening System 72 virgins