Liar, Liar, Pants On Fire

Liar, Liar, Pants On Fire

To catch a terrorist, you have to think like a terrorist or at least be able to get behind their lies and deception.

Terrorist want to gain entry, surveil their targets, plan their attack, assemble their weapons and tactics, avoid their pursuers, and execute maximum human, economic, and political damage.

To succeed, terrorists have to use lies and deceit to make their way through all the obstacles that the good guys put up.

Wired Magazine (February 2013) addresses some new interrogation technology being tested by the Department of Homeland Security (DHS) to catch the lies and the liars.

First of all, “people are really good at lying, and it’s incredibly hard to tell when we’re doing it.”

Moreover, most people “lie 10 times a day,” so it is routine and comes naturally to them.

In terms of detecting lies, we are not very good at it–in fact, we’re just better than chance–able to tell when someone is lying only 54% of the time.
Apparently, even with polygraph exams–their success is dependent more on the experience and finesse of the examiner and less on the polygraph tool.

However, with new research and development, DHS has come up with an automated interrogator–that enhances the success of catching a liar by combining multiple detection technologies.

The “interrogator bot” has three different sensors in use by the Embodied Avatar Kiosk.

– Infrared camera–“records eye movements and pupil dilation at up to 250 times per second–the stress of lying tends to cause the pupils to dilate”

– High-definition video camera–“captures fidgets such as shrugging, nodding, and scratching, which tend to increase during a deceptive statement”

– Microphone–“collects vocal data, because lies often come with minute changes in pitch” as well as “hesitation, changes in tempo and intonation, and spoken errors”

In the future, a additional sensors may be added for:

– Weight-sensing platform–to “measure leg and foot shifts or toe scrunches”

– 3-D camera–to “track the movements of a person’s entire body”

Aside from getting better deception-detection results from multiplying the sensing techniques, the interrogation kiosk benefits from communicating in multiple languages and being “consistent, tireless, and susceptible to neither persuasion not bribery.”

Another very cool feature being tested is tan interrogation avatar that actually resembles the person being interrogated using a camera and morphing software and making it look uncanny and “disturbing” at the same time–this can be quite familiar, disarming and unnerving.

By aggregating data points from many types of sensors and using behavioral analysis as a first line of defense followed by human questioning of those found to be lying, homeland security can proverbially light a fire under the pants of would be infiltrators and terrorists–and catch them before they make it to their next target. 😉

(Source Photo: here with attribution to Cosmic Jans)

>Big Brother is Watching and Enterprise Architecture


The enterprise architecture for law enforcement and security in the next decade will be focused on using technology to identify the bad guys and stop them in their tracks.

ComputerWorld, 14 January 2008, reports that “Homeland Security is bankrolling futuristic technology to nab terrorists before they strike.”

Here’s the target architecture:

“The year is 2012 [probably a bit optimistic on the date]. As soon as you walk into the airport, the machines are watching. Are you a tourist—or a terrorist posing as one? As you answer a few questions at the security checkpoint, the systems begin sizing you up. An array of sensors—video, audio, laser, infrared—feeds a stream of real-time data about you to a computer that uses specially developed algorithms to spot suspicious people. The system interprets your gestures and facial expressions, analyzes your voice and virtually probes your body to determine your temperature, heart rate, respiration rate, and other physiological characteristics—all in an effort to determine whether you are trying to deceive. Fail the test, and you’ll be pulled aside for a more aggressive interrogation and searches.”

Last July, The Department of Homeland Security, “human factors division asked researchers to develop technologies to support Project Hostile Intent, an initiative to build systems that automatically identify and analyze behaviors and physiological cues associated with deception.”

The intent is to use these screening technologies at airports, border crossings, as well as possibly in the private sector for building access control and candidate screening.

Sharla Rausch, director of DHS’s human factors division says that “in controlled lab setting, accuracy rates are in the range of 78% to 81%.”

Where is the current research focused?

  1. Recognition of gestures and microfacial expressions
  2. Analysis of variations in speech (i.e. pitch, loudness)
  3. Measurement of physiological characteristics

The hope is that by combining all three modalities, “the overall predictive accuracy rate” will improve.

What are some of the challenges with these technologies?

  1. Currently, too many false positives
  2. Existing technologies, like the polygraph have “long been questioned by scientists…and remain inadmissible in court.”
  3. Ability of algorithms to “correctly interrupt” suspicious behavior or cues
  4. Profiling is continuously objected too based on discriminatory grounds
  5. Privacy concerns about the personal data collected
  6. Testing is limited by security concerns in the field
  7. Deployment will be limited due to cost, leaving soft targets potentially at risk

Will this Big Brother screening technology come to fruition?

Absolutely. The challenges with the technologies will be resolved, and putting aside the profiling and privacy issues, these screening technologies will become essential to our protecting ourselves.