Border security: Avatars seek out suspicious behaviour
New research aims to create an airport security system that spots potential trouble better than humans can.
Border security and passenger profiling is a hot topic on both sides of the Atlantic. The 2009 Christmas Day mid-air terrorist attack may have been foiled. But it raised serious questions about why the alleged culprit was even allowed on board.
Could better profiling be the key to grounding potentially dangerous travellers? That's the question a new research initiative is trying to answer.
Humans are notoriously poor at spotting lies and other signs of trouble. But researchers hope that an avatar-based screening system will be more successful at flagging suspicious behaviour that should be looked into more closely.
Kiosks that screen for deception
Two institutions - the National Center for Border Security and Immigration at The University of Arizona (UA/BORDERS) and Laurea University of Applied Sciences/Safety and Security Management, Finland (Laurea) - are working together on this initiative.
They're exploring how an automated screening system can spot suspicious behaviour more successfully than border security or airport security agents.
The idea is to produce a kiosk that can screen for deception and intent in the field. The kiosk will use a battery of techniques to spot potential suspects. Controlling the process: an intelligent computer agent that interviews and monitors travellers' responses.
Take security at airports: A screen-based avatar might ask people questions as they pick up their tickets. That would enable the system to screen a vast number of people in a short time without laying a hand on them.
The avatar could give a signal indicating each person's trustworthiness. That would alert security personnel to look more closely at certain travellers.
Getting past the poker face
The most reliable and informative markers of deception are how people say things, rather than on what they say. But people are notoriously bad at reading the signs.
"We're trying to bring some science to what people do in a poker game," says Jay Nunamaker, from UA/BORDERS.
"Everybody thinks they can analyse what people are doing, that they know who's bluffing. But the literature from thousands of studies shows that - when faced with a single scenario - people are only 54 percent accurate in (spotting) deception."
In contrast, Nunamaker says his equipment is right 80 percent of the time.
People are only 54 percent accurate in spotting deception.
Nunamaker's equipment is right 80 percent of the time.
No matter how good a liar you are, no one has the perfect poker face. Everyone projects a catalogue of telltale signs, or cues.
"You've got hundreds and hundreds of cues," Nunamaker says. "Linguistic, vocality, psychometric cues - heart rate, the face heating up. But while humans can only focus on one or two cues at a time, border security agents are now trying to track hundreds at a time."
Rights versus responsibilities
In addition to airport security and border security, the technology has a host of other potential uses, such as:
- Gun licensing
- Insurance screening
- Visa applications
- Immigration procedures
The researchers acknowledge that their approach could raise issues about civil liberties and privacy. They counter that it could also reduce the need for other potentially sensitive screening techniques, such as full body scanning or racial profiling.
"Risk is not related to nationality, but to personality," says Nunamaker. "It is important to concentrate on the person."
The future of airport security?
Future implementations and regulation will determine which possible scenarios mature into everyday use.
International cooperation plays a vital role in the current project. Field studies in the United States and Europe will look at how cultural and environmental factors affect the performance of the system.
Researchers are studying interactions at borders between:
- The United States and Mexico
- The United States and Canada
- Finland and Russia
Finland is the first country to start testing the automated interviewing techniques. These tests might ultimately help airport security and border security experts to adopt a similar exit-entry-system across the European Union.