Andrew Tarantola — In an obvious attempt to hasten the destruction of mankind, a team of scientists at the University of Texas at Austin built a neural network and promptly induced in the system a state of digital Schizophrenia. The DISCERN neural network was created, in part, to test a hypothesis that schizophrenic symptoms are a byproduct of the sufferer's inability to efficiently ignore stimuli — they can't forget information as easily as the average person. The team mimicked this effect by programming DISCERN to learn natural language — storing information based on the relationship of the words and sentences. As the team increased the system's rate of learning, DISCERN could not properly parse the data, resulting "language abnormalities that suggest schizophrenia," according to one researcher. DISCERN eventually started putting itself at the center of delusional stories, even going so far as to claim responsibility for a terrorist bombing.
It's been far too long since we've been frightened by what are clearly attempts to help SkyNet come to pass. I give you: tiny robotic drones flying in formation!
Japanese university can’t stop building robotic babies
Cyriaque Lamar — Researchers at Osaka University previously brought us Affetto, the disembodied baby head who was supposed to teach us empathy (or something). Now roboticists at Osaka's Hosoda Lab have built Pneuborn-7II and Pneuborn-13, pint-sized robots that emulate infants' movements.
The robots get their names from their tiny pneumatic muscles — Pneuborn-7II mimics the movement of a seven-month-old, whereas Pneuborn-13 imitates a 13-month-old walking. These bots will debut at the International Conference on Robotics and Automation in Shanghai later this week.
While it's well and good that researchers are applying biomechanical principles to machines, a nagging suspicion tells me this won't end well. It's like we're witnessing the birth of some adorable lil' Cylons.
Paleontologist Jack Horner is like a lot of us: as a child, he wanted a pet dinosaur — a dream he's working to make reality. In a recent TED talk, Horner explained why the Chickensaurus is dinosaur's best hope.
In March, Horner gave a talk in Long Beach, explaining his quest to make dinosaurs from chickens. He starts where you would expect: with Jurassic Park. Paleontologists did undertake a search for dinosaur DNA with the hope of possibly using it to reconstruct a dinosaur in the modern era, but when the search proved less than fruitful, Horner turned his attention to the humble chicken.
He jokes that, while birds have technically been classed as dinosaurs, that hasn't satisfied the sixth graders. "Velociraptors are cool. Chickens aren't cool." But he is trying to make chickens more like dinosaurs through the process of atavistic activation — the "switching on" of earlier evolutionary characteristics. He notes that embryonic pigeons, for example, have an arm formation similar to that of a velociraptor, but adult pigeons do not. He and his fellow researchers and looking for the genes that activate those changes and find ways to inhibit the changes. He envisions a multi-fingered, toothed, and long-tailed chicken he has dubbed the "Chickensaurus" — sure to be every dinosaur-loving child's best friend.
Scientists have been saying for years that chickens and birds are the most direct descendants of dinosaurs and we could do a DNA graft to bring new dinosaurs about.
This isn't a DNA graft, it's using existing DNA within the chicken genome to revert some of their physiology to dinosaur levels. The graphic looks pretty badass, and I could really go for a chickenasaurus familiar right about now.
They should make an anime about how that reporter girl has to find the spheres friends, who are the ancient spheres of a civilization past. Using the ancients ones in the modern BATTLE-MEKU~ not only allows them to fly, but also gives them super powers if they make ecchi poses first!
IN JUST a few weeks single-celled yeast have evolved into a multicellular organism, complete with division of labour between cells. This suggests that the evolutionary leap to multicellularity may be a surprisingly small hurdle.
Multicellularity has evolved at least 20 times since life began, but the last time was about 200 million years ago, leaving few clues to the precise sequence of events. To understand the process better, William Ratcliff and colleagues at the University of Minnesota in St Paul set out to evolve multicellularity in a common unicellular lab organism, brewer's yeast.
Their approach was simple: they grew the yeast in a liquid and once each day gently centrifuged each culture, inoculating the next batch with the yeast that settled out on the bottom of each tube. Just as large sand particles settle faster than tiny silt, groups of cells settle faster than single ones, so the team effectively selected for yeast that clumped together.
Sure enough, within 60 days - about 350 generations - every one of their 10 culture lines had evolved a clumped, "snowflake" form. Crucially, the snowflakes formed not from unrelated cells banding together but from cells that remained connected to one another after division, so that all the cells in a snowflake were genetically identical relatives. This relatedness provides the conditions necessary for individual cells to cooperate for the good of the whole snowflake.
"The key step in the evolution of multicellularity is a shift in the level of selection from unicells to groups. Once that occurs, you can consider the clumps to be primitive multicellular organisms," says Ratcliff.
In some ways, the snowflakes do behave as if they are multicellular. They grow bigger by cell division and when the snowflakes reach a certain size a portion breaks off to form a daughter cell. This "life cycle" is much like the juvenile and adult stages of many multicellular organisms.
After a few hundred further generations of selection, the snowflakes also began to show a rudimentary division of labour. As the snowflakes reach their "adult" size, some cells undergo programmed cell death, providing weak points where daughters can break off. This lets the snowflakes make more offspring while leaving the parent large enough to sink quickly to the base of the tube, ensuring its survival. Snowflake lineages exposed to different evolutionary pressures evolved different levels of cell death. Since it is rarely to the advantage of an individual cell to die, this is a clear case of cooperation for the good of the larger organism. This is a key sign that the snowflakes are evolving as a unit, Ratcliff reported last week at a meeting of the Society for the Study of Evolution in Norman, Oklahoma.
Other researchers familiar with the work were generally enthusiastic. "It really seemed to me to have the elements of the unfolding in real time of a major transition," says Ben Kerr, an evolutionary biologist at the University of Washington in Seattle. "The fact that it happened so quickly was really exciting."
Sceptics, however, point out that many yeast strains naturally form colonies, and that their ancestors were multicellular tens or hundreds of millions of years ago. As a result, they may have retained some evolved mechanisms for cell adhesion and programmed cell death, effectively stacking the deck in favour of Ratcliff's experiment.
"I bet that yeast, having once been multicellular, never lost it completely," says Neil Blackstone, an evolutionary biologist at Northern Illinois University in DeKalb. "I don't think if you took something that had never been multicellular you would get it so quickly."
Even so, much of evolution proceeds by co-opting existing traits for new uses - and that's exactly what Ratcliff's yeast do. "I wouldn't expect these things to all pop up de novo, but for the cell to have many of the elements already present for other reasons," says Kerr.
Ratcliff and his colleagues are planning to address that objection head-on, by doing similar experiments with Chlamydomonas, a single-celled alga that has no multicellular ancestors. They are also continuing their yeast experiments to see whether further division of labour will evolve within the snowflakes. Both approaches offer an unprecedented opportunity to bring experimental rigour to the study of one of the most important leaps in our distant evolutionary past.
Maybe I've been living in Asia for too long, because the first thing I thought when I saw those digital shopping advertisements was 'wow, that is a very empty subway station.' I imagine it's much more clusterfuck-y than shown in the video, but it apparently still works anyway. What a great idea.
BERKELEY — Imagine tapping into the mind of a coma patient, or watching one’s own dream on YouTube. With a cutting-edge blend of brain imaging and computer simulation, scientists at the University of California, Berkeley, are bringing these futuristic scenarios within reach.
Using functional Magnetic Resonance Imaging (fMRI) and computational models, UC Berkeley researchers have succeeded in decoding and reconstructing people’s dynamic visual experiences – in this case, watching Hollywood movie trailers.
As yet, the technology can only reconstruct movie clips people have already viewed. However, the breakthrough paves the way for reproducing the movies inside our heads that no one else sees, such as dreams and memories, according to researchers.
“This is a major leap toward reconstructing internal imagery,” said Professor Jack Gallant, a UC Berkeley neuroscientist and coauthor of the study published online today (Sept. 22) in the journal Current Biology. “We are opening a window into the movies in our minds.”
Eventually, practical applications of the technology could include a better understanding of what goes on in the minds of people who cannot communicate verbally, such as stroke victims, coma patients and people with neurodegenerative diseases.
It may also lay the groundwork for brain-machine interface so that people with cerebral palsy or paralysis, for example, can guide computers with their minds.
However, researchers point out that the technology is decades from allowing users to read others’ thoughts and intentions, as portrayed in such sci-fi classics as “Brainstorm,” in which scientists recorded a person’s sensations so that others could experience them.
Previously, Gallant and fellow researchers recorded brain activity in the visual cortex while a subject viewed black-and-white photographs. They then built a computational model that enabled them to predict with overwhelming accuracy which picture the subject was looking at.
In their latest experiment, researchers say they have solved a much more difficult problem by actually decoding brain signals generated by moving pictures.
“Our natural visual experience is like watching a movie,” said Shinji Nishimoto, lead author of the study and a post-doctoral researcher in Gallant’s lab. “In order for this technology to have wide applicability, we must understand how the brain processes these dynamic visual experiences.”
Nishimoto and two other research team members served as subjects for the experiment, because the procedure requires volunteers to remain still inside the MRI scanner for hours at a time.
They watched two separate sets of Hollywood movie trailers, while fMRI was used to measure blood flow through the visual cortex, the part of the brain that processes visual information. On the computer, the brain was divided into small, three-dimensional cubes known as volumetric pixels, or “voxels.”
“We built a model for each voxel that describes how shape and motion information in the movie is mapped into brain activity,” Nishimoto said.
The brain activity recorded while subjects viewed the first set of clips was fed into a computer program that learned, second by second, to associate visual patterns in the movie with the corresponding brain activity.
Brain activity evoked by the second set of clips was used to test the movie reconstruction algorithm. This was done by feeding 18 million seconds of random YouTube videos into the computer program so that it could predict the brain activity that each film clip would most likely evoke in each subject.
Finally, the 100 clips that the computer program decided were most similar to the clip that the subject had probably seen were merged to produce a blurry yet continuous reconstruction of the original movie.
Reconstructing movies using brain scans has been challenging because the blood flow signals measured using fMRI change much more slowly than the neural signals that encode dynamic information in movies, researchers said. For this reason, most previous attempts to decode brain activity have focused on static images.
“We addressed this problem by developing a two-stage model that separately describes the underlying neural population and blood flow signals,” Nishimoto said.
Ultimately, Nishimoto said, scientists need to understand how the brain processes dynamic visual events that we experience in everyday life.
“We need to know how the brain works in naturalistic conditions,” he said. “For that, we need to first understand how the brain works while we are watching movies.”
Sometimes I feel like I get a little jaded toward amazing scientific discoveries because they resemble things I've read in science fiction. I can safely say that taking video recordings of dreams is one of the most novel scientific concepts I've ever heard of, and I will sign up for such a procedure IMMEDIATELY when the technology is available to the public.
I am really afraid of this technology, because as soon as I'm hooked up to a machine that can take images out of the mind, I'm going to be afraid of them seeing anything embarrassing in my brain, and then I'm just going to start thinking about penises.
My fears exactly. While I would love to be able to review certain dreams, I would not want anyone else to have a direct peek into my subconscious. Especially if I'm unconscious at the time.
I don't think I'd mind as long as I had a chance to review the results first. I'm not sure a mere visualization really could carry the full weight of a dream though. There's audio and a mix of internal thoughts and feelings involved in dreams.
In the technological journey towards artificial intelligence, Israeli researchers have made the next giant leap: the RoboRat.
Matti Mintz of Tel Aviv University in Israel and his fellow scientists have built a rodent-sized artificial cerebellum that when implanted onto the skull of a rat with brain damage, allows him to function normally again.
The cyborg cerebellum consists of a computer chip that is electrically wired into the rat’s brain with electrodes. Since the cerebellum is normally responsible for coordinating movement, this chip was programmed to take in sensory information from the body, interpret it, and communicate messages back out to the brain stem and in turn, the rest of the body.
To test the computer chip brain, scientists conditioned a rat to blink whenever it heard a tone. When the researchers disabled the rat’s cerebellum, however, the rat could no longer coordinate this behavior. Once the artificial cerebellum was hooked up, the rat went back to blinking at the sound of the beep.
“It’s proof of concept that we can record information from the brain, analyze it in a way similar to the biological network, and return it to the brain,” Mintz, who presented the work this month at the Strategies for Engineered Negligible Senescence meeting in Cambridge, UK, told NewScientist.
Though scientists have successfully wired artificial limbs to the brain to restore function, the days of a full-on human cyborg brain implant are far off, researchers say.
Comments
http://io9.com/#!5799943/japanese-university-cant-stop-building-robotic-babies
Scientists have been saying for years that chickens and birds are the most direct descendants of dinosaurs and we could do a DNA graft to bring new dinosaurs about.
Same with wooly mammoths and elephants.
Japanese daytime television has the best fluff pieces. Like this one about a tiny spherical rover that can fly and hover!
Such is the way anime works.
In Biology news:
Lab yeast make evolutionary leap to multicellularity
Sometimes I feel like I get a little jaded toward amazing scientific discoveries because they resemble things I've read in science fiction. I can safely say that taking video recordings of dreams is one of the most novel scientific concepts I've ever heard of, and I will sign up for such a procedure IMMEDIATELY when the technology is available to the public.