The guilt was written all over her face

7 02 2011

A poet and cognitive neuroscience enthusiast sent me this link:

May 19, 2008, 10:56 pm <!– — Updated: 8:06 pm –>The Most Curious Thing, By ERROL MORRIS explaining “how a photograph aided and abetted a terrible miscarriage of justice…about Sabrina Harman, one of the notorious ‘seven bad apples’ convicted of abuse in the notorious Abu Ghraib scandal”.

The focus of the article is her smile, the photographed record of her smile, which could have been the same smile you would’ve captured in an awkward family photo or at the Eiffel Tower, but eerily misplaced in the context of a prison full or rotting and damaged bodies, covered with ice in an attempt to diminish foul odors of decay.  It was shocking for audiences to view the photos, but more shocking that a person in that situation would give a ‘thumbs up’ and grin.

But if we look more closely at the photos, is the smile really the same smile captured in an awkward family photo or in front of the Eiffel Tower.  The writer looked for a smile expert, and found Paul Ekman, Professor Emeritus of Psychology at the University of California, San Francisco who is an expert on facial expressions.  So well versed in reading facial expressions he created F.A.C.E Training, a tool marketed to businessmen and government officials alike to help develop skills in “reading” another’s emotions and even lie detection just by critical review of someone’s face. I had to wonder when looking at his profile picture, how much time he had put into capturing the right facial expression for his audiences-to encourage trust and willingness to buy his product (you can see pictures of his own smiles used as example for his 2003 book in the article).

Why would being able to read her facial expression be important to discuss.  If he detects remorse or lack of enjoyment would that absolve her of her actions, or maybe it would help people have a restored faith in humanity–“See, at least she felt bad about it.” Ekman’s assessment of one photograph is something this “..she is showing a social smile or a smile for the camera. The signs of an actual enjoyment smile are just not there. There’s no sign of any negative emotion. She’s doing what people always do when they pose for a camera. They put on a big, broad smile, but they’re not actually genuinely enjoying themselves. We would see movement in the eye cover fold…”

Hmm…sounds pretty scientific.  But let’s get back to exploring F.A.C.E.  FACE stands for Facial expression Awareness Compassion Expression. What is it, what is the evidence behind it, and how is it being used? This is what the F.A.C.E. Training page says for the Advanced METT (Micro Expression Training Tool) says–

“This training is meant for those whose work requires them to evaluate truthfulness and detect deception – such as police and security personnel, and those in sales, education, and medical professions . If you should achieve the minimum target score of 80% or higher on the post test, a certificate of completion will be emailed to you.”

How is it currently being used? I found this article:

Airport security: Intent to deceive?

| May 26, 2010 |


According to this article, up to 1000 TSA (Transportation Security Administration) screeners have been trained with Paul Ekman’s techniques. In addition, “There are about 3,000 of these officers working at some 161 airports across the United States, all part of a four-year-old program called Screening Passengers by Observation Technique (SPOT), which is designed to identify people who could pose a threat to airline passengers” primarily terrorists. SPOT is a technique based on Ekman’s work. And interest in the technique is growing from the U.S. Department of Homeland Security and various intelligence agencies.

Does it work?  Ekman’s work has sparked a Fox T.V. series: Lie to Me, so there are probably a lot of people (general audiences and specialists alike) out there who are fascinated by (and maybe even like to fantasize about) the idea of it working.

But is Micro Expression identification actually a reliable tool for lie detection or intent to do bad things? According to this article, it seems that his colleagues remain skeptical.  Primarily, his scientific colleagues have problems replicating and corroborating his results. Other psychologists find that “many peer-reviewed studies seem to show that people are not better than chance when it comes to picking up signs of deception.” A 2007 report composed by a panel of “credibility assessment experts say that, “Simply put, people (including professional lie-catchers with extensive experience of assessing veracity) would achieve similar hit rates if they flipped a coin.”

In addition, his studies lack proper controls and his more recent work lacks peer-review.  But Ekman claims the lack of peer-review is intentional: “Ekman maintains that this publishing strategy is deliberate–that he no longer publishes all of the details of his work in the peer-reviewed literature because, he says, those papers are closely followed by scientists in countries such as Syria, Iran and China, which the United States views as a potential threat.” But peer-review is an important checks and balances system for scientists, as experts, to evaluate the works of one another. “As a scientist, I want to see peer-reviewed journal articles, so I can look at procedures and data and know what the training procedures involve, and what the results do show,” says Bella DePaulo, a social psychologist at the University of California, Santa Barbara.

Ekman claims that examining micro expressions can give you up to 70% accuracy in determining deception.  But you can go up to 100% if you use remaining body cues. According to TSA statistics from 2006-2009,”behavior-detection officers referred more than 232,000 people for secondary screening…But 1,710 were arrested, which the TSA cites as evidence for the program’s effectiveness. And in this 1% of people accurately identified, those arrests were for criminal activity unrelated to terrorist activity. Although, I found this article link from the TSA’s blog stating that at least one individual bearing explosive was caught by behavior detection officers.

Actually, the TSA’s blog expresses a lot of enthusiasm about Ekman and his techniques for identifying guilty travelers. They claim that After passing along his skills to US Customs, their “hit rate” for finding drugs during passenger searches rose to 22.5 percent from 4.2 percent in 1998. Some examples of suspicious criteria are illustrated here.

Now enter FAST (Future Attribute Screening Technology), a project being funded $10 million/year.  With FAST, travelers would walk through a portal while a myriad of sensors would monitor their vital signs remotely for ‘malintent’. The Department of Homeland Security (DHS) has a host of Human Factors Behavioral Sciences Projects, FAST is just one of them. According to the DHS website blurb on FAST, “FAST is grounded in research on human behavior and psychophysiology.”

Supporters of FAST technology feel that the future of security screening should be more about the people and less about their things. What does the research say about being able to reliably detect ‘mal-intent’ based on a combination of skilled observation and vital signs? In an interview for, Carnegie Mellon’s Stephen Fienberg, a university professor in the statistics and machine learning departments, said, “I haven’t seen any research that shows that those measures from the autonomic nervous system … measuring blood pressure, measuring breathing, measuring heat on the face, are at all related to intent.” Indeed, it sounds a bit like what would otherwise be part of any routine doctor’s visit, except done by a machine that has been programmed to calculate the statistical significance of my responses.

Polygraph lie detection, which measures four parameters—heart rate, blood pressure, respiration, and sweating has been around since the 1920’s. In a 2001 National Academy of Sciences panel discussion about lie detection, Dr. Richard Davidson, a neuroscientist and Director of the Laboratory for Affective Neuroscience from University of Wisconsin Madison claims that lie detection is more likely to detect “fear of detection” than actual lies. As a neuroimager, he says we need to go to the brain, “And if there’s one emotion that we have really learned a lot about in the last decade, it’s fear.” But lie detection in the brain will have to be another blog post on its own.

FAST is supposed to be more sophisticated in its goals than polygraph detection.  It’s not just trying to detect guilt.  FAST aims to detect intent. (And, for part of the testing, we might even see the Wii Balance Board thrown into the mix.).

This has sparked a lot of fearful titles like “Homeland Security Detects Terrorist Threats by Reading Your Mind.” But this isn’t mind reading.  If it was all you would need is $37 anyway.

Mind reading implies a certain ability to get a 1 to 1 level of accuracy to your thoughts.  This is simply taking into account your vital signs, your body and facial language, to really make an educated guess (using statistics) on whether you *plan* to do something bad. And this educated guess is maybe supported by some experts.

The TSA says that the data will be recorded and dumped for unsuspicious passengers so what’s the harm? Proponents could say, “Well, what harm does it for me to be a little uncomfortable to potentially save lives?”  Skeptics might say, “Well, this is an invasion of my privacy!”  But actually, maybe it’s something else.  What if it’s not really relevant to saving lives?  Think about what happens when you accuse someone of anything? What kind of psychological resonating effects will we see?  Will cultural differences be accounted for?

While many exciting technologies are being developed to give new insights into human behavior, most data point toward correlation and mathematical constructs (as in neuroimaging and EEG). The findings from these technologies do not predict or detect a 1:1 relationship of the human mind.

While the brain is a powerful decision-making machine, one thing might be important to consider: Bodies have brains; people have minds. Stay-tuned for another post on morality and brain-imaging as I have recently co-authored a commentary in the American Journal of Bioethics Neuroscience on just this topic.