Monday, July 19, 2010


0 comentarios

We meet a multitude of people on a daily basis: the nice waitress in the coffee shop around the corner, the bus driver or the colleagues at the office. Without the ability to recognize faces at first glance we would not be able to distinguish between people.

Monkeys also possess the remarkable ability to differentiate faces of group members and to extract the relevant information about the individual directly from the face. With the help of the so called Thatcher illusion, scientists of the Max Planck Institute for Biological Cybernetics in Tübingen, Germany, have examined how people and macaque monkeys recognize faces and process the information in the brain. They found out that both species perceive the faces of their kin immediately, while the faces of the other species are processed in a different way.

"From an early age on we are accustomed to the faces of other humans: a long nose, the swing of the lips or the bushy eyebrows. We learn to recognize the small differences which contribute to an individual appearance", explains Christoph Dahl, researcher at the Max Planck Institute for Biological Cybernetics". It is similar in monkeys. They learn to recognize the features of their fellow monkeys (so called conspecifics) and can grasp the identity of every group member quickly. "However in humans, as well as in macaque monkeys, this principle only works with individuals of the same kind", says Dahl. Even though the recognition of conspecific faces is achieved by means of holistic processing, the separate parts such as mouth, nose and eyes as well as the facial proportions are still important. "Although we look at the eyes first our neural functions still grasp the whole picture", Christoph Dahl describes the processing mechanisms behind the facial recognition.

With the help of the "Thatcher illusion" the scientists examined the facial recognition of macaque monkeys and humans. Local changes in facial features are hardly noticeable when the whole face is upside down, but strikingly grotesque when the face is right side up. "The faces in which the eyes and the mouth were rotated 180 degrees look grotesque - but only if we see them the right side up. Upside-down the differences between a normal face and a ‘thacherized’ face are hardly recognizable", explains Christian Wallraven, one of the scientists involved in the study. The effect can be explained by the lack of processing capabilities for locally rotated facial features when the face is turned upside down. The holistic processing mechanisms allow us to recognize fine changes in the arrangement of the separate facial parts. If the whole face is rotated 180 degrees, this ability gets lost.

The usual recognition mechanisms do not function with either inverted faces or with the faces of foreign species. The scientists discovered that the Thatcher illusion in macaque monkeys only works for the faces of their conspecifics, while they paid no special attention to the extremely grotesque human faces. Vice versa it behaved the same with humans for whom the manipulated monkey faces remained inconspicuous. "It must have been of great advantage for us as well as for our next relatives, the monkeys, in the course of the evolution to recognize especially the faces of our kind and also to develop similar processing mechanisms." Wallraven sums up. Besides, the ability to recognize faces on the first sight, the holistic processing opens another possibility: the identification of different conspecifics with in no time.

(Photo: Christoph Dahl)

Max Planck Institute


0 comentarios

Clusters of heated, magnetic nanoparticles targeted to cell membranes can remotely control ion channels, neurons and even animal behavior, according to a paper published by University at Buffalo physicists in Nature Nanotechnology.

The research could have broad application, potentially resulting in innovative cancer treatments that remotely manipulate selected proteins or cells in specific tissues, or improved diabetes therapies that remotely stimulate pancreatic cells to release insulin.

The work also could be applied to the development of new therapies for some neurological disorders, which result from insufficient neuro-stimulation.

"By developing a method that allows us to use magnetic fields to stimulate cells both in vitro and in vivo, this research will help us unravel the signaling networks that control animal behavior," says Arnd Pralle, PhD, assistant professor of physics in the UB College of Arts and Sciences and senior/corresponding author on the paper.

The UB researchers demonstrated that their method could open calcium ion channels, activate neurons in cell culture and even manipulate the movements of the tiny nematode, C. elegans.

"We targeted the nanoparticles near what is the 'mouth' of the worms, called the amphid," explains Pralle. "You can see in the video that the worms are crawling around; once we turn on the magnetic field, which heats up the nanoparticles to 34 degrees Celsius, most of the worms reverse course. We could use this method to make them go back and forth. Now we need to find out which other behaviors can be controlled this way."

The worms reversed course once their temperature reached 34 degrees Celsius, Pralle says, the same threshold that in nature provokes an avoidance response. That's evidence, he says, that the approach could be adapted to whole-animal studies on innovative new pharmaceuticals.

The method the UB team developed involves heating nanoparticles in a cell membrane by exposing them to a radiofrequency magnetic field; the heat then results in stimulating the cell.

"We have developed a tool to heat nanoparticles and then measure their temperature," says Pralle, noting that not much is known about heat conduction in tissue at the nanoscale.

"Our method is important because it allows us to only heat up the cell membrane. We didn't want to kill the cell," he said. "While the membrane outside the cell heats up, there is no temperature change in the cell."

Measuring just six nanometers, the particles can easily diffuse between cells. The magnetic field is comparable to what is employed in magnetic resonance imaging. And the method's ability to activate cells uniformly across a large area indicates that it also will be feasible to use it in in vivo whole body applications, the scientists report.

In the same paper, the UB scientists also report their development of a fluorescent probe to measure that the nanoparticles were heated to 34 degrees Celsius.

"The fluorescence intensity indicates the change in temperature," says Pralle, "it's kind of a nanoscale thermometer and could allow scientists to more easily measure temperature changes at the nanoscale."

(Photo: U. Buffalo)

University at Buffalo


0 comentarios

A new type of heat pump being developed at Purdue University could allow residents in cold climates to cut their heating bills in half.

The research, funded by the U.S. Department of Energy, builds on previous work that began about five years ago at Purdue's Ray W. Herrick Laboratories, said James Braun, a professor of mechanical engineering.

Heat pumps provide heating in winter and cooling in summer but are not efficient in extreme cold climates, such as Minneapolis winters.

"With this technology we can maintain the efficiency of the heat pump even when it gets pretty cold outside," said Eckhard Groll, a professor of mechanical engineering who is working on the project with Braun and W. Travis Horton, an assistant professor of civil engineering.

The innovation aims to improve efficiency in general but is especially practical for boosting performance in cold climates. The new heat pumps might be half as expensive to operate as heating technologies now used in cold regions where natural gas is unavailable and residents rely on electric heaters and liquid propane.

"We'll be able to extend the geographical range where heat pumps can apply," Horton said. "So this could open up a whole new market."

Researchers expect to complete a prototype by the end of the three-year, $1.3 million project. The research, which also involves three doctoral students, is a partnership with Emerson Climate Technologies Inc. and Carrier Corp. Emerson will work with researchers to create the prototype heat pump, and Carrier will integrate the new heat pump into a complete system.

The new technology works by modifying the conventional vapor-compression cycle behind standard air conditioning and refrigeration.

"This could be a relatively simple modification to existing heat pumps, refrigeration and air conditioning systems," Braun said.

The standard vapor-compression cycle has four stages: refrigerant is compressed as a vapor, condenses into a liquid, expands to a mixture of liquid and vapor, and then evaporates.

The project will investigate two cooling approaches during the compression process. In one approach, relatively large amounts of oil are injected into the compressor to absorb heat generated throughout the compression stage. In the second approach, a mixture of liquid and vapor refrigerant from the expansion stage is injected at various points during compression to provide cooling. The added steps improve the compression process while also reducing energy losses due to friction in the expansion stage.

"Cooling the compressor keeps the refrigerant dense, and that's important because it takes less energy to compress something that's more dense," Braun said.

The researchers are developing a system for precisely controlling the flow of refrigerant from the evaporation stage into the compression stage using a series of small valves. A critical component of the new heat pump is a "scroll compressor," which uses a rotating, scroll-shaped mechanism to compress refrigerant. Domestic heat pumps normally use reciprocating compressors, in which a piston compresses refrigerant.

"You can't inject a liquid into a reciprocating compressor, whereas you can with a scroll compressor, which is uniquely suited for this modification," Groll said. "Also, an important part of our project will be to determine the efficiency of a machine that pumps liquid while also compressing gas, so there will be a lot of computational modeling involved."

The work grew out of research into the Ericsson cycle, an exotic refrigeration technology in which liquid is added to coolant as it is being compressed. The Ericson cycle, however, does not use the vapor-compression cycle because the gas never turns to liquid.

(Photo: Purdue University/Mark Simons)

Purdue University


0 comentarios
Faced with threats such as habitat loss and climate change, thousands of rare flowering plant species worldwide may become extinct before scientists can even discover them, according to a paper published by a trio of American and British researchers in the journal Proceedings of the Royal Society B.

"Scientists have estimated that, overall, there could be between 5 million and 50 million species, but fewer than 2 million of these species have been discovered to date," says lead author Lucas Joppa of Microsoft Research in Cambridge, U.K., who received his doctorate from Duke University earlier this year. "Using novel methods, we were able to refine the estimate of total species for flowering plants, and calculate how many of those remain undiscovered."

Based on data from the online World Checklist of Selected Plant Families at the Royal Botanic Gardens, Kew, the scientists calculated that there are between 10 and 20 percent more undiscovered flowering plant species than previously estimated. This finding has "enormous conservation implications, as any as-yet-unknown species are likely to be overwhelmingly rare and threatened," Joppa says.

The new, more accurate estimate can be used to infer the proportion of all threatened species, says coauthor David Roberts of the Durrell Institute of Conservation and Ecology at the University of Kent. "If we take the number of species that are currently known to be threatened, and add to that those that are yet to be discovered, we can estimate that between 27 percent and 33 percent of all flowering plants will be threatened with extinction," he says.

"That percentage reflects the global impact of factors such as habitat loss. It may increase if you factor in other threats such as climate change," Joppa adds.

"The timing couldn't be more perfect," says co-author Stuart Pimm, Doris Duke Professor of Conservation Ecology at Duke's Nicholas School of the Environment. "The year 2010 is the International Year of Biodiversity. We wrote the paper to help answer the obvious questions: How much biodiversity is out there, and how many species will we lose before they are even discovered?"

Duke University


0 comentarios

Soccer referees may have an unconscious bias towards calling fouls based on a play's direction of motion, according to a new study from the of University of Pennsylvania School of Medicine. Researchers found that soccer experts made more foul calls when action moved right-to-left, or leftward, compared to left-to-right or rightward action, suggesting that two referees watching the same play from different vantage points may be inclined to make a different call. The study appears in the July 7 online edition of PLoS ONE.

It's been documented that individuals who read languages which flow left-to-right are more likely to have a negative bias for events moving in the opposite direction, from right-to-left. In the Penn study of twelve members of the University of Pennsylvania's varsity soccer teams (all native English speaking), researchers found that participants viewing the soccer plays were more likely to call a foul when seeing a right-to-left attack

"The effects are impressive considering that left-moving and right-moving images were identical, with the only difference being that they were flipped along the x-axis to create right-to-left and left-to-right versions," said lead researcher Alexander Kranjec, PhD, a post-doctoral fellow in the Neurology Department at the University of Pennsylvania School of Medicine. "If the spatial biases we observed in this population of soccer players have similar effects on referees in real matches, they may influence particular officials differently: referees on the field will more frequently be in positions that lower their threshold for calling fouls during an attack, compared to assistant referees working the lines."

In real matches, referees and linesmen tend to be exposed to different quantities of right-to-left or left-to-right attacking plays, as referees employ a system to help them cover the field efficiently. Referees are encouraged to use a diagonal patrolling technique, choosing to run either a left or a right diagonal, while the assistant referees are tasked with running the sidelines.

Based on this study, the left diagonal system would favor the offense (viewing more attacks from right-to-left), and the right diagonal system would favor the defense (viewing more attacks from left-to-right). Given the relational opposition, the authors suggest that referees should avoid switching diagonals at halftime.

"There could be an unfair advantage if one team goes into halftime with a lead and the referees switch to a right diagonal system in the second half, favoring both defenses," said Dr. Kranjec. "However, because referees viewing leftward action may be more likely to see a foul when no foul was actually committed, as seemed to be the case when the referee disallowed what should have been the US team's third goal against Slovenia, the bias could work against the offense sometimes."

Study participants called approximately three more fouls when images of soccer plays where viewed from right-to-left (66.5 fouls) compared to mirror images moving left-to-right (63.3 fouls). Participants were statistically more likely to call a foul when seeing a right-to-left attack.

Previous studies suggest that similar directional effects are reversed in populations that read right-to-left languages, but other populations (e.g. Arabic or Hebrew readers) would need to be tested directly to see if the effects reported in this study correlate with reading habits.

(Photo: Alexander Kranjec, PhD, University of Pennsylvania School of Medicine; PLoS One)

University of Pennsylvania


0 comentarios

Robotic cars attracted attention last decade with a 100-mile driverless race across the desert competing for a $1 million prize put up by the U.S. government.

The past few years have given rise to a growing number of microrobots, miniaturized mobile machines designed to perform specific tasks. And though spectators might need magnifying glasses to see the action, some think the time has come for a microrobotics challenge.

"I'd like to see a similar competition at the small scale, where we dump these microrobots from a plane and have them go off and run for days and just do what they've been told," said Karl Böhringer, a University of Washington professor of electrical engineering. "That would require quite an effort at this point, but I think it would be a great thing."

Researchers at the UW and Stanford University have developed what might one day be a pint-sized contender. Böhringer is lead author of a paper in the June issue of the Journal of Microelectromechanical Systems introducing an insectlike robot with hundreds of tiny legs.

Compared to other such robots, the UW model excels in its ability to carry heavy loads -- more than seven times its own weight -- and move in any direction.

Someday, tiny mobile devices could crawl through cracks to explore collapsed structures, collect environmental samples or do other tasks where small size is a benefit. The UW's robot weighs half a gram (roughly one-hundredth of an ounce), measures about 1 inch long by a third of an inch wide, and is about the thickness of a fingernail.

Technically it is a centipede, with 512 feet arranged in 128 sets of four. Each foot consists of an electrical wire sandwiched between two different materials, one of which expands under heat more than the other. A current traveling through the wire heats the two materials and one side expands, making the foot curl. Rows of feet shuffle along in this way at 20 to 30 times each second.

"The response time is an interesting point about these tiny devices," Böhringer said. "On your stove, it might take minutes or even tens of minutes to heat something up. But on the small scale it happens much, much faster."

The legs' surface area is so large compared to their volume that they can heat up or cool down in just 20 milliseconds.

"It's one of the strongest actuators that you can get at the small scale, and it has one of the largest ranges of motion," Böhringer said. "That's difficult to achieve at the small scale."

The microchip, the robot's body and feet, was first built in the mid 1990s at Stanford University as a prototype part for a paper-thin scanner or printer. A few years later the researchers modified it as a docking system for space satellites. Now they have flipped it over so the structures that acted like moving cilia are on the bottom, turning the chip into an insectlike robot.

"There were questions about the strength of the actuators. Will they be able to support the weight of the device?" Böhringer said. "We were surprised how strong they were. For these things that look fragile, it's quite amazing."

The tiny legs can move more than just the device. Researchers were able to pile paper clips onto the robot's back until it was carrying more than seven times its own weight. This means that the robot could carry a battery and a circuit board, which would make it fully independent. (It now attaches to nine threadlike wires that transmit power and instructions.)

Limbs pointing in four directions allow the robot flexibility of movement.

"If you drive a car and you want to be able to park it in a tight spot, you think, 'Wouldn't it be nice if I could drive in sideways,'" Böhringer said. "Our robot can do that -- there's no preferred direction."

Maneuverability is important for a robot intended to go into tight spaces.

The chip was not designed to be a microrobot, so little effort was made to minimize its weight or energy consumption. Modifications could probably take off 90 percent of the robot's weight, Böhringer said, and eliminate a significant fraction of its power needs.

As with other devices of this type, he added, a major challenge is the power supply. A battery would only let the robot run for 10 minutes, while researchers would like it to go for days.

Another is speed. Right now the UW robot moves at about 3 feet per hour -- and it's far from the slowest in the microrobot pack.

(Photo: University of Washington)

University of Washington




Selected Science News. Copyright 2008 All Rights Reserved Revolution Two Church theme by Brian Gardner Converted into Blogger Template by Bloganol dot com