Can a Machine Tell When You're Lying? Research Suggests the Answer Is 'Yes'

Inspired by the work of psychologists who study the human face for clues that someone is telling a high-stakes lie, UB computer scientists are exploring whether machines can also read the visual cues that give away deceit.

Results so far are promising: In a study of 40 videotaped conversations, an automated system that analyzed eye movements correctly identified whether interview subjects were lying or telling the truth 82.5 percent of the time.

That's a better accuracy rate than expert human interrogators typically achieve in lie-detection judgment experiments, said Ifeoma Nwogu, a research assistant professor at UB's Center for Unified Biometrics and Sensors (CUBS) who helped develop the system. In published results, even experienced interrogators average closer to 65 percent, Nwogu said.

"What we wanted to understand was whether there are signal changes emitted by people when they are lying, and can machines detect them? The answer was yes, and yes," said Nwogu, whose full name is pronounced "e-fo-ma nwo-gu."

The research was peer-reviewed, published and presented as part of the 2011 IEEE Conference on Automatic Face and Gesture Recognition.

Nwogu's colleagues on the study included CUBS scientists Nisha Bhaskaran and Venu Govindaraju, and UB communication professor Mark G. Frank, a behavioral scientist whose primary area of research has been facial expressions and deception.

In the past, Frank's attempts to automate deceit detection have used systems that analyze changes in body heat or examine a slew of involuntary facial expressions.

The automated UB system tracked a different trait -- eye movement. The system employed a statistical technique to model how people moved their eyes in two distinct situations: during regular conversation, and while fielding a question designed to prompt a lie.

People whose pattern of eye movements changed between the first and second scenario were assumed to be lying, while those who maintained consistent eye movement were assumed to be telling the truth. In other words, when the critical question was asked, a strong deviation from normal eye movement patterns suggested a lie.

Previous experiments in which human judges coded facial movements found documentable differences in eye contact at times when subjects told a high-stakes lie.

What Nwogu and fellow computer scientists did was create an automated system that could verify and improve upon information used by human coders to successfully classify liars and truth tellers. The next step will be to expand the number of subjects studied and develop automated systems that analyze body language in addition to eye contact.

Nwogu said that while the sample size was small, the findings are exciting.

They suggest that computers may be able to learn enough about a person's behavior in a short time to assist with a task that challenges even experienced interrogators. The videos used in the study showed people with various skin colors, head poses, lighting and obstructions such as glasses.

This does not mean machines are ready to replace human questioners, however -- only that computers can be a helpful tool in identifying liars, Nwogu said.

She noted that the technology is not foolproof: A very small percentage of subjects studied were excellent liars, maintaining their usual eye movement patterns as they lied. Also, the nature of an interrogation and interrogators' expertise can influence the effectiveness of the lie-detection method.

The videos used in the study were culled from a set of 132 that Frank recorded during a previous experiment.

In Frank's original study, 132 interview subjects were given the option to "steal" a check made out to a political party or cause they strongly opposed.

Subjects who took the check but lied about it successfully to a retired law enforcement interrogator received rewards for themselves and a group they supported; Subjects caught lying incurred a penalty: they and their group received no money, but the group they despised did. Subjects who did not steal the check faced similar punishment if judged lying, but received a smaller sum for being judged truthful.

The interrogators opened each interview by posing basic, everyday questions. Following this mundane conversation, the interrogators asked about the check. At this critical point, the monetary rewards and penalties increased the stakes of lying, creating an incentive to deceive and do it well.

In their study on automated deceit detection, Nwogu and her colleagues selected 40 videotaped interrogations.

They used the mundane beginning of each to establish what normal, baseline eye movement looked like for each subject, focusing on the rate of blinking and the frequency with which people shifted their direction of gaze.

The scientists then used their automated system to compare each subject's baseline eye movements with eye movements during the critical section of each interrogation -- the point at which interrogators stopped asking everyday questions and began inquiring about the check.

If the machine detected unusual variations from baseline eye movements at this time, the researchers predicted the subject was lying.

New 'Thermal' Approach to Invisibility Cloaking Hides Heat to Enhance Technology

In a new approach to invisibility cloaking, a team of French researchers has proposed isolating or cloaking objects from sources of heat -- essentially "thermal cloaking." This method, which the researchers describe in the Optical Society's (OSA) open-access journal Optics Express, taps into some of the same principles as optical cloaking and may lead to novel ways to control heat in electronics and, on an even larger scale, might someday prove useful for spacecraft and solar technologies.

Recent advances in invisibility cloaks are based on the physics of transformation optics, which involves metamaterials and bending light so that it propagates around a space rather than through it. Sebastien Guenneau, affiliated with both the University of Aix-Marseille and France's Centre National de la Recherche Scientifique (CRNS), decided to investigate, with CRNS colleagues, whether a similar approach might be possible for thermal diffusion.

"Our key goal with this research was to control the way heat diffuses in a manner similar to those that have already been achieved for waves, such as light waves or sound waves, by using the tools of transformation optics," says Guenneau.

Though this technology uses the same fundamental theories as recent advances in optical cloaking, there is a key difference. Until now, he explains, cloaking research has revolved around manipulating trajectories of waves. These include electromagnetic (light), pressure (sound), elastodynamic (seismic), and hydrodynamic (ocean) waves. The biggest difference in their study of heat, he points out, is that the physical phenomenon involved is diffusion, not wave propagation.

"Heat isn't a wave -- it simply diffuses from hot to cold regions," Guenneau says. "The mathematics and physics at play are much different. For instance, a wave can travel long distances with little attenuation, whereas temperature usually diffuses over smaller distances."

To create their thermal invisibility cloak, Guenneau and colleagues applied the mathematics of transformation optics to equations for thermal diffusion and discovered that their idea could work.

In their two-dimensional approach, heat flows from a hot to a cool object with the magnitude of the heat flux through any region in space represented by the distance between isotherms (concentric rings of diffusivity). They then altered the geometry of the isotherms to make them go around rather than through a circular region to the right of the heat source -- so that any object placed in this region can be shielded from the flow of heat (see image).

"We can design a cloak so that heat diffuses around an invisibility region, which is then protected from heat. Or we can force heat to concentrate in a small volume, which will then heat up very rapidly," Guenneau says.

The ability to shield an area from heat or to concentrate it are highly desirable traits for a wide range of applications. Shielding nanoelectronic and microelectronic devices from overheating, for example, is one of the biggest challenges facing the electronics and semiconductor industries, and an area in which thermal cloaking could have a huge impact. On a larger scale and far into the future, large computers and spacecraft could also benefit greatly. And in terms of concentrating heat, this is a characteristic that the solar industry should find intriguing.

Guenneau and colleagues are now working to develop prototypes of their thermal cloaks for microelectronics, which they expect to have ready within the next few months.

Butterfly Wings' 'Art of Blackness' Could Boost Production of Green Fuels

Butterfly wings may rank among the most delicate structures in nature, but they have given researchers powerful inspiration for new technology that doubles production of hydrogen gas -- a green fuel of the future -- from water and sunlight.

The researchers presented their findings in San Diego on March 26 at the American Chemical Society's (ACS') 243rd National Meeting & Exposition.

Tongxiang Fan, Ph.D., who reported on the use of two swallowtail butterflies -- Troides aeacus (Heng-chun birdwing butterfly) and Papilio helenus Linnaeus (Red Helen) -- as models, explained that finding renewable sources of energy is one of the great global challenges of the 21st century. One promising technology involves producing clean-burning hydrogen fuel from sunlight and water. It can be done in devices that use sunlight to kick up the activity of catalysts that split water into its components, hydrogen and oxygen. Better solar collectors are the key to making the technology practical, and Fan's team turned to butterfly wings in their search for making solar collectors that gather more useful light.

"We realized that the solution to this problem may have been in existence for millions of years, fluttering right in front of our eyes," Fan said. "And that was correct. Black butterfly wings turned out to be a natural solar collector worth studying and mimicking," Fan said.

Scientists long have known that butterfly wings contain tiny scales that serve as natural solar collectors to enable butterflies, which cannot generate enough heat from their own metabolism, to remain active in the cold. When butterflies spread their wings and bask in the sun, those solar collectors are soaking up sunlight and warming the butterfly's body.

Fan's team at Shanghai Jiao Tong University in China used an electron microscope to reveal the most-minute details of the scale architecture on the wings of black butterflies -- black being the color that absorbs the maximum amount of sunlight.

"We were searching the 'art of blackness' for the secret of how those black wings absorb so much sunlight and reflect so little," Fan explained.

Scientists initially thought it was simply a matter of the deep inky black color, due to the pigment called melanin, which also occurs in human skin. More recently, however, evidence began to emerge indicating that the structure of the scales on the wings should not be ignored.

Fan's team observed elongated rectangular scales arranged like overlapping shingles on the roof of a house. The butterflies they examined had slightly different scales, but both had ridges running the length of the scale with very small holes on either side that opened up onto an underlying layer.

The steep walls of the ridges help funnel light into the holes, Fan explained. The walls absorb longer wavelengths of light while allowing shorter wavelengths to reach a membrane below the scales. Using the images of the scales, the researchers created computer models to confirm this filtering effect. The nano-hole arrays change from wave guides for short wavelengths to barriers and absorbers for longer wavelengths, which act just like a high-pass filtering layer.

The group used actual butterfly-wing structures to collect sunlight, employing them as templates to synthesize solar-collecting materials. They chose the black wings of the Asian butterfly Papilio helenus Linnaeus, or Red Helen, and transformed them to titanium dioxide by a process known as dip-calcining. Titanium dioxide is used as a catalyst to split water molecules into hydrogen and oxygen. Fan's group paired this butterfly-wing patterned titanium dioxide with platinum nanoparticles to increase its water-splitting power. The butterfly-wing compound catalyst produced hydrogen gas from water at more than twice the rate of the unstructured compound catalyst on its own.

"These results demonstrate a new strategy for mimicking Mother Nature's elaborate creations in making materials for renewable energy. The concept of learning from nature could be extended broadly, and thus give a broad scope of building technologically unrealized hierarchical architecture and design blueprints to exploit solar energy for sustainable energy resources," he concluded.

The scientists acknowledged funding from National Natural Science Foundation of China (No.51172141 and 50972090), Shanghai Rising-star Program (No.10QH1401300).

Tiny Reader Makes Fast, Cheap DNA Sequencing Feasible

Researchers have devised a nanoscale sensor to electronically read the sequence of a single DNA molecule, a technique that is fast and inexpensive and could make DNA sequencing widely available.

The technique could lead to affordable personalized medicine, potentially revealing predispositions for afflictions such as cancer, diabetes or addiction.

"There is a clear path to a workable, easily produced sequencing platform," said Jens Gundlach, a University of Washington physics professor who leads the research team. "We augmented a protein nanopore we developed for this purpose with a molecular motor that moves a DNA strand through the pore a nucleotide at a time."

The researchers previously reported creating the nanopore by genetically engineering a protein pore from a mycobacterium. The nanopore, from Mycobacterium smegmatis porin A, has an opening 1 billionth of a meter in size, just large enough for a single DNA strand to pass through.

To make it work as a reader, the nanopore was placed in a membrane surrounded by potassium-chloride solution, with a small voltage applied to create an ion current flowing through the nanopore. The electrical signature changes depending on the type of nucleotide traveling through the nanopore. Each type of DNA nucleotide -- cytosine, guanine, adenine and thymine -- produces a distinctive signature.

The researchers attached a molecular motor, taken from an enzyme associated with replication of a virus, to pull the DNA strand through the nanopore reader. The motor was first used in a similar effort by researchers at the University of California, Santa Cruz, but they used a different pore that could not distinguish the different nucleotide types.

Gundlach is the corresponding author of a paper published online March 25 by Nature Biotechnology that reports a successful demonstration of the new technique using six different strands of DNA. The results corresponded to the already known DNA sequence of the strands, which had readable regions 42 to 53 nucleotides long.

"The motor pulls the strand through the pore at a manageable speed of tens of milliseconds per nucleotide, which is slow enough to be able to read the current signal," Gundlach said.

Gundlach said the nanopore technique also can be used to identify how DNA is modified in a given individual. Such modifications, referred to as epigenetic DNA modifications, take place as chemical reactions within cells and are underlying causes of various conditions.

"Epigenetic modifications are rather important for things like cancer," he said. Being able to provide DNA sequencing that can identify epigenetic changes "is one of the charms of the nanopore sequencing method."

Coauthors of the Nature Biotechnology paper are Elizabeth Manrao, Ian Derrington, Andrew Laszlo, Kyle Langford, Matthew Hopper and Nathaniel Gillgren of the UW, and Mikhail Pavlenok and Michael Niederweis of the University of Alabama at Birmingham.

The work was funded by the National Human Genome Research Institute in a program designed to find a way to conduct individual DNA sequencing for less than $1,000. When that program began, Gundlach said, the cost of such sequencing was likely in the hundreds of thousands of dollars, but "with techniques like this it might get down to a 10-dollar or 15-minute genome project. It's moving fast."

Research: 'Buckliball' Opens New Avenue in Design of Foldable Engineering Structures

Motivated by the desire to determine the simplest 3-D structure that could take advantage of mechanical instability to collapse reversibly, a group of engineers at MIT and Harvard University were stymied -- until one of them happened across a collapsible, spherical toy that resembled the structures they'd been exploring, but with a complex layout of 26 solid moving elements and 48 rotating hinges.

The toy inspired the engineers to create the "buckliball," a hollow, spherical object made of soft rubber containing no moving parts, but fashioned with 24 carefully spaced dimples. When the air is sucked out of a buckliball with a syringe, the thin ligaments forming columns between lateral dimples collapse. This is the engineering equivalent of applying equal load on all beams in a structure simultaneously to induce buckling, a phenomenon first studied by mathematician Leonhard Euler in 1757.

When the buckliball's thin ligaments buckle, the thicker ligaments forming rows between dimples undergo a series of movements the researchers refer to as a "cooperative buckling cascade." Some of the thick ligaments rotate clockwise, others counterclockwise -- but all move simultaneously and harmoniously, turning the original circular dimples into vertical and horizontal ellipses in alternating patterns before closing them entirely. As a result, the buckliball morphs into a rhombicuboctahedron about half the size (46 percent) of the original sphere.

The researchers named their new structure for its use of buckling and its resemblance to buckyballs, spherical all-carbon molecules whose name was inspired by the geodesic domes created by architect-inventor Buckminster Fuller. The buckliball is the first morphable structure to incorporate buckling as a desirable engineering design element. The buckling process induces folding in portions of the sphere -- similar to the way paper folds in origami -- so the researchers place their buckliball in a larger framework of buckling-induced origami they call "buckligami."

Because their collapse is fully reversible and can be achieved without moving parts, morphable structures such as the buckliball have the potential for widespread applications, from the micro- to macroscale. They could be used to create large buildings with collapsible roofs or walls, tiny drug-delivery capsules or soft movable joints requiring no mechanical pieces. They also have the potential to transform Transformers and other kinds of toys. (The toy that provided the researchers' epiphany is the Hoberman Twist-O.)

The researchers -- Jongmin Shim MS '05, PhD '10, a postdoc at Harvard; Claude Perdigo, a visiting graduate student at MIT; Elizabeth Chen, a recent graduate of the University of Michigan who will join Harvard as a postdoc in the fall; Katia Bertoldi, an assistant professor in applied mechanics at Harvard; and Pedro Reis, the Esther and Harold E. Edgerton Assistant Professor of Civil and Environmental Engineering and Mechanical Engineering at MIT -- wrote a paper about this work that appears this week in the Proceedings of the National Academy of Sciences.

"In civil engineering, buckling is commonly associated with failure that must be avoided. For example, one typically wants to calculate the buckling criterion for columns and apply an additional safety factor, to ensure that a building stands," Reis says. "We are trying to change this paradigm by turning failure into functionality in soft mechanical structures. For us, the buckliball is the first such object, but there will be many others." For instance, a robotic arm could be built from a single piece of material using a precisely engineered pattern of dimples at the intended hinging points that, when activated by a pressure signal, would bend.

"The buckliball not only opens avenues for the design of foldable structures over a wide range of length scales, but may also be used as a building block for creating new materials with unusual properties, capable of dramatic contraction in all directions," Bertoldi says.

Reis's research uses precision tabletop-scale lab tests and mathematical analysis to determine the basic physics underlying the mechanical behavior of materials. Bertoldi's research group uses tools from continuum and computational mechanics to unravel the mechanics of soft structures. The two teams collaborated on the buckliball: Reis' team performed the lab experiments with the help of digital fabrication techniques (such as 3-D printing) to create objects with precise geometry, and Bertoldi's group used computation to further analyze the detailed mechanics of the process.

Chen, who was visiting Harvard at the time, determined that only five spherical geometric structures have the potential for reversible buckling-induced collapse. (The specific example of Fuller's 12-hole rhombicuboctahedron that collapses into a cuboctahedron is one of these five.) Design parameters for buckliballs include dimple size, the thickness of the thin shell inside the dimple and the stiffness of the material used to fabricate the buckliball.

Nature, it appears, has already figured this out. Viruses inject their nucleic acids into a host through a reversible structural transformation in which 60 holes open or close based on changes in the acidity of the cell's environment, a different mechanism that achieves a similar reversible collapse at the nanoscale.

"What's exciting about this work is that it uses instabilities to basically amplify small or moderate pressures into dramatic motion," says Carmel Majidi, an assistant professor of mechanical engineering at Carnegie Mellon University whose research in soft robotics focuses on stretchable skin-like materials containing sensors. "One limitation of working with soft-material robotics is that they're soft; they can't produce the high pressures you get with heavy machines, so you're left with machines that provide only fairly moderate pressures. This makes it difficult to achieve dramatic deformations. If you use a robotic skin as an assistive medical device on a human, it can monitor motion. But with advancements like the buckliball, the skin may even be able to actively change its shape and directly help with motor tasks."

The work was funded through a National Science Foundation grant to the Harvard Materials Research Science and Engineering Center and by funds from Harvard University and MIT.