Running and neurogenesis; the plastic brain

A new paper online in the Journal of Physiology ("Physical exercise increases adult hippocampal neurogenesis in male rats provided it is aerobic and sustained," Nokia et al.), and described here by the NYT, reports that running is good for the brain.  At least the rat brain.

From the paper (emphasis mine):

Adult hippocampal neurogenesis (AHN) is a continuous process through which cells proliferate in the subgranular zone of the dentate gyrus, mature into granule cells, and ultimately become incorporated into hippocampal neuronal networks. In rodents, adult-born hippocampal neurons seem crucial for a variety of adaptive behaviors such as learning, pattern separation, and responses to stress. Aerobic exercise, e.g. running, increases AHN and improves cognitive performance in both male and female adult rodents. The increase in AHN in response to running is reported to be in part due to an increase in the number of surviving neuronal precursor cells (type 2) rather than to the shortening of the cell cycle. There are also studies indicating that running increases the survival and incorporation of newly divided hippocampal cells, born days before commencing training, to increase net neurogenesis. [See the paper for citations for reported findings, which I've removed here for length.]

It has already been well-established that aerobic exercise is associated with an increase in adult hipocampal neurogenesis, the number of neurons in the hippocampus, the region of the brain associated with producing long-term memory among other functions.  But, Nokia et al. wondered if it was only aerobic exercise, or whether other kinds of exercise have the same effect.

So they compared the number of neuronal cells of mice subjected to high-intensity interval training, resistance training and distance running.  They found no increase in the rats who did resistance training compared to sedentary rats, and a smaller than expected increase in rats that did the interval training.  It was only in the brains of the rats who did aerobic exercise that neurogenesis was significantly increased.  The authors hypothesize that this is because running stimulates the production of  brain-derived neurotrophic factor and insulin-like growth factor, which are associated with neurogenesis. The more aerobic exercise the animal does, the more of these the animal produces, and thus the more neurons.

Currently the best advice for preventing dementia in old age is to maintain a social life, quit smoking, and exercise.  And, if this rat study can be applied to humans, this should at least qualify that as aerobic exercise; running or biking, say.  As with all such lifestyle advice, this surely won't work for everyone, but the evidence is increasingly in its favor, at least on a population basis.

But there are deeper implications of this work, I think.  If exercise changes the architecture of the brain in ways that can affect learning, even in adults, and, as has been repeatedly demonstrated, stimulating children by reading to them, using lots of words, playing music to them, and so on, or the reverse, growing up in poverty,  or with disease, or amid famine all can affect brain architecture and thus cognitive ability for better or for worse, why do so many continue to privilege genes and genes alone -- or even more, a single gene -- for the creation of intelligence?

Source: "Effects on brain development leading to cognitive impairment:  A worldwide epidemic," Olness,
Journal of Developmental & Behavioral Pediatrics:
April 2003 - Volume 24 - Issue 2 - pp 120-130

It seems that the brain responds to experience at all ages, but it's possible that there's a 'sensitive period' for cognition.  As just one example, the cognitive abilities of children reared in institutions in Bucharest were compared to that of children never placed in an institution to those whose lives began there but who moved to foster care before age two.  Those who were reared entirely in institutions had much lower cognitive ability than the other two groups; the cognitive abilities of those who were moved to foster care before age two significantly improved.  The authors of this study suggest that there may be a sensitive period for developing cognitive ability, just as there is one for learning language, and many other aspects of brain function.

Of course, as with any trait, genes play a crucial role in the development of the brain.  But they don't do it alone.  E.g., a 2010 paper in Child Development describes the genetic underpinnings of the developing brain, but its plasticity as well.

The foundations of brain architecture are established early in life through a continuous series of dynamic interactions between genetic influences and environmental conditions and experiences (Friederici, 2006; Grossman, 2003; Hensch, 2005; Horn, 2004; Katz & Shatz, 1996; Majdan & Shatz, 2006; Singer, 1995). There is increasing evidence that environmental factors play a crucial role in coordinating the timing and pattern of gene expression, which in turn determines initial brain architecture. Because specific experiences potentiate or inhibit neural connectivity at key developmental stages, these time points are referred to as sensitive periods (Hess, 1973; Knudsen, 2004). Each one of our perceptual, cognitive, and emotional capabilities is built upon the scaffolding provided by early life experiences. Examples can be found in both the visual and auditory systems, where the foundation for later cognitive architecture is laid down during sensitive periods for basic neural circuitry.  

Genetic determinists might acknowledge the plasticity of the brain but then say that how the brain responds to experience is what's genetically determined, and thus that there are children who just aren't genetically equipped to be the next Einstein, or even to learn calculus.  We know this is true at least at one extreme of the distribution of intelligence, because there are many alleles known to be associated with low cognitive ability.  These usually cause syndromic conditions, however, so aren't related only to how quickly synapses are crossed, or memories made, or whatever it is that underlies -- or defines -- intelligence.  As with many other trait distributions, what happens at the extremes doesn't necessarily represent what's going on in the middle, so I think the jury is still out as to the overriding importance of single or even a small number of alleles in the development of normal or above normal intelligence (again, whatever that is -- for the moment, let's call it the ability to score well on IQ tests).  And indeed no genes with large effects on intelligence have yet been identified, despite decades of looking.  That has so far included comparison of the tails of the distribution among individuals without a clear-cut pathology.

So, of course there are genes involved in how quickly people think, or make connections between ideas, or memorize, or invent things, or remember -- how people learn.  But it's not either mainly genes or environment.  It's both, interacting, and molding the reactive brain.  There is enough evidence now to show that the brain is a hungry organ, soaking up and responding to experience at all times, throughout life.  Whether or not we believe that society should be investing in optimizing the environment of every child to maximize their potential is a social and political decision, not a scientific one.

Brain plasticity -- why should intelligence be an exception?

We live in an age that demands we multitask if we're going to get everything done that we need to do.  Answering email, picking up the children, submitting grants for every deadline, getting in 30 minutes of exercise everyday, eating right, keeping up with the literature -- so much pressure.  Fortunately someone's got our backs, and we can now answer email on one screen at our treadmill desks and work on that grant proposal on another, all while we have lunch.  So much easier, so much time saved.

But wait, psychology tells us that, despite appearances, we can't multitask after all, we can't do two cognitive things at once.  Instead we're 'task-switching', reading then speaking, writing a paragraph then answering the text from the child we forgot to pick up. So tread milling, eating and emailing we can do but tread milling, eating, emailing and writing a methods section we can't.

Unless we're musicians.  A new paper in Cognitive Science ("Musical Training, Bilingualism, and Executive Function: A Closer Look at Task Switching and Dual-Task Performance," Moradzadeh et al.) reports that musicians are better at task-switching and 'dual-tasking' than non-musicians.  Task-switching is just what it sounds like, the ability to switch between tasks, and the speed and ease with which this can be done is what was measured.  Dual-tasking is the ability to do two or more things at once. There must be a reason this isn't just 'multi-tasking' but I don't know what it is.

I'm also not sure what constitutes a 'task'.  Indeed, how many tasks is reading music, with all the separate bits it involves (remembering which key has 5 flats, how long to hold a black flagged note compared with an empty oval, what that marking over the final note means, all the Italian notations telling you how to play the piece, turning notes on a page into a melody, keeping time, etc.) or playing the horn, with all the separate bits that involves (how to blow into the mouthpiece, how to press the keys, which fingering to use for each note and how to do that, how to synchronize your breathing with your fingers to get a note, how to play loudly or softly, playing in tune, all while remembering what each of the conductor's hand movements signifies, and staying in time with the players around you)?  Some of these tasks get relegated to muscle memory after enough practice, certainly, but much of musicianship still involves cognition.

Anyway, the researchers compared the ease with which a group of 153 bilingual and monolingual musicians and non-musicians switched between tasks or accomplished more than one at once.  These were apparently standard psychological tests, with task switching involving tracking numbers on a computer screen, and dual-tasking involving tracking a white dot while looking at flashing letters, while being asked to note when an X appears.

Results demonstrated reduced global and local switch costs in musicians compared with non-musicians, suggesting that musical training can contribute to increased efficiency in the ability to shift flexibly between mental sets.... These findings demonstrate that long-term musical training is associated with improvements in task switching and dual-task performance.

The researchers point out that there can be 'far transfer effects' of training or experience on cognition. This is a well-studied area of psychology, and it's known that many hours of things like physical exercise or video-gaming can affect how we think, or remember, and so forth.  So, that something as complex as musical training might affect other mental skills isn't a surprise.

And, it has long been known that longterm musical training has effects on brain structure, including on sensorimotor and auditory areas, but on grey matter as well (references here).  And, London taxi drivers are known to have larger hippocampi, related to spatial navigation, than London bus drivers who spend as much time driving.  Multilinguists have denser grey matter in brain areas related to language and communication than do monolingual people.  And so on.

This is all overwhelming evidence for brain plasticity.  It beats me why anyone would insist that intelligence is an exception, hard-wired, and not at all contingent upon experience.

Two-eyed cyclops -- the plasticity of the brain

The brain is a remarkable thing.  Part of what's so remarkable about it is how it responds to and molds itself around experience.  Alfred Wallace exempted humans from the march of evolution because we are able to do so many things that can't be attributed to natural selection: calculus, the invention of televisions and robots, smell tar and Twinkies, none of which are abilities that we specifically can thank natural selection for since they are all recent.  We can do them because of our brain's adaptability, its ability to make sense of input it clearly isn't hardwired to understand.

Toy tin robot in the show. Boston MA United States. Picture taken by Jonathan McIntosh, 2003; Wikimedia

I remember lying in bed when I was a child, before I was even in kindergarten, closing one eye and then the other and noticing that I could clearly see the books on the bookshelf across the room with one eye but the same books were a blurry mass with the other.  This was just a fact of life, of idle interest to my 4-year old self but nothing more, and I don't think I ever thought to mention it to anyone. I was fine; I could read up close, I could see in the distance, just not with both eyes at once, so it didn't occur to me that anything was weird or wrong about that.  A routine eye exam at school found me out and I finally got glasses to correct this thing that wasn't really a problem.

As I've gotten older, my vision has gotten worse, each eye in its own way.  My eyes are equidistant from 20:20 in opposite directions, one myopic, the other hyperopic; I can still see without glasses, though not perfectly. And still, without my glasses, it's one eye working at a time.

Vision pathway; Weiss and Buchanan, The Mermaid's Tale, 2009

But think about what that means.  Without my glasses, light is pouring into both eyes, hitting my retina at essentially the focally right place in one eye, but all wrong in the other.  The curious thing, to me, is that my brain long ago learned not to pay attention to the blurry input, to only interpret the light waves hitting my retina in the 'right' place. How did it know which was right?

And at some point, in managing input anywhere along the continuum from my eyes to the furthest point I can see, my brain switches from paying attention to my right eye to paying attention to my left.  All the light waves are getting passed along in the same way to both eyes and on to my visual cortex -- I know this because if I close the 'good' eye, of course I'm seeing something, it's just blurry -- but at the very final step in the vision pathway, when my visual cortex is coordinating all the input into a single image, my brain dumps the blurry images and retains the clear.

But it's even more impressive, I think -- with my glasses on, my brain allows input from both eyes to make its way to the final image.  It's switching from monocular to binocular vision all the time.  Again, how does it know to do that?

My eyes as a metaphor for life
The plasticity of the brain isn't confined to the vision pathway, of course.  Plasticity defines the brain -- it's why we can meet new people, learn things, have new experiences, create memories, and then make sense of it all as we go.  Not only are we constantly making new synapses between neurons, we are still making new neurons well into old age, which is what makes our brains able to successfully make sense of all the information with which we're bombarded all the time.  Adaptability, or facultativeness, is so fundamental to evolutionary success that we think of it as a basic principle of life (see chapter 3, The Mermaid's Tale).

And yes, there's a larger point here.  The idea that some of us evolved 'for' sprinting, ping-pong, money-lending, economic prowess, or the ability to do well in 20th century school systems is based on, we think, a superficial understanding of evolution, and the way the brain works.  But it's an appealing one, one that too many scientists and journalists still believe.

Ideology assumes, science asks.

The Professor of the Environment

In the Paris EvoDevo meeting that we attended a couple of weeks back, one speaker was the developmental biologist Scott Gilbert. Scott spends his time between Swarthmore and the University of Helsinki, and is the author of the world's foremost text on developmental biology but is also author of a long list of books and articles that put the subject in its historical context.

One point he's been making recently, as in his Paris talk, is that there has been too narrow a focus recently on genes as essentially the only cause of development or its evolution -- genes are not all that's inherited in life. Environmental effects of various kinds can be inherited in different ways, too. Some developmental biologists criticize this message for being 'anti-genetic', but Scott does his own genetics work, and certainly doesn't dismiss its importance.  His examples certainly do not undermine genetics, but simply show, convincingly, that other factors contribute to development and evolution.

We pointed this out in our book Mermaid's Tale as well, and in fact plasticity -- varying responses to environmental factors during development -- was a major theme of the EED meetings. Our gut bacteria (E coli) are vital for survival, and newborns are 'infected' from their mothers or their environment. Gilbert provided numerous other examples. Some fly eggs receive bacteria (Wolbachia) that are needed for their proper gene expression and development (as in the image to the left; Credit: PLoS / Scott O'Neill) -- without that, the egg dies. Maternal uterine conditions can cause fetal gene expression that affects the baby for its whole future life, in terms of things like obesity and blood pressure.

In this sense, laboratory organisms may be in such artificial environments that we don't really get a good picture from them, of how things are out there in the real world.

Since these various kinds of commensalism involve the genomes of more than just the species in question (and other non-genetic environmental factors are also transmitted to or needed by an organism's genome), understanding development requires a broader perspective. Information from the environment can be transmitted 'horizontally' in the sense that it is not transmitted 'vertically' from parent to offspring.

As Gilbert cleverly put it, the environment is not just the sieve of natural selection, deciding who shall live and who shall die. It is a source of information to an organism. And, because response to that information potentially affects survival and evolutionary success, the environment is like a professor: It gives information.....and then gives the recipient a test!

And these tests, like university final exams, determine whether the organism shall graduate!

Did my genes make me do it?

David Brooks of the New York Times seems to be heavily into neuroscience these days. Having dismissed evolutionary psychology some time ago, he's apparently not going at this to understand the genetics behind why some people vote Republican and others Democratic (though there are people doing just this), but perhaps instead in the hopes of understanding how people can be taught to vote Republican. He tells in today's column of his recent attendance at a conference on neuroscience, to his surprise a conference of and by the young. He says

Young scholars have been drawn to this field from psychology, economics, political science and beyond in the hopes that by looking into the brain they can help settle some old arguments about how people interact.

These people study the way biology, in the form of genes, influences behavior. But they’re also trying to understand the complementary process of how social behavior changes biology. Matthew Lieberman of U.C.L.A. is doing research into what happens in the brain when people are persuaded by an argument.

Examples he gives of the kinds of studies being done include:

Reem Yahya and a team from the University of Haifa studied Arabs and Jews while showing them images of hands and feet in painful situations. The two cultures perceived pain differently. The Arabs perceived higher levels of pain over all while the Jews were more sensitive to pain suffered by members of a group other than their own.

Mina Cikara of Princeton and others scanned the brains of Yankee and Red Sox fans as they watched baseball highlights. Neither reacted much to an Orioles-Blue Jays game, but when they saw their own team doing well, brain regions called the ventral striatum and nucleus accumbens were activated. This is a look at how tribal dominance struggles get processed inside.

Brooks' point seems to be that people are malleable, and can be socialized to overcome tribalism or anti-social behaviors -- he has been writing about this for years, contending that poor Americans need to emulate the behaviors of middle and upper class families to have any chance of success. Once we understand how the brain works, he suggests, then policy wonks will "see people as they really are".

But how will that help, really? Even if neuroscientists show us that culture is taught, and people are teachable, who gets to decide who's the best role model? David Brooks thinks he knows, but so do Glen Beck and Rachel Maddow--and the rest of us. So, even if we ever do understand how the brain works, the politics won't get any cleaner or easier.

To us, though, the point is deeper than this. What it takes for major league pitchers to learn to throw a baseball so skillfully, and for violinists to learn to play the violin, and for anthropologists to learn to age and sex skeletons is practice--maybe 10,000 hours of practice. We didn't evolve 'to' do calculus, or play the violin, or throw a baseball or to agree or disagree with Rush Limbaugh, but instead to be able to learn how to do these things (although Alfred Wallace, the co-discoverer of evolution, became quite spiritual in later life, certain that humans were above nature because we could do calculus; he couldn't imagine how our ability to do math could have evolved).

The brain is plastic--learning changes synapses. Indeed, physiological changes in the brain from learning new tasks are measurable, as reported in a recent Nature Neuroscience paper on changes in the density of gray and white brain matter in subjects learning to juggle, described here. And, CNN is reporting on their website today about a woman who apparently had a stroke prenatally which destroyed half of her brain; many of the functions stereotypically performed by the missing part of her brain have been taken over by the active side.

It is natural, especially within western culture with its focus on cause and effect, to seek (or hunger for) simple causes for 'effects' we are interested in. We put the word in quotes because from the inquiry point of view what constitutes an 'effect' is often quite subjective. Is the party you vote for a meaningful effect, or is your vote based on deeper issues that, at present, you find affiliated with some particular party?

It is easy to think about finding the gene 'for' some specific effect, especially if you define the effect meaningfully. It's the view that evolution has hard-wired us for the effect, DNA being the prescriptive cause. But that may have things quite backwards. It makes more evolutionary sense that organisms be programmed to be facultative in sensing, assessing, and responding to the environment. If the human brain is anything it is like that rather than hard-wired for voting this way or that. Even ants, as Darwin observed, seem quite intelligent if we shed our anthropocentric biases.

However, it is much, much more difficult to think of understanding the genetic basis of facultative assessment and response, than of hard-wiring. We don't have good ways to define the trait, much less to find its genetic basis. Yet, clearly, that is the trait we should be trying to understand.

Now Hear This (Shape)!

About 1% of the population has synaesthesia, the conflation of several modes of perception into one. Some synasthetes always associate a particular number or letter with a specific color, so 6 is always blue, or B is always lavender. But, a story on the BBC website today suggests that everyone makes similar associations, if to a much more limited extent.

Experimental psychologist Charles Spence and colleagues at Oxford University have found that people generally tend to associate larger shapes with lower pitched sounds and smaller shapes with higher sounds.

It seems our brains may use these synaesthetic associations, says Professor Spence, "to combine all of the different sensory cues that are hitting our receptors at any one time".

Which one of these shapes is 'bouba' and which one is 'kiki'?

Most people label the amoeboid-like shape a bouba, and the star shape a kiki. Further, Spence says that people tend to associate certain sounds with foods.

He said that two of the best examples are brie, which is "very maluma", whereas cranberries are "very takete".

Spence is working with a world-renowned chef to combine name dishes in way that influences their taste buds. But, is there anything to this but amusement?

Synaesthesia can be the stuff of what might be called pathologic brilliance. The well-known autistic mathematical genius Daniel Tammet (his blog is here) says that he sees numbers as shapes and colors. He wrote the books Born on a Blue Day and recently Embracing the Wide Sky. There are other examples. Whatever autism is (and it's probably a single term for a host of different places on the range of neural variation), a kind of remoteness from the everyday world of most people may be associated with aspects of mapping associations in the brain.

But if these are all very interesting facts, but what do they tell us about how our brains work? There is a lot written these days about fMRIs, or functional magnetic resonance imaging, that purports to show which part of the brain is 'for' what function. But, in fact most published fMRI images are composites from multiple people, because the brain isn't as divisible and predictable as it might seem. And, studies of the brains of people who lose a sense, such as vision, show that the part of the brain that was once 'for' vision can be remapped to touch or hearing.

From an evolutionary genetic point of view, this work seems to point to the idea that mammal brains evolved to localize types of function in ways that sequester them from each other--perhaps to enable easier recall, the way books in a library are catalogued by category numbers for easier shelving.

But memories and thoughts cannot be wholly sequestered, or animals could not integrate information to make holistic sense of their environment--which is vital to find food, detect predators, recognize mates and kin, and so on. The partial nature of sequestration is a fundamental aspect of life--all life--and is one of the generalizations about life that we explore in our book (and this blog).

Clearly, cultural experience associates different kinds of information. In our culture, we associate low sounds with large size, because many things that are large, like empty barrels, tubas, and so on make large sounds, while small birds & piccolos make high ones. But the fact that we make such associations, and ones like the shape-test above, does not imply that we're hard-wired to make them.

Too much genetic hardwiring would potentially cripple an organism to conditions its ancestors dealt with and that selected the lucky hard-wired ones for reproductive success. But it could be a disaster if environments change. Clearly, we think, it is functionally better to be soft-wired: genetically enabled to make associations by experience, file them, and recall them systematically as needed--this kind of adaptability is another generality about life that we explore in our book. Whether it's harder to understand how genes could evolve to be soft-wired rather than hard-wired is debatable, but most research in evolutionary biology is about what's hard-wired and takes a hard-wiring perspective.

Synaesthesia shows that the wires can become crossed in some people (at least sometimes for reasons having to do with genetic variation). But, in general, mammals are master electricians who install the wiring in each case where it best needs to be.