Astrid Rosenthal-von der Putten

In new research, volunteers showed empathy while watching videos of a dinosaur robot being treated violently.

By Tanya Lewis
LiveScience

From R2-D2 in "Star Wars" to Furby, robots can generate surprisingly humanlike feelings. Watching a robot being abused or cuddled has a similar effect on people to seeing those things done to a human, new research shows.

Humans are increasingly exposed to robots in their daily lives, but little is known about how these lifelike machines influence human emotions.

Feeling bad for bots
In two new studies, researchers sought to measure how people responded to robots on an emotional and neurological level. In the first study, volunteers were shown videos of a small dinosaur robot being treated affectionately or violently. In the affectionate video, humans hugged and tickled the robot, and in the violent video, they hit or dropped him. [5 Reasons to Fear Robots]

Scientists assessed people's levels of physiological excitation after watching the videos by recording their skin conductance, a measure of how well the skin conducts electricity. When a person is experiencing strong emotions, they sweat more, increasing skin conductance.  

The volunteers reported feeling more negative emotions while watching the robot being abused. Meanwhile, the volunteers' skin conductance levels increased, showing they were more distressed.

In the second study, researchers use functional magnetic resonance imaging (fMRI) to visualize brain activity in the participants as they watched videos of humans and robots interacting. Again, participants were shown videos of a human, a robot, and, this time, an inanimate object being treated with affection or abuse.

In one video, for example, a man appears to beat up a woman, strangle her with a string and attempt to suffocate her with a plastic bag. In another, a person does the same things to the robot dinosaur.

Affectionate treatment of the robot and the human led to similar patterns of neural activity in regions in the brain's limbic system, where emotions are processed, fMRI scans showed. But the watchers' brains lit up more while seeing abusive treatment of the human than abuse of the robot, suggesting greater empathy for the human.

"We think that, in general, the robot stimuli elicit the same emotional processing as the human stimuli," said lead study author Astrid Rosenthal-von der Pütten of the University of Duisburg Essen, in Germany. Rosenthal-von der Pütten suspects that people still have greater empathy for humans than robots, as evidenced by the stronger effect of watching violence toward the human than the robot.

Still, the study only assessed people's immediate reactions to the emotional cues, Rosenthal-von der Pütten said. "We don't know what happens after the short term," she said.

Human-robot interactions
That humans would show empathy for the robot is not surprising, because the bot looked and behaved like an animal, roboticist Alexander Reben, founder of the company BlabDroid, LLC and a research affiliate at MIT, told LiveScience. Reben, who was not involved in the recent study, himself builds small cardboard robots that tap into the human affinity for cute creatures.

Some people find the idea of humans empathizing with robots concerning. But Reben compared trends in robot development with breeding dogs for companionship. "We have been doing this for millennia," he said. "I think we're doing the same thing with robots."

Humans have been known to show empathy for robots in a variety of contexts. For instance, soldiers form bonds with robots on the battlefield. Other research suggests that humans feel more empathy for robots the more realistic they seem, but not if they're too humanlike.

As robots become more and more ubiquitous, understanding human-robot interactions will take on increasing importance, Rosenthal-von der Pütten said.

The new research will be presented in June at the International Communication Association Conference in London.

Follow Tanya Lewis on Twitter and Google+. Follow us @livescience, Facebook and Google+. Original article on LiveScience.com.

• Gallery: The BioDigital Human
• Biomimicry: 7 Clever Technologies Inspired by Nature
• Humans Show Empathy For Robots | Video

Copyright 2013 LiveScience, a TechMediaNetwork company. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.

James Lever, U.S. Army's Cold Regions Research and Engineering Laboratory

The Yeti rover in Greeland in April 2012, with its ground-penetrating radar sled.

By Becky Oskin, OurAmazingPlanet

Meet Yeti, a faithful rover of the robotic kind that sniffs out dangerous crevasses for convoys crossing the glaciers of Antarctica and Greenland, explores ice caves on an active volcano and finds old buildings buried under the polar ice.

This southern summer, Yeti trundled ahead of the bright orange tractors that deliver supplies to research outposts throughout Antarctica. Towing a ground-penetrating radar, the rover alerts drivers to hidden breaks in the ice. Snow often bridges the fissures, obscuring hazards, but the little robot is light enough (150 pounds, or 68 kilograms) to ford the breaks if not stopped in time.

"It's not likely to fall through, and there's no danger to the robot except losing the robot," said Laura Ray, an engineer at Dartmouth College in Hanover, N.H., who led the robot's development. [Watch Yeti guide a tractor in Greenland]

Preventing injuries and equipment damage is crucial in Antarctica, where rescue operations are dangerous for both the injured and the rescue crews.

But Yeti offers more than protection for polar researchers. In December 2012, scientists mapped ice caves on Mount Erebus, an active volcano, with the radar. The robot also helped find abandoned, long-buried buildings from the original South Pole research station in December 2011. The structures were demolished after a tractor fell through ice atop one of the buildings. The robot also aids convoys in Greenland.

Yeti isn't the first robot to ply the Antarctica ice, but it is the smallest now roving on the continent's frozen surface. Researchers are also peering at buried glacial lakes with micro-submarines or mapping the seafloor with remotely operated vehicles. Earlier robotic explorers include the spiderlike Dante, sent into Mount Erebus in 1992; the Nomad rover, which hunted for meteorites in 2000; and a robotic rover tested by Kansas University in 2005.

Ray thinks Yeti or similar robots can further complement polar research, for example, by finding buried streams and rivers. A robot will handily finish a survey of gridded points that would be tedious for people, she said.

"I do hope there's more of this because it's so costly to do science in these regions. Whatever you can do to reduce the cost enables you to do more science," Ray told OurAmazingPlanet.

Logistical costs — moving people — account for about 70 to 75 percent of most research budgets in Antarctica, Ray said. A National Science Foundation (NSF) report put the total cost of transportation and support even higher, at 90 percent of the polar science research budget. With the median NSF research grant at $125,000 in July 2012, any cost-savings in logistics frees up money for science. The Yeti-led supply convoys across the ice to McMurdo Station in Antarctica save $2 million annually over plane trips, the NSF estimates.

The Yeti robot, conceived and built by Dartmouth engineering students, cost about $25,000, Ray said. The Mars Curiosity Rover cost $2.5 billion to design, construct and land on another planet.

Constructed from off-the-shelf parts and batteries sturdy enough for very low temperatures, the rover has been a popular addition to the United States polar science program.

"It's always in the field or in transit," Ray said. "I would love to make a second or a third robot, or transfer this technology to a company so we can get them into more hands."

Email Becky Oskin or follow her @beckyoskin. Follow us @OAPlanet, Facebook or Google+. Original article on LiveScience's OurAmazingPlanet .

• Extreme Living: Scientists at the End of the Earth
• Infographic: Antarctica – 100 Years of Exploration
• The Harshest Environments on Earth

Copyright 2013 LiveScience, a TechMediaNetwork company. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.

If your job involves tasks such as sorting springs and screws or unloading dishwashers, a robot replacement may soon be on the way.

For now, the granular-gripper robot demonstrated in the video above is perhaps best suited as a sidekick for bar games you might play while trying to grab the attention of a potential flesh-and-bone soul mate.

That is, assuming the potential mate doesn't fall for the robot instead. After all, its barroom athleticism is tough to beat — able to sink mini-basketball shots with uncanny accuracy and hit the bull's eye on the dartboard time and time again.

The tossing ability of jamming robot gripper is a new trick from roboticists working on the grasping technology at Cornell University and the University of Chicago. 

The gripper itself is essentially a balloon filled with granular material, in this case coffee grounds. This squishy balloon hand conforms to whatever object it touches. When the air is sucked out of the balloon, a tight grip is created. To toss the object, the gripper is rapidly re-inflated with air. 

While this all seems simple, anyone who's tried to consistently sink baskets on the court or in a bar knows that picking up balls and tossing them repeatedly through the hoop isn't nearly as easy as it seems.

From the roboticists' perspective, the technology is an improvement over other throwing robots.

"Certainly throwing has been demonstrated with robot arms before, but the momentum for throwing is typically provided by the arm motion while the gripper simply releases the object at the optimum time," the researchers write in a FAQ accompanying their paper to appear in IEEE Transactions on Robotics.

"Here, the entirety of the shooting function is provided by the gripper."

While the shooting skill of the robot isn't good enough for it to go to work tossing together electronic components, which requires higher precision, it is good enough to pick up trash after a good house party.

Other potential applications, the team notes, include picking up and quickly disposing of improvised explosive devices (IEDs). After all, the research is supported by DARPA.

— Via IEEE

More on throwing robots and the robotic workforce:

• Robot to throw first pitch at Phillies game
• Robot folds, throws paper plane
• Tosser bot: Dog's best friend?
• Duke grad builds beer-tossing fridge
• More work for robots in China
• Nine jobs that humans may lose to robots
• Underwater robots at work in Japan

Ford.com

A screenshot from a Ford video shows how Active Park Assist works in the Flex model. Drivers just need to target a spot, and the car uses ultrasonic range finders to park itself.

Do you want a future where C-3P0 and his robotic pals do the driving as you text your friends the 411 on the next stop in a crosstown pub crawl? Minds capable of making this dream come true want your opinion.

Students with Stanford University's Center for Automotive Research have prepared on online survey to find out your robotic car desires. After providing a few generic personal details, you can weigh in on questions such as:

• How much control you're willing to give up to an automated self-driving technology? All? Some, like an airplane pilot? None at all?
• Would you take a cab driven by a robot? Choices range from "Definitely not" to "Definitely would. There is no way a computer can drive worse than current human cab drivers :)".
• What are your feelings about a car that could drive you without any input? Choices include: "Excitement – where can I get one," "Party time – I can go out partying without having to worry about drinking and driving," and "Fear – That's it. Run for the hills. The robots are taking over."

To take the survey, click here. When the results are out this spring, we'll share the details.

More on robotic cars:

• Road rage at driverless cars? It's possible
• GM researching driverless cars
• With these autonomous cars, who needs to drive?
• Cars are approaching 'auto' pilot mode
• Audi to climb Pikes Peak without a driver 
• Google tests cars that steer without drivers
• Google self-driving car crash caused by human

John Roach is a contributing writer for msnbc.com. To learn more about him, check out his website. You can also follow him on Twitter.

For more of our Future of Technology series, watch the featured video below.

Mat Evans / University of Sheffield

A Roomba robot outfitted with whiskers and reprogrammed with monkey smarts determines what type of flooring is beneath it.

The next time you find yourself trapped under a pile of rubble, your savior might be a Roomba — souped-up with whiskers and a monkey brain.

Such a robot was recently shown to outperform other whiskered robots in characterizing its environment, using technology that could wend its way into next-generation search and rescue robots, the University of Sheffiled reports.

Researchers have long known that rats sense their environment with whiskers. But models of how their brains interpret these signals vary. 

One approach, for example, has assumed that rats looked at whisker movement patterns and vibrations over a set period of time and then used that information to make a decision.

But various robots created with this model, Science Now explains, correctly guessed the floor beneath them only 50 to 80 percent of the time, after 0.4 seconds of exposure.

Nathan Lepora at the University of Sheffield in England wondered if outfitting these robots with a model of how the monkey brain makes decisions would be an improvement.

Previous research shows that individual neurons in monkey brains ramp up their firing rates when making decisions about the direction of motion for a group of random dots flashing on a screen.

A decision is made when the firing of these neurons cross a certain threshold. If the neurons responding to the up motion cross the threshold first, for example, the monkey would say the dots are moving up.

Lepora and his team fitted a brain model based on this monkey study into an existing Roomba with rat whiskers and found that it nearly flawlessly correctly identified the type of flooring beneath it.

The findings are reported Jan. 25 in the Journal of the Royal Society Interface.

In addition to improved rescue robots, the result suggests that rat brains may function similar to those of monkeys — in fact, they "suggest the possibility of a common account of decision-making across mammalian species," the team conclude.

[Via: Science Now and University of Sheffield]

More on whiskers, rats, monkeys, and brains:

• Virtual whiskers have the touch
• RatCar takes to the robo-road
• 3-D model of rat brain circuit created
• Cat brain inspires computers of the future
• How whiskers help rats find their way

Paul Sakuma / AP

Stanford graduate student Mick Kritayakirana shows the computer system inside a driverless car on the Stanford University campus in Palo Alto, Calif.

The road to a future where we jump in our cars, enter a destination, and let them do the driving could be filled with rage, according to an expert on driverless car technology.

For starters, driverless cars will likely be programmed to obey all traffic laws. They won't speed and will always come to a complete stop at stop signs, for example.

Throw just a few of those law-abiding robots on roads clogged with 250 million human-controlled cars, and there's bound to be some shaken fists, or worse.

"Let's face it, … [we] don’t always follow exactly the traffic rules," Sven Beiker, the executive director of the Center for Automotive Research at Stanford University in California, told me Friday. 

"An autonomous car would probably need to because there's a company putting code into a system and that obviously then becomes a legal action."

20-year vision?
The road rage-at-the-robot scenario came up as we discussed the evolution of driverless car technology and how we might eventually realize the dream of texting while the robot does the driving.

It'll likely remain a dream, Beiker said, for the foreseeable future.

Some experts in the field, he noted, call it a 20-year vision. "Quite frankly, if someone says 20 years, that's basically telling you we don't really know," he said.

But, driver-assisted technologies such as cars that can park themselves, maintain a safe distance from other cars on the road, and have other crash-avoidance technologies are increasingly available on cars today.

All of these technologies, Beiker said, still require drivers to keep their hands on the wheel and their eyes on the road. But those aids are becoming more common, and not just in luxury models.

"These things are definitely happening, and basically you can expect something new every year in that regard," he noted.

Technological, legal, cultural hurdles
When the field will reach the point where we can relinquish control of the car will depend, in part, on further technological developments, a new set of laws — and a cultural shift.

From the technological standpoint, cars can and do drive themselves today (see the Google Street View cars, for example). So, in a sense, we are technologically there.

But a future of roads full of driverless cars would be enhanced by the development and deployment of a wireless communication system that lets the robots anywhere on the road talk to each other.

Such a system, for example, would let cars know if the car in front of it was planning to turn left or right, as well as provide points of traffic congestion that alert robot drivers to alternate routes.

Think of such a system as a radio traffic report on steroids.

Roads full of autonomous vehicles all talking to each other could be much safer than they are today, Beiker noted. After all, human error contributes to 95 percent of all accidents. 

But, "no technology is 100 percent safe," he said.

When a wreck happens, who gets the blame? That's unclear today. Stanford's automotive center has a legal fellow, Bryant Walker Smith, on staff precisely to help answer these types of questions.

It'll probably shake out one of two ways: Either the car owner and/or passenger will be legally responsible just as drivers are today for most accidents, or the manufacturer will be.

But until such laws are written — and there are some are in the works, such as in Nevada where a law has been passed to make driverless cars legal — it's unlikely that autonomous cars will rule the roads.

And then there's the question of how to deploy the robots once we're technologically and legally ready. Perhaps at first autonomous cars will be restricted to one lane of travel on certain roads, such HOV lanes.

"But mixing the conventional vehicle and the autonomous vehicles?" Beiker said. "That's quite a challenge."

More on driverless car technology:

• GM researching driverless cars
• With these autonomous cars, who needs to drive?
• Cars are approaching 'auto' pilot mode
• Audi to climb Pikes Peak without a driver
• Google tests cars that steer without drivers
• Google self-driving car crash caused by human

John Roach is a contributing writer for msnbc.com. To learn more about him, check out his website. For more of our Future of Technology series, watch the featured video below.

Mary Levin / UW Photography

The latest version of the Raven has mechanical wrists that hold tiny pincers. Coming soon is a piece that will allow research groups to attach the same tools used by commercial surgical robots.

Surgical robots named Ravens are flocking to university labs around the U.S. where researchers will be encouraged to hack their software.

This reprogramming could accelerate innovation in surgical robotics, which is stifled due in part to a lock on the market held by the only company with a FDA-approved robot, according to Blake Hannaford, the director of the Biorobotics Laboratory at the University of Washington in Seattle.

That robot, da Vinci from Intuitive Surgical, has successfully performed more than 200,000 procedures — mostly hysterectomies and the removal of prostate glands — in hospitals around the world.

In these types of procedures, surgeons use the robots to make small incisions and wield tiny instruments that are difficult to handle with their own hands. The result is shortened recovery times and less post-operative pain, which increases the demand for robot-assisted surgery.

Academic researchers would like to innovate in this space, but "da Vinci costs $1.8 million and it's a closed system, you're not allowed to program it," Hannaford told me Friday.

This makes sense given that the da Vinci has FDA approval and Intuitive Surgical owns the patents, he added, but until now researchers wanting to experiment had to build their own robots from scratch or come up with enough funds to buy a da Vinci for research purposes.

The Raven program overcomes these hurdles. The robots were purchased with a $1.1 million grant from the National Science Foundation. They are being shipped to five universities with an open-source software license. 

"The [researchers] will modify that software, invent their cool things, and then share them within this community so that we can build off of each other's advances," Hannaford said. 

Similar to da Vinci, each two-armed Raven has mechanical wrists with tiny pincers that can wield surgical tools. A person sitting at a screen can look through Raven's cameras and guide the instruments to do tasks such as suturing.

Unlike da Vinici, the Raven platform lacks FDA approval, which means that it will not be removing a human prostate gland any time soon. 

But innovations created using the platform could be licensed by an existing medical robotics company or used to start a new one, Hannaford noted.

And since some of Intuitive Surgical's key patents are soon to expire, many medical companies that have been sitting on the sidelines "waiting for the right time to jump in" may find the time is ripe to do so, he added.

To see Raven in action, check out the video below:

More stories on surgical robots:

• Robots invade the operating room
• Surgery goes social: Robotic operation to be webcast, tweeted
• Robot folds, throws paper plane
• Robot prostate surgery comes with trade-off

John Roach is a contributing writer for msnbc.com. To learn more about him, check out his website. For more of our Future of Technology series, watch the featured video below.

Hajime Research Institute

Hajime Sakamoto, president of Japan's Hajime Research Institute, is soliciting sponsors to help him build a 13-foot-tall humanoid robot.

If all goes reasonably well, the future will be full of friendly humanoid robots that that bake cookies, fold laundry, fetch beers, and clean up our messes. We'll likely think of them as unthreatening helpers. After all, most prototypes are unmistakably robots and aren't quite as tall as a full-grown adult.

That may change if Hajime Sakamoto gets his way. The president of Japan's Hajime Research Institute is soliciting sponsors to help him build a 13-foot-tall humanoid robot. A robot that big around the house might be better suited to make sure you (and your kiddos) do your own household chores.

The robot will be the largest humanoid robot in the world and capable of bipedal walking, the company notes. It will come with a cockpit, presumably so that humans can use it as alternative mode of transportation — or just to scare the daylights out of friends and neighbors.

Bipedalism, that is two-footed, upright walking, is hallmark of primate evolution that separates monkeys from early humans. Doing it is no simple task (just watch a toddler learning to walk). Designing a robot that moves around with human efficiency on two feet has been an ongoing challenge in robotics. 

A breakthrough came in 2005 when a trio of robots that walk in a human-like manner was reported in the journal Science. One was about as tall as an adult woman, according to the researchers, and all three were lauded for shedding light on the biomechanics of human walking.

That is, by building robots that walk like humans, we are forced to learn more about how we actually do the task. This in turn results in more human-like robots and could also lead to more life-like robotic prosthesis.

If the video below showcasing Sakamoto's 7-foot bipedal robot playing soccer is any indication, his quest for a 13 footer, and even taller bots, appears driven by an obsession with the "Mobile Suits" seen in the anime series Gundam. 

— via PopSci

More on walking robots:

• One giant leap for walking robots
• This robot can run — it's the fastest one ever
• Robot walks 40.5 miles non-stop
• Robot suit for rent in Japan to help people walk
• This robot is all downhill