Mozak nije kao kompjuter,tvrde studije

• 0Niko još nije pohvalio poruku.
  Registruj se da bi pohvalio/la poruku!

As you read this sentence, your brain is processing the letters into words. One popular theory associates this activity with a computer that inputs each bit of data – in this case letters – one after the other.

But a new study finds that language comprehension is not broken up into discrete chunks. Indeed, the brain may work in a more continuous, analog fashion – in which the yes-no, on-off, one-zero precision of the digital computer is only gradually achieved.

Michael Spivey, a psycholinguist from Cornell University, tracked mouse movements on a computer screen of 42 student volunteers. When the students heard a word, such as “candle," they were instructed to click on one of two images that corresponded to the word.

Struggling with ambiguity

When presented with images whose names did not sound alike – for instance, candle and jacket – the subjects moved the mouse in a straight line to the correct image. However, when the images had similar names – like candle and candy – the subjects took longer to click.

"When there was ambiguity, the participants briefly didn't know which picture was correct and so for several dozen milliseconds, they were in multiple states at once," Spivey said.

The evidence for “multiple states" is the fact that the mouse trajectories in the ambiguous cases were no longer straight, but curved.

If the brain worked like a computer, one might expect the students to wait until they had processed the whole word before moving. Or perhaps they would make a preliminary guess towards one image, and then correct themselves and change direction.

But a curved line seems to indicate that the students started moving the mouse after only processing part of the word. And yet they appear to hedge their bet by staying somewhere in between the two guesses.

"The degree of curvature of the trajectory shows how much the other object is competing for their interpretation; the curve shows continuous competition," Spivey said. “[The students] sort of partially heard the word both ways, and their resolution of the ambiguity was gradual rather than discrete."

Shades of gray

Neurons in the brain may still work like electrical circuits or a computer network, but this activity may not correspond to the black and white clarity of a computer. Spivey and his collaborators are advocating a “biological" model of the brain that allows for shades of gray.

"In thinking of cognition as working as a biological organism does,“ Spivey said, “you do not have to be in one state or another like a computer, but can have values in between – you can be partially in one state and another, and then eventually gravitate to a unique interpretation."

This sounds a bit like Schroedinger’s Cat – a paradox from quantum physics in which an unfortunate feline can be both dead and alive. So perhaps a quantum computer – whenever one of those finally gets built – will make a better analogy to the human brain.

http://www.livescience.com/humanbiology/050707_brain_computer.html

• 0Niko još nije pohvalio poruku.
  Registruj se da bi pohvalio/la poruku!

Citat:But a curved line seems to indicate that the students started moving the mouse after only processing part of the word.
Kad ovo citam, pomislim da mozak radi kao pipeline...

Citat:“[The students] sort of partially heard the word both ways, and their resolution of the ambiguity was gradual rather than discrete."

To je logicno, posto nista nije diskretno u prirodi, sve je kontinualno... Kompjuteri ce se razvijati tako da dobijaju vise na kontinualnosti.

• 0Niko još nije pohvalio poruku.
  Registruj se da bi pohvalio/la poruku!

Kompjuteri ce se razvijati koliko im mi dozvolimo,mozda ce "Murov zakon",biti pogresan(mislim da se nece skoro razviti AI).

• 0Niko još nije pohvalio poruku.
  Registruj se da bi pohvalio/la poruku!

Kako mislis "koliko im dozvolimo"?

Pitanje je sta ce biti sa razvitkom kvantnih racunara.

• 0Niko još nije pohvalio poruku.
  Registruj se da bi pohvalio/la poruku!

Pa koliko mi se cini vec se javljaju problemi pri proizvodnji CPU jezgra(velika kolicina tranzistora na jednom mestu=vatra),a do kvantnih racunara ima vremena,prvo se trebaju razviti organski tranzistori,pa tako dalje.

• 0Niko još nije pohvalio poruku.
  Registruj se da bi pohvalio/la poruku!

@ Vladimir

Lako je za tu 'vatru' u procesoru. Vodeno hladjenje...
Nego, silicijum dolazi do kraja svojih mogucnosti (isto kao i hdd).

Bilo je na 2rts sta se zbiva sa licnom koja zatvara el. kolo i pali sijalicu. Sve to radi, ali da bi sto vise zica (tranzistora) stavili na mali (isti) prostor, moramo da ih smanjimo.
Oni (naucnici) su stanjili licnu i kada su pustili struju kroz nju desilo se ocekivano. Licna je pregorela.
Tako bi se isto desilo i sa silicijumskim cipovima.
Nasli su izlaz u nekim tranzistorima sa grafitom

• 0Niko još nije pohvalio poruku.
  Registruj se da bi pohvalio/la poruku!

@Snoop,
Zaboravio si da je IBM vec napravio (jos uvek je u lab), tranzistor velicine 1 molekula. Ima na njihovom sajtu. Malo je problem jos to staviti u masovnu proizvodnju, ali doci ce se i do toga.
Znaci vise se nece meriti u nanometrima, to ce biti ogromna jedinica.

• 0Niko još nije pohvalio poruku.
  Registruj se da bi pohvalio/la poruku!

Researchers at the National Institute of Mental Health (NIMH), part of the National Institutes of Health, have discovered a genetically controlled brain mechanism responsible for social behavior in humans — one of the most important but least understood aspects of human nature. The findings are reported in Nature Neuroscience, published online on July 10, 2005.

Abnormal regulation of the amygdala in participants with Williams Syndrome (right) compared to controls (left). The amygdala activates more for threatening scenes (bottom), but less for threatening faces (top). (Image courtesy of NIH/National Institute of Mental Health)

The study compared the brains of healthy volunteers to those with a genetic abnormality, Williams Syndrome, a rare disorder that causes unique changes in social behavior. This comparison enabled the researchers to both define a brain circuit for social function in the healthy human brain, and identify the specific way in which it was affected by genetic changes in Williams Syndrome.

People with Williams Syndrome, who are missing about 21 genes on chromosome seven, are highly social and empathetic, even in situations that would elicit fear and anxiety in healthy people. They will eagerly, and often impulsively, engage in social interactions, even with strangers. However, they experience increased anxiety that is non-social, such as fear of spiders or heights (phobias) and worry excessively.

For several years, scientists have suspected that abnormal processing in the amygdala, an almond-shaped structure deep in the brain, may be involved in this striking pattern of behavior. The amygdala's response and regulation are thought to be critical to people's social behavior through the monitoring of daily life events such as danger signals. Scientists know from animal studies that damage to the amygdala impairs social functioning.

"Social interactions are central to human experience and well­being, and are adversely affected in psychiatric illness. This may be the first study to identify functional disturbances in a brain pathway associated with abnormal social behavior caused by a genetic disorder," said NIMH Director Thomas R. Insel, M.D.

In this study, investigators used functional brain imaging (fMRI) to study the amygdala and structures linked to it in 13 participants with Williams Syndrome who were selected to have normal intelligence (Williams Syndrome is usually associated with some degree of mental retardation or learning impairment) and compared to healthy controls. Andreas Meyer­Lindenberg, M.D., Ph.D., and Karen Berman, M.D., from the NIMH Genes, Cognition, and Psychosis Program, and colleagues, then showed participants pictures of angry or fearful faces. Such faces are known to be highly socially relevant danger signals that strongly activate the amygdala. The fMRI showed considerably less activation of the amygdala in participants with Williams Syndrome than in the healthy volunteers (see graphic below). These findings suggest that reduced danger signaling by the amygdala in response to social stimuli might be responsible for their fearlessness in social interactions.

Next, researchers showed the study participants pictures of threatening scenes (a burning building or a plane crash) which did not have any people or faces in them and thus had no immediate social component. In remarkable contrast to the response to faces, the amygdala response to threatening scenes was abnormally increased in participants with Williams Syndrome (see graphic below), mirroring their severe non-social anxiety.

"The amygdala response perfectly reflected the unique profile of social and non-social anxiety in Williams Syndrome," said Meyer-Lindenberg. "Because our data showed that the amygdala did still function, although abnormally, in Williams Syndrome, we wondered whether it might be its regulation by other brain regions that was the cause of the amygdala abnormalities."

To investigate this, the scientists looked at the whole brain to identify other regions where reactivity was different between Williams participants and healthy volunteers. They identified three areas of the prefrontal cortex, located in the front part of the brain, that have been implicated in decision-making, representation of social knowledge, and judgment. Those regions are the dorsolateral, the medial, and the orbitofrontal cortex. Specifically, the dorsolateral area is thought to establish and maintain social goals governing an interaction; the medial area has been associated with empathy and regulation of negative emotion; and orbitofrontal region is involved in assigning emotional values to a situation.

The researchers found a delicate network by which these three regions modulate amygdala activity. In Williams Syndrome, this fragile system was significantly abnormal, particularly the orbitofrontal cortex. This area did not activate for either task and was not functionally linked to the amygdala, as it was in healthy controls. Instead, the scientists observed increased activity and linkage in the medial region, which is consistent with the high level of empathy exhibited by people with Williams Syndrome.

"We had previously seen that the orbitofrontal cortex is structurally abnormal in Williams Syndrome, but we didn't know what role it played functionally in the disorder; it is now clear that this area can play a major role in producing social behavioral abnormalities," said Berman. "The over-activity of the medial-prefrontal cortex may be compensatory, but the result is still an abnormal fear response. The medial-prefrontal cortex still works and in fact it is working overtime because it may be the only thing that still regulates the amygdala in Williams Syndrome."

###

In addition to the NIMH Intramural Research Program, the research was also funded by a grant from the National Institute on Neurological Disorders and Stroke (NINDS) to co-author Dr. Carolyn Mervis, University of Louisville.

Also participating in the research were Dr. Ahmad Hariri, Karen Munoz, Dr. Venkata Mattay, NIMH, and Dr. Colleen Morris, University of Nevada.


http://www.sciencedaily.com/releases/2005/07/050710201243.htm

• 0Niko još nije pohvalio poruku.
  Registruj se da bi pohvalio/la poruku!

For more than a century some of the biggest minds in science have debated whether brain size has anything to do with intelligence. A new study suggests it does.

Bigger brains make for smarter people, says Michael McDaniel, an industrial and organizational psychologist at Virginia Commonwealth University.

"For all age and sex groups, it is now very clear that brain volume and intelligence are related," McDaniel said.

Tale of the tape

In the old days, measuring brain size involved either the imprecise method of putting a tape around a person's skull or waiting until they died to examine their noggins. Over the past five years or so, various research groups have studied brain size using new imaging techniques that provide a more accurate gauge.

McDaniel examined 26 mostly recent brain-imaging studies to reach his conclusion.

Still controversial is whether the standard intelligence tests used to measure smarts in the studies are valid. Do IQ tests really reveal intelligence? And how relevant are they to the real world?

"When intelligence is correlated with a biological reality such as brain volume, it becomes harder to argue that human intelligence can't be measured or that the scores do not reflect something meaningful," McDaniel said.

The work was published June 16 in the online version of the journal Intelligence.

A gray matter

The human mind eludes complete understanding, however.

A study last year found that IQ is related to the amount of gray matter in the brain, and that intelligence does not reside in one location but is spread throughout the brain. Importantly, that research discovered that abundant gray matter in certain locations was strongly correlated to IQ.

"This may be why one person is quite good at mathematics and not so good at spelling, and another person, with the same IQ, has the opposite pattern of abilities," said that study's leader, Richard Haier of the University of California at Irvine.

Another investigation led by Haier and reported earlier this year found that men think more with their gray matter and women tend to rely more on white matter, the other primary type of brain tissue.

McDaniel works with employers to screen job applicants and measure their performance. He now thinks intelligence tests are the single best predictor of job performance.

"On average, smarter people learn quicker, make fewer errors, and are more productive," McDaniel said. "The use of intelligence tests in screening job applicants has substantial economic benefits for organizations."

http://www.livescience.com/humanbiology/050620_big_brains.html

• 0Niko još nije pohvalio poruku.
  Registruj se da bi pohvalio/la poruku!

Howard Hughes Medical Institute researchers have discovered a novel way in which the brain size of developing mammals may be regulated. They have identified a signaling pathway that controls the orientation in which dividing neural progenitor cells are cleaved during development.

The image shows neural progenitor cells in the proliferative ventricular zone of the developing mouse neocortex. The progenitor cells divide with the mitotic spindle (alpha-tubulin, green; pericentrin, blue; nuclei, red) perpendicular or parallel to the ventricular (apical) surface (the ventricular surface is pointing downward). The G-beta-gamma subunits of heterotrimeric G-proteins and their activator AGS3 are required for proper apical-basal division of neural progenitors and asymmetric cell fate decisions of their progeny. (Image: Courtesy of Li-Huei Tsai, HHMI at Harvard Medical School.)

The way these cells are sliced during development is critical because at later stages of neurogenesis, vertical cleavage gives rise to two mature neurons that are incapable of further division, while horizontal cleavage yields one neuron and one progenitor cell that can continue to support brain growth.

The researchers speculate that this type of regulatory decision point may play a powerful role in determining the ultimate size of the mammalian brain. Inherited disorders that cause the brain to develop too small or too large may also influence this developmental pathway.

Howard Hughes Medical Institute investigator Li-Huei Tsai and postdoctoral fellow Kamon Sanada, both at Harvard Medical School, published their findings in the July 15, 2005, issue of the journal Cell.

The researchers drew on studies by other researchers that showed that the orientation of cleavage planes in dividing neural progenitor cells in the neocortex determines the fate of the resulting daughter cells. However, nothing was known about the molecular signaling mechanism that regulates the decision to cleave one way or another, said Tsai.

Studies in fruit flies and roundworms had hinted that major regulatory molecules called heterotrimeric G proteins play a role in orienting the mitotic spindles that govern the orientation of cell cleavage during cell division, or mitosis.

"Based on that knowledge, we knew that heterotrimeric G proteins were very good candidates as regulators of the plane of cell division in neural precursors," said Tsai.

"And this is a very important question, because how these cells divide ultimately determines the final number of cells that will be generated during brain development," she said.

Using the embryonic mouse brain as a model, Sanada and Tsai sought to determine whether G proteins play a role in the developing mammalian brain. In their initial experiments, they impaired the function of the Gβγ subunits of heterotrimeric G proteins, in the developing mouse brain. They saw a dramatic interference with orientation of cell cleavage to overproduce "postmitotic" neurons -- cells that could no longer divide -- at the expense of progenitor cells, which could still divide. "This observation led us to want to further test whether impairment of Gβγ has any consequences for cell fate in division, because it has been speculated that the different division planes dictate the ultimate cell fate adoption of daughter cells," said Tsai.

To determine definitively whether impairment of Gβγ had a direct effect on cell fate, the researchers impaired Gβγ signaling in the mouse brains in utero, then isolated the progenitor cells to study the effects in vitro. Those studies revealed that impairing Gβγ did result in overproduction of neurons as a result of both daughter cells adopting the neuronal fate. Thus, the researchers concluded that Gβγ does control the orientation of cleavage and the identity of the daughter cells -- whether they will become neurons or progenitor cells.

Sanada and Tsai also sought to determine how Gβγ is regulated. Studies on asymmetric cell division in fruit flies and roundworms had implicated a particular class of proteins -- known as AGS3 and mPins in mammals -- as upstream regulators of G proteins.

The researchers' analyses showed that AGS3 is, indeed, expressed in progenitor cells. And when they "silenced" AGS3 expression in embryonic mouse brains, the resulting abnormalities mimicked those produced when Gβγ signaling was disrupted.

Now that the role of Gβγ in neural cell proliferation has been discovered, said Tsai, further studies will try to pinpoint how its signaling determines the orientation of spindles during mitosis, and thus the orientation of cell cleavage.

"While this is basic research, we do know that this mechanism is very important in determining the ultimate size of the brain," said Tsai. "And, there are humans born with too few neurons, called microcephaly, or too many neurons, called macrocephaly. I would speculate that many of these cases are the outcome of some sort of impairment in the regulation of cell division, perhaps in the plane of division," said Tsai.

"And with the possibility that neural stem cells may find therapeutic use, the role of G protein signaling in the differentiation of such progenitor cells is going to be a very exciting area to explore," she said.


http://www.sciencedaily.com/releases/2005/07/050715065906.htm