OK, the original press release is at http://www.ox.ac.uk/media/news_stories/2010/101104.html . Note the date: November, 2010. I was wrong that Discovery News did no extra research: they clearly reinterviewed the researcher and obtained quotes that were not in the press release. The release gives size of the sample: 15 undergraduates. They were divided into three groups (presumably of 5 each, although the release didn't say), which got left-to-right current, right-to-left current, or no current. Right-to-left performed better than no current, but again, there is no mention of effect size. Left-to-right performed *worse* than no current. The persistent-effect claim came from reexamining the right-to-left group after six months -- apparently the other groups were not reexamined. The next step, if anybody cares, would be to read the actual paper, which is referenced at the press release. I repeat my prediction that the effect size is tiny. On Wed, May 9, 2012 at 5:46 PM, Allan Wechsler <acwacw@gmail.com> wrote:
There is no mention of effect size. There never is, in this kind of story.
This kind of story starts because institutions, in this case Oxford University, have public relations offices whose only job is to create media buzz about what's happening at the institution. The PR offices are constantly canvassing the academic departments and labs, and looking at publication lists, hoping for something they can spin. Science news organizations like Discovery News are constantly bombarded with press releases from these PR offices. For 99% of science news stories, these releases are the source, and the story merely re-frames the original release, often with no further investigation on the part of the news organization.
I haven't even started to investigate this, but I can predict what I will find out: a tiny effect size, a study that doesn't actually say what the press release wishes it did, a news story that simply parrots the press release, and somewhere in the background, a poor researcher dancing around moaning, "But I didn't mean ...." I'm going to send this message before I investigate.
On Wed, May 9, 2012 at 4:42 PM, ed pegg <ed@mathpuzzle.com> wrote:
http://news.discovery.com/human/brain-electricity-math.html It's barely enough to light a light bulb, but passing a very mild current of electricity through the brain can turn on a metaphorical light bulb in a person's brain. Scientists from the University of Oxford have shown that they can improve a person's math abilities for up to six months. The research could help treat the nearly 20 percent of the population with moderate to severe dyscalculia (math disability), and could probably aid students in other subjects as well. "I am certainly not advising people to go around giving themselves electric shocks," said Roi Cohen Kadosh, a scientist at the University of Oxford and a co-author of a new paper. "But we are extremely excited by the potential of our findings." The UK scientists used a method known as transcranial direct current stimulation, or TDCS. This non-invasive technique involves passing electricity through the skull to increase or decrease the activity of neurons, usually for less than 15 minutes. The amount of electricity is tiny, so small that most patients don't even know it is happening. In fact, many scientists were initially skeptical it would have any effect at all, said Jim Stinear, Director of the Neuralplasticity Laboratory at the Rehabilitation Institute of Chicago. For this experiment the scientists directed the current into the brain's parietal lobe, which is involved in number processing. Instead of learning familiar Arabic numerals, however, the scientists had the participants learn a new series of symbols that represented numbers. Then, while their brains were being stimulated, they tested the participants ability to organize those numbers. Patients who were on TDCS showed an improved ability to order the numbers. The electric current makes it subtly easier or more difficult to stimulate a particular group of nerves, depending on the needs of the researchers and the patient. For example, if researchers want to make it easier for a patient to learn, then the nerves will fire more readily. Other studies have shown that TDCS can improve a variety of brain functions, from pain management to rehabilitation after traumatic events, said Jim Stinear, Director of the Neuralplasticity Laboratory at the Rehabilitation Institute of Chicago. But what is "really remarkable," about this new research is how long the effects lasted: six months. If TDCS can improve number processing in normal people, it should be able to improve number processing in people who have lower than normal number processing skills, and that's who the Oxford scientists will be testing next. TDCS should be able to improve other types of learning, such as language, as long as they are near the surface of the brain. Structures like the hippocampus, which are buried under entire lobes of the brain, are likely beyond the reach of TDCS, said Cohen Kadosh. While the Oxford scientists don't advocate plugging yourself into a wall socket, they do eventually hope to create a device that will provide an appropriate amount of electrical current to the brain, and have filed a patent on such a device. Such a device won't instantly make you better at math, help you recover from a stroke faster, or manage pain better, said Stinear. Anybody using a device will still have to put in a significant amount of effort. Drawing a parallel between a popular stimulant, Stinear said that coffee can help you wake up,but if you just sit on the couch you still aren't being productive. The same goes for TDCS. "Electrical stimulation will most likely not turn you into Albert Einstein," said Kadosh, "but if we're successful it might be able to help some people to cope better with math." _______________________________________________ math-fun mailing list math-fun@mailman.xmission.com http://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun