Recognizing voices depends on language ability

Anne Trafton, MIT News Office  Jul 29,2011

Distinguishing between other people's voices may seem like a trivial task. However, if those people are speaking a language you don't understand, it becomes much harder. That's because you rely on individuals' differences in pronunciation to help identify them. If you don't understand the words they are saying, you don't pick up on those differences. Recognizing voices

Read more: Recognizing voices depends on language ability

Idle Minds

by The Gabrieli Lab on Mon, 01/25/2010

Scientific American recently ran the article Idle Minds and What They May Say about Intelligence by John and Sue Gabrieli.

When smarter people's brains are scanned while "at rest," long-distance connections appear stronger

man daydreaming

iStock/Amanda Rohde.

For many years now, neuroscientists have been telling the subjects of experiments something like this: “Please lie in the MRI scanner and relax. When you see the task instructions come onto the screen in front of you, do your best.” The researcher would then use the brain’s activity during the “lie there and relax” period as a mere control condition; the object of scientific interest was always what “lights up” when a subject reads, makes financial decisions or performs some other task.

That has changed. It is now appreciated that the mind never rests.  And that if we measure brain activation while a person lies in a scanner doing nothing, naturally occurring fluctuations will reveal networks that help elucidate the functional organization of the brain in fascinating new ways. Initial studies indicate that these “resting state”networks may help cast light on mental illness. And now, tantalizing new results suggest a significant link between these networks and intelligence.  

Intelligence is a complex and historically controversial topic, in large part because it is difficult to define and to measure.  Psychometricians have developed paper-and-pencil tests of general intelligence that tend to predict performance on a wide range of other tests and a number of life outcomes, like salary.  Neuroscientists have used modern imaging methods to discover the neural correlates of intelligence as measured by these widely used tests.  Many of these studies have examined the relations of IQ to brain anatomy, generally finding that greater grey matter volume or thickness across many brain regions correlates with higher IQ scores. Others have looked at functional measures taken while people perform tasks, generally finding that bilateral frontal and parietal regions are most often associated with performance on intelligence tests. 

But now, for the first time, functional measures of the resting brain are providing new insights into network properties of the brain that are associated with IQ scores. In essence, they suggest that in smart people, distant areas of the brain communicate with each other more robustly than in less smart people.

In a recent paper,
 researchers at the Chinese Academy of Sciences, led by Ming Song, examined how resting brain networks differ between people who have superior versus average IQ scores. They used graph theory to quantify the network properties of the brain, such as how strong the communication is among distant brain regions. A graph is a mathematical representation that is composed of nodes (or brain regions) and connections between them (functional connectivity or temporal correlations), and can be used to characterize neural networks. Like prior researchers, they found that the posterior cingulate cortex is the hub of the human brain – it is the most widely and intensively connected region of the human brain at rest.  Moreover, the strength of connectivity among distant brain regions was greater in people with superior than average IQ scores. Another 2009 study came to a similar conclusion, and noted that the strongest relations between resting connectivity and IQ were observed in the frontal and parietal brain regions, which have been most associated with performance on IQ tests.

Thus, remarkably, the strength of long-distance connections in the resting brain can be related to performance on IQ tests.  We are often impressed when people make creative connections between ideas – perhaps long-range connectivity in the brain empowers such mental range.

These “at rest” findings fit well into what we know of how intelligence develops in children. Previous work discovered that in typical brain development there is a progression from local to distributed network connectivity.  In children, there is strong local connection and weak distant connection. That changes with age:  local connectivity decreases and long-distance connectivity increases.  Intelligence by almost any measure increases with age until young adulthood.  Interestingly, Earlier research also found that slower thinning of the neocortex (often interpreted as pruning of synapses) was associated with higher IQs in children; perhaps the slower pruning allowed for the establishment of long-lasting long-distance connections.  Thus, the strength of long-distance connections in the brain may support the growth of intelligence and influence variation in adult intelligence.

A caveat: The discovery of neurobiological correlates of IQ cannot be construed as evidence that IQ is determined by genes and fixed at birth.  Most research favors the view that IQ scores are influenced both by genes and experience (like almost all human traits).  Further, recent behavioral research has show that intensive training with adults can increase IQ-type scores.

These intriguing findings on intelligence are only the latest to emerge from the new strategy of using the resting mind to help discover the functional organization of the human brain.  Early reports of such resting brain activity around 1995 were viewed skeptically, but many findings since 2000 that relate well to other kinds of neuroscience knowledge have made the study of resting-brain connectivity an exciting research area, in subjects ranging from human infants to primates.  In at-rest studies, children or patients do not have to perform complex tasks (or any tasks). So at-rest scans are proving useful for a wide range of studies examining development and clinical disorders. Indeed, many researchers are now mining old data sets with resting periods from scanning sessions, retroactively transforming a modest control condition into a cutting-edge brain analysis.  Moreover, because resting scans involve no specific task, they are readily pooled across sites into large data bases.

Already, striking differences have been observed in the patterns of resting connectivity found in autism and schizophrenia.  People with autism exhibit reduced connectivity in the brain network most associated with introspection. That lower connectivity may reflect their reduced ability to focus on their own thoughts and feelings, and reduced appreciation of the inner mental worlds of others. In contrast, people with schizophrenia have exaggerated connectivity in this network.  This could reflect problems of self-reference and paranoia in schizophrenia, with an overactive network encouraging patients to interpret events in their environment as having special relevance to themselves. Characteristic changes in the resting brain have been found also in depression and Alzheimer’s disease, raising the possibility that resting-brain scans may help with diagnoses.

Beyond disease, and beyond IQ, these exciting discoveries about the resting human brain raise the question of whether we are gaining the novel capacity to measure quantitatively our most intimate and unique inner selves. Are you most “you” when you’re racing through work? Or when you’re simply sitting in a chair, mind adrift, just being?

What blame can tell us about autism

Anne Trafton, MIT News Office Feb 1,2011

In the mid-1980s, a team of autism researchers theorized that one of the major features of autism is an inability to infer the thoughts of other people. This skill, known as theory of mind, comes naturally to most people — we are constantly evaluating other people’s mental states and trying to determine what they know, what they want and why they are happy or sad, angry or scaredRead the article

Read more: What blame can tell us about autism

Specialization:the connections have it

Saad Jbabdi & Timothy E J Behrens  Feb, 2012

Brain regions exhibit specialization for different functions, but such functions are constrained by anatomical connections to other brain regions. A study now finds that, by measuring these connections, we can predict complex functional responses before the subject has even performed the task.Specialization:the Connections have it

Read more: Specialization:the connections have it

Additional information