Muting the Mozart effect

Children get plenty of benefits from music lessons. Learning to play instruments can fuel their creativity, and practicing can teach much-needed focus and discipline. And the payoff, whether in learning a new song or just mastering a chord, often boosts self-esteem.

But Harvard researchers now say that one oft-cited benefit — that studying music improves intelligence — is a myth.

Though it has been embraced by everyone from advocates for arts education to parents hoping to encourage their kids to stick with piano lessons, a pair of studies conducted by Samuel Mehr, a Harvard Graduate School of Education (HGSE) doctoral student working in the lab of Elizabeth Spelke, the Marshall L. Berkman Professor of Psychology, found that music training had no effect on the cognitive abilities of young children. The studies are described in a Dec. 11 paper published in the open-access journal PLoS One.

“More than 80 percent of American adults think that music improves children’s grades or intelligence,” Mehr said. “Even in the scientific community, there’s a general belief that music is important for these extrinsic reasons. But there is very little evidence supporting the idea that music classes enhance children’s cognitive development.”

The notion that music training can make someone smarter, Mehr said, can largely be traced to a single study published in Nature. In it, researchers identified what they called the “Mozart effect.” After listening to music, test subjects performed better on spatial tasks.

Though the study was later debunked, the notion that simply listening to music could make someone smarter became firmly embedded in the public imagination, and spurred a host of follow-up studies, including several that focused on the cognitive benefits of music lessons.

Though dozens of studies have explored whether and how music and cognitive skills might be connected, when Mehr and colleagues reviewed the literature they found only five studies that used randomized trials, the gold standard for determining causal effects of educational interventions on child development. Of the five, only one showed an unambiguously positive effect, and it was so small — just a 2.7 point increase in IQ after a year of music lessons — that it was barely enough to be statistically significant.

“The experimental work on this question is very much in its infancy, but the few published studies on the topic show little evidence for ‘music makes you smarter,’” Mehr said.

To explore the connection between music and cognition, Mehr and his colleagues recruited 29 parents and 4-year-old children from the Cambridge area. After initial vocabulary tests for the children and music aptitude tests for the parents, each was randomly assigned to one of two classes, one that had music training, or another that focused on visual arts.

“We wanted to test the effects of the type of music education that actually happens in the real world, and we wanted to study the effect in young children, so we implemented a parent-child music enrichment program with preschoolers,” Mehr said. “The goal is to encourage musical play between parents and children in a classroom environment, which gives parents a strong repertoire of musical activities they can continue to use at home with their kids.”

Among the key changes Mehr and his colleagues made from earlier studies were controlling for the effect of different teachers — Mehr taught both the music and visual arts classes — and using assessment tools designed to test areas of cognition, vocabulary, mathematics, and two spatial tasks.

“Instead of using something general, like an IQ test, we tested four specific domains of cognition,” Mehr said. “If there really is an effect of music training on children’s cognition, we should be able to better detect it here than in previous studies, because these tests are more sensitive than tests of general intelligence.”

The study’s results, however, showed no evidence for cognitive benefits of music training.

While the groups performed comparably on vocabulary and number-estimation tasks, the assessments showed that children who received music training performed slightly better at one spatial task, while those who received visual arts training performed better at the other.

“Study One was very small. We only had 15 children in the music group, and 14 in the visual arts,” Mehr said. “The effects were tiny, and their statistical significance was marginal at best. So we attempted to replicate the study, something that hasn’t been done in any of the previous work.”

To replicate the effect, Mehr and colleagues designed a second study that recruited 45 parents and children, half of whom received music training, and half of whom received no training.

Just as in the first study, Mehr said, there was no evidence that music training offered any cognitive benefit. Even when the results of both studies were pooled to allow researchers to compare the effect of music training, visual arts training, and no training, there was no sign that any group outperformed the others.

“There were slight differences in performance between the groups, but none were large enough to be statistically significant,” Mehr said. “Even when we used the finest-grained statistical analyses available to us, the effects just weren’t there.”

While the results suggest studying music may not be a shortcut to educational success, Mehr said there is still substantial value in music education.

“There’s a compelling case to be made for teaching music that has nothing to do with extrinsic benefits,” he said. “We don’t teach kids Shakespeare because we think it will help them do better on the SATs. We do it because we believe Shakespeare is important.

“Music is an ancient, uniquely human activity. The oldest flutes that have been dug up are 40,000 years old, and human song long preceded that,” he said. “Every single culture in the world has music, including music for children. Music says something about what it means to be human, and it would be crazy not to teach this to our children.”

No Two People Smell the Same

A difference at the smallest level of DNA — one amino acid on one gene —  can determine whether you find a given smell pleasant. A different amino acid on the same gene in your friend’s body could mean he finds the same odor offensive, according to researchers at Duke University.

There are about 400 genes coding for the receptors in our noses, and according to the 1000 Genomes Project, there are more than 900,000 variations of those genes. These receptors control the sensors that determine how we smell odors. A given odor will activate a suite of receptors in the nose, creating a specific signal for the brain. 

But the receptors don’t work the same for all of us, said Hiroaki Matsunami, Ph.D., associate professor of molecular genetics and microbiology at the Duke University School of Medicine. In fact, when comparing the receptors in any two people, they should be about 30 percent different, said Matsunami, who is also a member of the Neurobiology Graduate Program and the Duke Institute for Brain Sciences. 

"There are many cases when you say you like the way something smells and other people don’t. That’s very common," Matsunami said. But what the researchers found is that no two people smell things the same way. "We found that individuals can be very different at the receptor levels, meaning that when we smell something, the receptors that are activated can be very different (from one person to the next) depending on your genome."

The study didn’t look at the promoter regions of the genes, which are highly variable, or gene copy number variation, which is very high in odor receptors, so the 30 percent figure for the difference between individuals is probably conservative, Matsunami said.

While researchers had earlier identified the genes that encode for odor receptors, it has been a mystery how the receptors are activated, Matsunami said. To determine what turns the receptors on, his team cloned more than 500 receptors each from 20 people that had slight variations of only one or two amino acids and systematically exposed them to odor molecules that might excite the receptors. 

By exposing each receptor to a very small concentration — 1, 10, or 100 micromoles — of 73 odorants, such as vanillin or guaiacol, the group was able to identify 27 receptors that had a significant response to at least one odorant. This finding, published in the December issue of Nature Neuroscience, doubles the number of known odorant-activated receptors, bringing the number to 40.

Matsunami said this research could have a big impact for the flavors, fragrance, and food industries.

"These manufacturers all want to know a rational way to produce new chemicals of interest, whether it’s a new perfume or new-flavored ingredient, and right now there’s no scientific basis for doing that," he said. "To do that, we need to know which receptors are being activated by certain chemicals and the consequences of those activations in terms of how we feel and smell."

Listening to the inner voice

Perhaps the most controversial book ever written in the field of psychology, was Julian Janes’ mid-seventies classic, “The Origin of Consciousness in the Breakdown of the Bicameral Mind.” In it, Jaynes reaches the stunning conclusion that the seemingly all-pervasive and demanding gods of the ancients, were not just whimsical personifications of inanimate objects like the sun or moon, nor anthropomorphizations of the various beasts, real and mythical, but rather the culturally-barren inner voices of bilaterally-symmetric brains not yet fully connected, nor conscious, in the way we are today.

In his view, all people of the day would have “heard voices”, similar to the schizophrenic. They would have been experienced as a hallucinations of sorts, coming from outside themselves as the unignorable voices of gods, rather than as commands originating from the other side of the brain. After a long hiatus, the study the inner voice, and the larger mental baggage that comes along with having one, has returned to the fore. Vaughan Bell, a researcher from King’s College in London, recently published an insightful call to arms in PlOS Biology for psychologists and neurobiologists to create a new understanding of these phenomena.

A coherent inner narrative in synch with our actions, is something most of us take for granted. Yet not everyone can take such possession. The congenitally deaf, for example, may later acquire auditory and communicative function through the use of cochlear implants. However, their inner experiences of sound-powered word, which they acquire through the reattribution of percepts of a previous gestural or visual nature, is something not typically shared or appreciated at the level of the larger public. A similar lack of comprehension at the research community level exists regarding those with physically intact senses, but with some other mental process gone awry. We may note with familiarity the shuffling and muttering of a homeless schizophrenic, yet have no systematic way to comprehend their intuitions, no matter how deluded they may appear.

Bell notes that current neurocognitive theories tend to ignore how those who hear voices first acquire what he describes as “internalized social actors.” In addition to live social interactions, “offline” social interaction with an internal model of those individuals holding significant power in our lives would seem like a handy feature to have. We can readily imagine entirely non-pathological situations where such a model would be of benefit. A young child cut from a school basketball team which they worked hard to make, may be temporality devastated, but hardly traumatized. If they renew their efforts to make the team the next year and practice each day in their backyard, they might imagine the coach who cut them watching their every shot with a critical eye. While this hallucinated guidance would be entirely benign, if the person they imagine is instead an abusive parent or classmate, the internal model might eventually take on a more sinister nature.

It would seem that at least in some individuals, the internal model seems able to get the upper hand, particularly when that hand is forced. We might imagine a school child tasked with the tedium of a seemingly endless recitation—saying the rosary beads, for example, in the catholic school days of yore. The familiar “Hail Mary, full of Grace……” might, after so many repetitions, transform in the mind into something else, despite the earnestness of the professor of faith. “Hail Mary, full of …..” might instead be completed with a different choice word that intrudes from another collective in the brain despite the alarmed child’s efforts to suppress it. In the situation where this is vocalized externally, completely out of control as in full blown Tourette’s syndrome, the child now has a problem.

The idea that separate voices represent separate hemispheres may be a good starting point, but it can readily be dispatched as far as being the whole story. Auditory hallucinations can take the form of multiple social actors, clearly outnumbering our hemispheres, and all with different tones, personalities, and persistence of identity. Attempts have been made to localize brain activity to a particular narrative using EEG recording, or to elicit a hallucination using magnetic stimulation. While the occasional inciteful anecdote may be gleaned from these kinds of investigations, we should not expect much fine detail to ever be had from them. The cortical area known as the temporoparietal junction routinely emerges as a favorite among brain imagers because of its geometric location at the pinnacle of the major fold in the brain. Unfortunately, until there exists a large scale minimally damaging recording technology we are probably going to have to content ourselves with looking closer at what subjects have to say about their own auditory hallucinations, than what their brains might have to say.

As children we learn to talk by talking to ourselves. Unless marooned on an island, we tend to abandon this behavior in polite company for fear of stigmatization, among other things. If the line between normalcy and pathology for hearing voices, or even talking to them, (so long as they do not command undesirable physical actions), is drawn with a greater acceptance for normalcy, a clearer understanding of the inner voice might be sooner in hand.

Multi-dog study points to canine brain’s reward center

After capturing the first brain images of two alert, unrestrained dogs last year, researchers at Emory University have confirmed their methods and results by replicating them in an experiment involving 13 dogs.

The research, published by the Public Library of Science One (PLOS One), showed that most of the dogs had a positive response in the caudate region of the brain when given a hand signal indicating they would receive a food treat, as compared to a different hand signal for “no treat.”

“Our experiment last year was really a proof of concept, demonstrating that dogs could be trained to undergo successful functional Magnetic Resonance Imaging (fMRI),” says the lead researcher Gregory Berns, director of Emory’s Center for Neuropolicy. “Now we’ve shown that the initial study wasn’t a fluke: Canine fMRI is reliable and can be done with minimal stress to the dogs. We have laid the foundation for exploring the neural biology and cognitive processes of man’s best, and oldest, friend.”

Co-authors of the paper include Andrew Brooks, a post-doctoral fellow at the Center for Neuropolicy, and Mark Spivak, a dog trainer and the owner of Comprehensive Pet Therapy.

Both the initial experiment and the more recent one involved training the dogs to acclimatize to an fMRI machine. The task requires dogs to cooperatively enter the small enclosure of the fMRI scanner and remain completely motionless despite the noise and vibration of the machine.

Only those dogs that willingly cooperated were involved in the experiments. The canine subjects were given harmless fMRI brain scans while they watched a human giving hand signals that the dogs had been trained to understand. One signal indicated that the dog would receive a hot dog for a treat. The other hand signal meant that the dog would not receive a hot dog.

The most recent experiment involved the original two dogs, plus 11 additional ones, of varying breeds. Eight out of the 13 showed the positive caudate response for the hand signal indicating they were going to receive a hot dog.

The caudate sits above the brain stem in mammals and has the highest concentration of dopamine receptors, which are implicated in motivation and pleasure, among other neurological processes.

“We know that in humans, the caudate region is associated with decision-making, motivation and processing emotions,” Berns says.

As a point of reference, the researchers compared the results to a similar experiment Berns had led 10 years previously involving humans, in which the subjects pressed a button when a light appeared, to get a squirt of fruit juice.

Eleven of 17 humans involved in that experiment showed a positive response in the caudate region that was similar to the positive response of the dogs. “Our findings suggest that the caudate region of the canine brain behaves similarly to the caudate of the human brain, under similar circumstances,” Berns says.

Six of the dogs involved in the experiment had been specially bred and trained to assist disabled people as companion animals, and two of the dogs (including one of the service dogs) had worked as therapy dogs, used to help alleviate stress in people in hospitals or nursing homes. All of the service/therapy dogs showed a greater level of positive caudate activation for the hot dog signal, compared to the other dogs.

“We don’t know if the service dogs and therapy dogs showed this difference because of genetics, or because of the environment in which they were raised, but we hope to find out in future experiments,” Berns says. “This may be the first hint of how the brains of dogs with different temperaments and personalities differ.”

He adds: “I don’t think it was because they liked hot dogs more. I saw no evidence of that. None of the dogs turned down the hot dogs.”

One limitation of the experiments is the small number of subjects and the selectivity of the dogs involved, since only certain dogs can be trained to do the experiments, Berns says.

“We’re expanding our cohort to include more dogs and more breeds,” Berns says. “As the dogs get more accustomed to the process, we can conduct more complicated experiments.”

Plans call for comparing how the canine brain responds to hand signals coming from the dog’s owner, a stranger and a computer. Another experiment already under way is looking at the neural response of dogs when they are exposed to scents of members of their households, both humans and other dogs, and unfamiliar humans and dogs.

“Ultimately, our goal is to map out canine cognitive processes,” says Berns, who recently published a book entitled “How Dogs Love Us: A Neuroscientist and His Adopted Dog Decode the Canine Brain.”

Even in an increasingly technical era, the role of dogs has not diminished, Berns says. In addition to being popular pets, he notes that dogs are important in the U.S. military, in search-and-rescue missions, as assistants for the disabled and as therapeutic stress relievers for hospital patients and others.

“Dogs have been a part of human society for longer than any other animal,” Berns says. He cites a genetic analysis recently published in Science suggesting that the domestication of dogs goes back 18,000 to 32,000 years, preceding the development of agriculture some 10,000 years ago.

“Most neuroscience studies on animals are conducted to serve as models for human disease and brain functions,” Berns says. “We’re not studying canine cognition to serve as a model for humans, but what we learn about the dog brain may also help us understand more about how our own brains evolved.”