The aim of science is to establish facts, as accurately as possible. It is therefore crucially important to determine whether an observed phenomenon is real, or whether it's the result of pure chance. If you declare that you've discovered something when in fact it's just random, that's called a false discovery or a false positive. And false positives are alarmingly common in some areas of medical science.

In 2005, the epidemiologist John Ioannidis at Stanford caused a storm when he wrote the paper 'Why Most Published Research Findings Are False',focusing on results in certain areas of biomedicine. He's been vindicated by subsequent investigations. For example, a recent article found that repeating 100 different results in experimental psychology confirmed the original conclusions in only 38 per cent of cases. It's probably at least as bad forbrain-imaging studies and cognitive neuroscience. How can this happen?

The problem of how to distinguish a genuine observation from random chance is a very old one. It's been debated for centuries by philosophers and, more fruitfully, by statisticians. It turns on the distinction between induction and deduction. Science is an exercise in inductive reasoning: we are making observations and trying to infer general rules from them. Induction can never be certain. In contrast, deductive reasoning is easier: you deduce what you would expect to observe if some general rule were true and then compare it with what you actually see. The problem is that, for a scientist, deductive arguments don't directly answer the question that you want to ask.

What matters to a scientific observer is how often you'll be wrong if you claim that an effect is real, rather than being merely random. That's a question of induction, so it's hard. In the early 20th century, it became the custom to avoid induction, by changing the question into one that used only deductive reasoning. In the 1920s, the statistician Ronald Fisher did this by advocating tests of statistical significance. These are wholly deductive and so sidestep the philosophical problems of induction.

Some 56 million years ago, carbon surged into Earth's atmosphere, raising temperatures by 5°C to 8°C and causing huge wildlife migrations—a scenario that might mirror the world's future, thanks to global warming. But what triggered this so-called Paleocene-Eocene thermal maximum (PETM) has remained a mystery.

Now, in new work presented on 27 September at the annual meeting of the Geological Society of America and published this week in Science, a group of scientists bolsters its claim that a small comet impact kicked off the PETM, stirring up the carbon just 10 million years after a similar event decimated the dinosaurs. The group announced the discovery of glassy, dark beads, set in eight sediment cores tied to the PETM's start—spheres that are often associated with extraterrestrial strikes.

The critical evidence was hardly the result of a targeted campaign, according to Morgan Schaller, a geochemist at the Rensselaer Polytechnic Institute in Troy, New York, who presented the team's work. The spheres were hiding in plain sight—in sediments off the coast of New Jersey.

Fanuc, maker of the industrial robots used to assemble Apple's iPhone and cars for Volkswagen and Tesla, is now partnering with Nvidia to add the company's graphics processing units to its massive machines.

Fanuc launched an initiative to bring artificial intelligence to its robots after investing $7.3 million in Preferred Networks, a machine learning company, in 2015.

Nvidia's graphics processing units and deep-learning technology will be used to help Fanuc robots recognize, process and respond to the environment around them. It's especially important for reinforcement learning, which is how machines use artificial intelligence to adopt new skills through practice.

A robot may capture video of itself to review how it well it did, then analyze and build on that information as it keeps improving over time. Fanuc's machines will feed what they learn into a neural network that other robots can learn from and contribute to as well, reported MIT Technology Review.

Astronomers have discovered two moons located behind Uranus after re-examining old data collected by NASA's space probe Voyager 2.

During a 1986 flyby, the space probe took a closer look at the planet and its satellites increasing the then-known number of moons around Uranus threefold. Since then, scientists believed there were 27 moons in orbit around the ice giant.

However, two planetary scientists from the University of Idaho's Moscow campus, Rob Chancia and Matthew Hedman, have re-examined Voyager 2's old data and found what they say are two exciting discoveries.

In an unprecedented move, U.S. intelligence agencies are teaming up with the nation's most prestigious scientific body in a bid to make better use of findings from the country's leading social and behavioral scientists.

The partnership between the Office of the Director of National Intelligence (ODNI) in Tysons Corner, Virginia, and the National Academies of Sciences, Engineering, and Medicine aims to build bridges between communities that historically have either ignored one another or butted heads. The effort includes the creation of a permanent Intelligence Community Studies Board at the academies, which will meet for the first time next week, as well as a first-ever study of how social and behavioral science research might strengthen national security.

David Honey, ODNI director of science and technology under Director of National Intelligence James Clapper, says he hopes that the new partnership will help the intelligence community improve how it collects and analyzes information. He and others are eager for help picking out useful and relevant research, as well as grasping where there is a lack of good science. Understanding "the limitations of our knowledge," says Robert Fein, a national security psychologist in Cambridge, Massachusetts, and a member of the new intelligence board, "will help to protect us against armies of snake oil salesmen."

One area in dire need of better research is figuring out when people are lying, Fein says. After the start of the wars in Afghanistan and Iraq, he notes, intelligence agencies poured money into research on both mechanical—think polygraphs—and behavioral—think interrogations—methods of detecting deception. But the results were disappointing, recalls Fein, who led a 2006 report on interrogation techniques for the director of national intelligence. "Researchers overpromised," he says, "and there were few useful results after millions of dollars were spent."

Just as a healthy human baby can grow into a healthy adult, so it is for plants. A variety of nutrients and exposure to different experiences often expose the human being to opportunities of expansion. A new study has unveiled why a field with a variety of plants seems to attract fewer plant-eating insects than farm land with just one type of crop. Scientists and farmers have puzzled over this pattern that makes protecting crops from pests a challenge.

Successful organic fruit-growing starts with selecting varieties that are inherently disease resistant. This important first step eliminates half the problem.

Research published in the current issue of Nature and led by William Wetzel, a new Michigan State University entomologist and the study's lead author, is shedding light on this interaction. Plants suppress their insect enemies by being variable, not just by being low quality on average as is typically thought.

Foxconn has deployed 40,000 robots in its factories in mainland China as it aims to reduce the number of workers at its plants creating digital devices.

Dai Chia-peng, general manager of the automation technology development committee of Foxconn, said during an interview with local Chinese media that those robots are basically made by Foxconn itself, except for some parts like servo motors and reducers that come from other parties. Those robots were deployed to Foxconn's manufacturing base in Zhengzhou, a panel factory in Chengdu, and computer and peripherals factories in Kunshan and Jiashan.

For the first time, scientists have helped a paralyzed man experience the sense of touch in his mind-controlled robotic arm.

For the cutting-edge experiment, a collaboration between the University of Pittsburgh and the University of Pittsburgh Medical Center, electrodes smaller than a grain of sand were implanted in the sensory cortex of the man's brain. The electrodes received signals from a robot arm. When a researcher pressed the fingers of the prosthesis, the man felt the pressure in the fingers of his paralyzed right hand, effectively bypassing his damaged spinal cord

The results of the experiment, which have been repeated over several months with 30-year-old Nathan Copeland, offer a breakthrough in the restoration of a critical function in people with paralyzed limbs: the ability not just to move those limbs, but to feel them.

The experiment with Copeland was a featured stop Thursday when President Obama visited Pittsburgh for a White House Frontiers Conference on advances in science, medicine and technology. The researchers described how neuroscience has been able to build a technology where simply imagining a motion translates into motion, in this case a robotic arm.

Bland food could soon be a thing of the past after scientists invented a revolutionary new device which completely alters our sense of taste and may prove to be a breakthrough in the fight against obesity.

British scientists are on the verge of making broccoli taste like chocolate and tofu taste like steak.

Researchers at London's City University have built a prototype for a device, called the Taste Buddy, which uses low-level electrical current to 'trick' your taste buds.

Astronomers have long pondered how many galaxies might exist in the universe - and the answer appears to be a staggering two trillion. The figure, obtained from Hubble Telescope images, has shocked scientists who estimated the number to be at least 10 times smaller.

The finding wasn't easy to come by. An international team of scientists examined deep space images taken by the Hubble from over the past 20 years, and converted them into 3D pictures. From there, they measured the number of galaxies at different times in the history of the universe, going back more than 13 billion years - near the time of the 'Big Bang.'

The researchers, led by Christopher Conselice of the University of Nottingham, also used new mathematical models which allowed them to infer the existence of galaxies which telescopes cannot observe. The team's findings are detailed in the Astrophysical Journal.

For the number of galaxies we now see and their numbers to add up, the researchers concluded that there must be a further 90 percent of galaxies in the observable universe that are too faint and too far away to be seen.