Connectome.
How the brain's wiring makes us who we are.
Amazon review (4 stars out of 5)
of Sebastian Seung book ''Connectome''.
May 1st, 2012 by Simon Laub.
We are all unique. And we all have unique brains.
According to Sebastian Seung:
In our brains, uniqueness resides in the pattern of connections between the brain's neurons.
A pattern, which change slowly over time as we learn and grow. The connectome is the entire collection of
our brain's neuronal connections, the totality of how we are wired together.
In the first chapters I found Sebastian Seungs often simple, chatty, informal style
a bit simplistic and too much in the direction of popular science. But,
the book grew on me as I read on.
Actually, throughout the book Sebastian Seung gives us many brilliant insights.
Complex issues are made understandable by good examples and
Seungs broad knowledge of the field.
Change and the Connectome:
If we are our neural connectome, it then follows
that we can change ourselves by changing the connectome.
|
|
Connectome changes:
All four ''R's'' of the connectome - Reweighting, reconnection, rewiring and regeneration -
are affected by our experiences.
Today, we can change our connectome by training our behaviours and thoughts. But,
eventually, we might be able to change our connectome with molecular
interventions (that promote the four R's of connectome change).
|
Indeed, according to Seung, there is good evidence that what we experience in life change our connectome.
And when people are different this might very well have a lot to do with different
connectomes.
Some differences are so big that they can ''easily'' be measured:
[p. 19] ''Researchers have studied many autistic children and found that their
head and brains are indeed enlarged on average - especially the frontal lobe,
which contains many areas involved in social and linguistic behaviour.''
[p. 111] ''Autism and schizophrenia are caused by some neuropathology, which is caused
by abnormal brain development, which is caused by a combination
of abnormal genetic and environmental influences.''
But what is neuropathology? The theory that makes the most sense to Seung is one
where autism and schizophrenia are connectopathies.
Seung gives us many stunning examples of the brains power to change and heal itself
(especially true for young children):
[p. 27] ''Children with very frequent and debilitating seizures are sometimes
treated by removing one hemisphere of the cerebrum entirely. This is one of the most
radical neurosurgical procedures, and it is astonishingly
that most children recover very well from it!
Some would perhaps argue that recovery after hemispherectomy is
not so surprising. Perhaps it is like losing a kidney?
But, remember
that some of the mental functions are lateralized, so the left and right
side of the brain are not equivalent.
Because the left hemisphere specializes in language, its removal almost
invariably leads to aphasia in adults.
This is not true for children.
Here linguistic functions migrate to the right hemisphere, demonstrating that
cortical areas can indeed change their functions!''
The Kennard principle [1] states that the earlier the brain damage,
the greater the recovery of function. Still, new treatments seems to suggest that the critical period (where recovery can be made)
can often be extended well into adulthood or eliminated altogether [p. 127].
Certainly, there is an amazing neuro plasticity in the brain.
But Seung doesn't really make us much wiser on why some changes are possible,
while others are not.
Perhaps, he seems to be saying, that we now need to know a lot more about e.g. the ''wiring''
of autism and schizophrenia. With this knowledge we would know more about how
it can (or cannot) be changed.
Along the way, Seungs gives us many interesting side stories about change.
I found it especially interesting that most of the (brain) recovery (and change) is done without new neurons, [p. 129]:
''The no new neurons dogma prevails in the neocortex. But neurons
are continually added to two regions of the adult brain, the hippocampus and
the olfactory bulb.
Perhaps the new neurons enhance learning potential!? The hippocampus belongs
to the medial temporal lobe, and some researchers believe that
the hippocampus serve as a gateway to memory. They theorize that
it stores information first and later transfers it to other regions
like the neocortex.
If this is the case, the hippocampus need to extremely plastic, and
new neurons would endow it with extra plasticity.
Similarly, the olfactory bulb might use new neurons to store
smell memories.''
Still, following Seungs presentation, one never doubts that there is a lot more to learn
(about the connectome and how it can be changed) than we know today.
The Importance of Finding Connectomes:
Obviously, to find connectomes, we will have to create whole new machines that produce clear images of neurons and synapses over a large field of view.
Seung is rather optimistic that it can be done: [p. 169] ''If connectomics experiences sustained exponential progress, then
finding entire human connectomes will become easy well
before the end of the twenty-first century.''
But we are certainly not there yet.
According to Seung, we don't have computers or tools powerful enough to analyze a cubic millimeter of a bird brain's connectome, much less a complete human brain a million times that size.
But Seung obviously seems pretty convinced that this is the way forward. And wants us onboard. In his words:
[p. 68] ''The act of discovery in science includes not only the
creation of a new idea, but also the act of pursuading others to accept it.
To receive full credit for a discovery, a person must influence others.''
Noone doubts the importance of connectivity though.
Brain centers certainly needs to be well connected in order to be of any use.
Take the Broca-Wernicke model of language. Here, the Broca area is linked to speech
production, and the Wernicke area is
involved in the understanding of written and spoken language.
Seung writes [p. 180]: ''Wernicke hypothesized a bundle of long axons connecting the areas.
Damage to the bundle, while leaving speech comprehension and production intact, would
make it impossible to repeat words after hearing someone else speak them.''
[p. 180] Wernicke called this disorder conduction aphasia (loss of signal conduction).
And patients with these symptomps were later discovered. Furthermore,
neuroanatomists identified the hypothetical connection between Broca's and Wernicke's regions
as a bundle of axons called the arcuate fasciculus.
// Note:
Probably not the whole story though. According to Wikipedia, the regions the arcuate fasciculus connects to is still in debate.
But these connections are obviously important: In nine out of ten people with tone deafness,
the superior arcuate fasciculus in the right hemisphere could not be detected. //
Connections are very important though:
Damage to a region impairs the corresponding elementary function.
Damage to a connection impairs complex functions requiring cooperation between regions.
Sure, fixing missing or damaged connections is (still) somewhat futuristic.
But, knowing more about the connections is obviously a first step!
Connectopathies:
Seung believes that [p. 217]: ''Certain mental disorders might be caused by connectopathies.
If thats the case, true cures would require establishing normal patterns of connectivity.''
Luckily the brain already has mechanisms for connectome change. Genes that
affect how reweighting, reconnection, rewiring and regenaration can be done. So, in theory,
maybe some drug could turn these genes on or off, and therefore in the
end help change the connectome?
According to Seung: Prozac lifts serotonin levels immidiately, but the
depression is not cured immidiately. According to one line of speculation,
other changes in the brain is what actually helps deal with the depression. Perhaps
an increase in the formation of neurons, synapses and branches in the hippocampus?
In short, a change to the connectome?
Memory (in Connections):
Maybe, Seung is right, and we are our connectome. But everyone will certainly agree that
we are our memories.
But, obviously, even memories could also turn out to be just connections in the brain.
Seung uses the PHCA procedure to explain the difference between short term memory
and long term memory.
[p. 91] In a dramatic medical procedure Profound Hypothermia and Circulatory Arrest (PHCA),
the heart is stopped and the entire body is cooled below 18 degrees Celcius,
slowing life processes to a glacial pace.
PHCA is so risky that it is used only when surgery is required
to correct a life threatening condition. But the success rate
is quite high, and patients usually survive with memories intact,
even though their brains were effectively shut down during the procedure.
The success of PHCA supports a doctrine known as ''dual-trace''
theory of memory. Persistent spiking is the trace of short term memory,
while persistent connections are the trace of long term memory.
To store information for long periods, the brain transfers it from
activity to connections.
The ''dual-trace'' theory explains why long term memory can be retained without neural activity.
During the period between storage and recall, the activity pattern can be latent
in the connections without actually being expressed.
Memory is obviously extremely interesting, and Seung has some brilliant
side stories to tell us more about how memory actually works [p. 187]:
''Henry Gustav Molaison (H.M.) underwent surgical treatment for severe epilepsy at the age of twenty seven.
The medial temporal lobe (MTL) was removed from both sides of H.M.s brain.
Afterwards he appeared normal. Personality, intellect, motor skills were intact.
But, for the rest of his life, he woke up in a hospital room with no idea why he was
there. Apparently, he could not learn new things. But, he could still remember
things from before the surgery.''
Notice, the MTL seems essential for storing new memories, but not for retaining old ones.
And, [p. 181], H.M.s amnesia only applied to declarative memory. He could
still learn new motor skills.
Following Seungs line of thought, knowing more about connections would allow us to
learn more about how memories are actually stored in the brain.
We might not be able to take total recall pills right away. But, certainly,
it would also be nice to know just a little bit more about how memories are actually formed.
As, indeed, memories, are everything.
Howto make Connection Maps:
In the brain, the gray matter is a mixture of all parts of neurons - cell bodies, dendrites, axons
and synapses - while the white matter contains only axons. I.e. The white matter
is all the ''wiring''.
[p. 207] Some axons don't travel very far, reentering the gray matter
close to where they started. But most axons of pyramidal neurons
project to other regions in the cortex, some going as far as the other side
of the brain. Some white-matter axons - a small minority - connect
the cortex with other structures in the brain, such as the cerebellum, the brainstem,
or even the spinal cord. These axons make up less than one-tenth
of the white matter. The cortex is highly self-centered, primarily ''talking''
with itself rather than the outside world.
Tracing the journey of every axon in the white matter seems like an impossible task, but Seung
thinks it could be done by ''slicing and imaging all of the white matter and using computers
to follow the path travelled by each axon in the image. The start and end points
of every path would define a connection between two locations in the cortex.''
Reconstructing one cubic millimeter of gray matter is still beyond the best tools.
But reconstructing hundreds of cubic centimeters of white matter might be possible,
as white matter is visible at a lower resolution.
Seung, optimistically, thinks it will possible to extract the relevant
information from these mountains of data.
What is a healthy wiring, what
is a connectopathy etc.
But, surely having the data will be a frist step.
Howto use data about connectomes:
Aldous Huxley imagined a future dystopia in Brave New World,
where humans are born in factories, have their minds and bodied
transformed by the state, and provided with
mind altering drugs in place of religion.
Indeed, knowledge might not guarantee us a happy future.
Sometimes, the more we know, the more perplexing and difficult the world actually
becomes. Take death. Seung has some rather interesting notes about the brainstem death criteria:
[p. 247] Our level of arousal goes up and down all the time, most dramatically
in the sleep-wake cycle. Several populations of brainstem neurons
neurons, collectively called the reticular activating system,
send their axons widely over the brain. These neurons secrete special neurotransmitters
known as neuromodulators, chemicals that ''wake up'' the thalamus and
cerebral cortex.
Without them the patient cannot be conscious, even if the rest of the brain is intact.
This makes it rather apparent that ''brain death'' can mean all sorts of
things. What if the brainstem is destroyed, while the rest of the brain is left intact?
[p. 247] The patient will never breathe again without a mechanical ventilator,
and will never regain consciousness. Yet, one could argue that the patient
still lives, assuming that memories, personality, and intelligence are
preserved in the cerebrum.
These properties seem more fundamental to personal identity than respiration,
circulation or brainstem function.
So, new knowledge leads to new ways of dealing with the world.
Seung certainly thinks we should have a new definition of death,
A more fundamental definition (than the brainstem
criteria).
A definition which would be true even when medicine is much more advanced in the future.
According to Seung, a good definition would be: ''Death is the destruction of
the connectome.''
Seung is always careful to note that ''we don't know yet whether a connectome
actually contains a person's memories, personality or intellect. Testing these ideas
will occupy neuroscientists for a very long time.''
Still, probably, more new knowledge about the connectome will eventually completely
change how we think about ourselves and how we should deal with the world.
Brain Simulations:
Seung also have some good comments about brain simulations.
I.e. finding complete connectomes might not be enough to give us
a full picture of what might be going on inside the brain.
We might also
need to run computer simulations, in order to figure out precisely how the brain might actually be working.
[p. 257] ''Simulation starts from the wish to reproduce an interesting phenomenon,
and tries to find the data necessary to do it. In the past, all kinds of
assumptions were put into the models, that were not backed up by empirical data.
But with better data (connectomics and other better measurements from real brains)
brain models will improve.''
Good models of the various neuron types in the brain are obviously essential
for running a good brain simulation.
[p. 267] ''You are your connectome plus models of neurons type (how the various types work).''
[p. 268] ''The parts list for the human brain is longer, so it will
take many years to model every neuron type in the human brain. But the
parts list is still far shorter than the total number of parts.
Thats why the organization of the parts (connectome) is so important.''
Seung is rather critical of Henry Markrams Blue Brain simulations:
[p.266] Markrams laboratory has characterized the electrical
properties of many neocortical neuron types through experiments in vitro.
Based on this data, they have modelled each neuron type as hundreds of interacting
electrical ''compartments'', which is an approximation to simulating the
millions of ion channels in the neuron. Markram deserves credit for the
realism of the multicompartmental model neurons used
in Blue Brain.
But Blue Brain is severely lacking in one respect. Since no cortical connectome
is known yet, it's not clear how to connect the model neurons with each other.
Markram assumes that conectivity is random - I.e. accidental collisions leads to contact points,
where a synapse occurs with some probability.
If the neural connectivity of Blue Brain is wrong (violations of random connectivity has
already been discovered), the simulation will be too.
But, obviously, Markram could always incorporate information from connectomes into Blue Brain
(later on).
And we certainly won't get the right results, if important operations are completely missing in the
simulations [p. 268]: ''Markram and Modha have included reweighting using mathematical models
of Hebbian synaptic plasticity. But it is also important to include
reconnection, rewiring, and regeneraion.''
(More) Problems for Brain Simulations:
And it gets worse:
[p. 268] ''One difficulty is that neurons can interact outside the confines of synapses.
For example, neurotransmitter molecules might escape from one synapse,
and diffuse away to be sensed by another more distant neuron.
The neurons might not even be in contact with each other.
Extrasynaptic interactions is not encompassed in the connectome.''
[p. 268] If extrasynaptic interactions turn out to be critical for brain function,
then it might be necessary to reject the hypothesis ''You are your
connectome''. The weaker statement: ''You are your brain'' might still
be true though.
A brain computer simulation would then have to simulate every atom in the brain
(or something like that). Faithful to reality, but not exactly easy.
Limited computer power is one problem. But then there is also the problem of
obtaining the information to initialize the simulation. Would it be necessary
to know the (start) position and velocities of all atoms in the brain (which is impossible according to quantum physics) ?
Brain Simulations and Uploading:
In the final chapters Seung manages to sneak in some comments about
running complete brains as computer simulations.
I.e. would it be possible to extract the connectome from a
real brain and then run a simulation of it on a computer?
Seung actually seems kind of sympathetic to the idea behind uploading the connectome
to a computer.
I.e. surely, noone likes death. Staying away from that makes sense.
And the uploading idea could be rephrased:
All efforts to achieve immortality can be viewed as attempts to
preserve information. So, uploading and cryonics is actually just about preserving the information in brains.
Makes sense.
The practicality of it is something else entirely though. Here Seung is rather harsh:
Obviously, brain modelling faces many problems (to name but a few): a) Insufficient neural
modelling b) Extrasynaptic Interactions c) Insufficient knowledge of
the laws of nature.
Connecting the brain model to the real world might really be just a small problem compared
to these other problems:
The connections between the brain and the external world are far less numerous than
the connections within the brain. The optic nerve, which connects the
eye to the brain, carries input through its million axons. That may sound like
a lot, but there are many more axons running within the brain
(most of the brains 100 billion neurons have axons). On the output side,
the pyramidal tract
carries signals from the motor cortex to the spinal cord, so that the
brain can control the movement of the body. Like the optic nerve, the
pyramidal tract contains a million axons.
Seung even has some ideas of what a self (involved in a upload) would be thinking:
''Your self-model would presumably be uploaded along with all of your other memories.
You would be able to check the fidelity of you simulation by continously
comparing your behaviour with the predictions of your self model.
The more accurate the simulation, the fewer the inconsistencies.''
But, according to Seung, not something we are going to see anytime soon....
Conclusion:
Its not all science fiction though. And knowing more about the connectome
will eventually put us on a road to very concrete results:
Is it indeed true that minds differ because connectomes differ?
If we succeed in answering that question, we will also be able
to identify desirable changes in the brain's wiring.
Where the next step will be to devise new methods of promoting such changes.
According to Seung: You might be sceptical that technology will never progress enough to find connectomes
quickly and cheaply.
But it is a well defined goal - so time, money and effort are likely to yield progress.
To understand how the brain work is a much broader goal, where it is
not very precisely defined what success actually means.
Reading Seungs book nevertheless leaves one with the feeling of being on the right track!
Amazon: [2],
[3].
-Simon
Simon Laub
www.simonlaub.net
Who is in Charge?
Amazon review (4 stars out of 5)
of Michael S. Gazzanigas book ''Who's in Charge - Free Will and the Science of the Brain''.
April 22nd, 2012 by Simon Laub.
Free will is a tricky subject.
But, in the book ''Who is in Charge'', Michael Gazzaniga has some good insights.
Gazzaniga first introduces us to some of the nuts and bolts of the physical brain. How things are wired
and what that might mean.
From thereon the book moves on to consciousness, and how consciousness might emerge from the
physical brain.
Eventually, free will and determinism are discussed.
Obviously, a lot of the issues in this book are only
superficially touched upon. And, obviously, it would have been nice with a more thorough discussion
about these super interesting issues.
But, in a relatively popular book, it is
probably not possible to given more details and be more thorough.
Still, the book is a nice read. And it does give some nice insights on what brains, consciousness and free will might be all about.
The physical brain:
It all starts with the brain:
The current view of the brain is that its largescale plan is genetic,
but specific connections at the local level are activity dependent and a function
of epigenetic factors and experience.
Both nature and nurture are important.
Training can indeed change the brain. E.g.:
Synaptic connections in a living mouse rapidly respond to motor skill training and permanently rewire.
Training a month-old mouse to reach with its forelimb caused rapid (within an hour!) formation
of new dentritic spines.
And wiring within the brain is obviously important [p.36]:
A new imaging technique, diffusion tensor imaging,
can actually map nerve fibers. The way that the human brain is organized on the local level is now obtainable, seeable, detectable
and quantifiable.
And now it is possible to see at least a little bit of what is going on. Different wiring, different functionality:
The white matter fiber tract that in humans is involved with language, the arcuate fasciculus, has a completely
different organization in the monkey, the chimp and the human.
Indeed, probably wiring is more important than just the size of the brain [p. 31]:
In split brain surgery, the large tract of nerves that connects the two hemispheres, the corpus callosum,
is severed to prevent the spread of electric impulses, seizures.
In these cases, the isolated left brain, however, which received no input from the right hemisphere (in essense
losing half its size), remains just as intelligen as intelligent as the whole brain.
If brain quantity is so important, you would think that there would be an effect on problem
solving and hypothesizing when half the brain is lost,
but there is not.
Gazzaniga has other interesting examples of different brain wiring [p. 197].
In an experiment, people are asked to name an upside down object.
This involves two processes. One right hemisphere process, which rotates an object in space,
and one left hemisphere process, which names an object:
I show you an up-side-down boat. Before you can name it, you first rotate
it right-side-up in your right hemisphere. Next you send the rotated image to the
left hemisphere, and the left hemisphere then names the object.
What we noticed is that some people are fast at this, and some are slow at it.
We found that the fast (at naming the object) people, use one part of their corpus callosum to transfer the
information between the brain hemispheres, and the slow people use a totally different
part to get to the their speech center.
Sure, the way in which the brain processes information is dependent on how these
fibers are connected.
Indeed, connections are everywhere [p. 32]:
The human cerebral cortex volume is 2.75 times larger than in chimpanzees, but has
only 1.25 times more neurons.
Which intimates that a good deal of the increased mass is due to the space between
cell bodies and what that space is filled with.
Indeed, the space, known as neuropil, is filled with the stuff of connections: Axons,
dendrites and synapses. In general, the larger the area, the better connected it is.
Consciousness:
And, somehow, from the brain comes consciousness:
Where we can only be conscious about something we can actually process....
I.e. if the brain
can't handle a thing (either because humans generally can't, or because a person has damage to some particular
brain area), we can't be conscious about it [p. 64]:
This has led us to realize that in order to be conscious about a particular
part of space, the part of the cortex that processses that part of space is involved.
If it is not functioning, then that part of space no longer exist for that brain or that person.
Consciousness is obviously only the tip of the cognitive iceberg. Beneath it, a lot of unconscious processing is going on,
which shapes the world we actually see (For visual illusions, try Michael Bachs page).
And there might not be any simple explanation for the construction of consciousness
(For more, see my notes, in danish, about Damasios theories in the HAL article):
The view in neuroscience today is that consciousness does not constitute a
single generalized process.
It is becoming increasingly clear that consciousness involves a multitude
of widely distributed specialized systems and disunited processes, the products of which are integrated
in a dynamic manner by the interpreter module. Consciousness is an emergent property.
From moment to moment, different modules or systems compete for attention and the winner emerges
as the neural system underlying that moment's conscious experience.
Our conscious experience is assembled on the fly, as our brains respond to constantly changing inputs.
Free will:
Still, consciousness is not fully understood, and then, obviously, free will is not fully understood.
The two main positions within that debate are the claim that determinism is false and thus that free will exists (or is at least possible); and hard determinism, the claim that determinism is true and thus that free will does not exist.
See more at Wiki.
The hard determinists will tell us, that:
1. The brain enables the mind and the brain is a physical entity.
2. The physical world is determined. so our brains must also be determined.
3. If our brains are determined, and if the brain is the necessary and sufficient organ that enables the mind, then we are left with the belief that the thoughts that arise from our mind also are determined.
4. Thus, free will is an illusion. And we must revise our thoughts about what it means to be personally responsible for our actions.
Put, differently, the concept of free will has no meaning.
Most people agree that the first claim is ok. Claim 2, however, is under attack.
From the 3 body problem and moving on to non linear complex systems it is seen that complex systems does not allow exact predictions of future states.
Which, obviously, puts claim 3 (and then claim 4) on shaky ground...
There are other problems though.
Can we derive the macro story from the micro story - find a mental state from neural state?
According to Gazzaniga, ''analyzing nerve states may be able to inform us how the thing could work, but not how it actually does.''
It has to do with emergence.
Gazzaniga quotes nobel prize winner Philip W. Andersen:
The ability to reduce everything to simple fundamental laws does not imply the ability to start from these laws and reconstruct the universe.
Emergence is not very much liked. Finally, we have gottan rid of the homunculus inside our brains, and finally, we have gottan rid of Descartes dualism - and other ghosts in the machine - and then people propose, that there are still ghosts in there ...
According to Gazzaniga:
You cannot analyze traffic at the level of the individual car.
You have to throw in location, time, weather, society and other drivers, then a new set of laws emerge, that can predict traffic.
So, once a mental state exist - there is downward causation? But, can a thought constrain the brain that produced it?
Not, really, it is more complex than that.
Take genes. Codons are controlling the construction of the whole (enzymes), but the whole is, in part, controlling the identification of the parts (translation) and the construction itself (protein synthesis). Its not upward or downward. Its complementary.
How do we get from brain state M1 to brain state M2?
M1 is produced by a physical state P1 and M2 is produced by P2.
If we just go from P1 to P2 - then there is no free will, and we are just along for the ride.
The tough question is though, does M1, in a downward constraining process, guide P2, thus affecting M2?
In genetics there is a multiplicity of events going on? And the same might be the case for action? Downward and upward causation working together.
Maybe:
Action is the result of complementary components arising from within and without.
And different levels and in different languages. Some levels completely emergent from the levels below?
Software and hardware:
Gazzaniga doesn't mention it, but I can't help thinking about software and hardware!
I.e. could the
p states be labelled hardware states and the m states software states?
If so, then it would make sense to say that we have software running in our brains
that guides where the p states will go next (''downward causation'').
And, language might be just one of many software packages that we have running inside
our heads, on our hardware of neurons, that controls the flow of action.
Social control:
And it gets ''worse''. Our free will is also constrained and controlled by outside (the brain) forces.
Social context and social constraints on the group level might also influence our actions. In a language that is completely emergent (not producible) from the level of brain hardware.
E.g.[p.157]:
Tomasello think that humans may have undergone a self-domestication
process, where overly aggressive or despotic others were either ostracized or killed by the group.
Thus, the gene pool was modified. Which resulted in the selection of
systems that controlled (that is, inhibited) emotional reactivity such as aggression.
Our imprimers (see my Minsky article) might indeed have set up
goals for us to chase after (with our ''free will''...)
[p. 183]:
The Greeks, more than any other ancient peoples, and in fact more than
most people on the planet today, had a remarkable sense of personal agency.
The sense that they were in charge of their own lives and free to act as they chose.
Happiness for the Greeks consisted in being able to exercise their powers in pursuit
of excellence in a life free from constraints.
Ancient Chinese differed in that their focus was on social obligation and collective
agency. The Chinese counterpart to Greek agency was harmony.
And we are certainly hardwired to a lot of things [p. 209]:
Renee Baillargeon and collegues have been hard at work with a group of toddlers
and have shown that a sense of fairness is present not only in two-and-a-half-year-olds,
but also sixteen-month-olds. The older group when asked to distribute
treats to animated puppets will do so evenly, and the sixteen-month-old infants prefer
cartoon characters that divide prizes equally.
Free will is a tricky thing ....
The action is at the interface of these layers.
It is where downward causation meets upward causation (p and m layers)
in the brain. And it (action) is in the space between brains that interact
with each other.
And we didn't even throw in quantum mechanics and the observer problem to make it tricky.... :-)
Amazon: [4], [5].
Wordpress: [6].
-Simon
Simon Laub
www.simonlaub.net
Stress and the Prefrontal Cortex.
Normally, the prefrontal cortex serve as our control center that controls
concentration, planning, decision making, insight, judgement, and the ability to retrieve memories
(For more, see John Duncans How Intelligence Happens).
When things are going well, it easily keeps our baser emotions and inpulses in check.
Not so when we are stressed out.
In an interesting Scientific American article, April 2012 - Amy Arnsten, Carolyn M. Mazure and Rajita Sinha
explains what happens, when untrollable stress sets of a series of chemical events that
weakens the prefrontal cortex control:
And strengthen older parts of the brain. In essence, stress transfers
high-level control of thought and emotion from the prefrontal cortex to the hypothalamus
and other earlier evolved structures.
As the older parts take over, we may find ourselves either consumed by
paralyzing anxiety or else subject to impulses that we usually manage
to keep in check: Indulgences in excesses of food, drink, drugs or spending sprees.
...
The prefrontal cortex is so sensitive to stress, because of its special status within the
hierarchy of brain structures. It is the most highly evolved brain region,
bigger proportionally in humans than in other primates, and makes up a full
third of the human cortex. It matures more slowly than any other brain
area, and reaches full maturity only after the teen years have passed.
The (prefrontal) area houses the neural circuitry for abstract thought and allows
us to concentrate and stay on task, while storing information in mental sketch pad
of working memory.
When we are stressed, the brain stem starts flooding our brains with
arousal chemicals (norepinephrine, dopamine).
In the prefrontal cortex, elevated levels of these chemicals, shuts off neuron
firing....
Network activity diminishes, as does the ability to regulate behaviour.
Other areas of the brain then takes over.
E.g. the basal ganglia,
that regulate cravings and habitual emotional and motor responses, gets a bigger say in what
behaviour we should have. And the stress chemicals also make the Amygdala more
active. Eventually, the Amygdala alerts the rest of the nervous systems to prepare
for danger, and also strengthen memories that are related to fear and similar emotions.
But why weaken the highest cognitive functions when stressed?
According to one theory, it makes sense to freeze and be very scared
when in danger (e.g. standing in front of a large tiger). Not a good time to make elaborate
plans and think rationally!?
Actually, going ''blank'' might be a good thing!?
For more, see my notes on Memories (d).
-Simon
Simon Laub
www.simonlaub.net
Limits of Intelligence.
Why don't we have larger brains? And why don't we think faster?
David Robson answers these two question in a NewScientist article, September 24th 2011.
Basicly, maybe we have reached a point where the advantages of bigger brains
started to be outweighed by the dangers of giving birth to children with big heads?
And, if we want to speed up our thinking with faster neurons (neurons that can fire
more times per second) - that would take more energy. E.g. to support a small increase in the ''clock speed''
of our neurons we would burn energy faster than anyone running a 100 metre sprint, all the time....
Not very practical.
Maybe, rewiring of the brain is the only realistic way forward for improvements
(In the future this could be done articially by laying out new artificial wiring links in the brain....?).
Maybe, nature has already begun!? Indeed, in the last 10 - 15.000 years the human
brain, compared with our body, has shrunk by 3 or 4 percent?
Douglas Fox deals with the same issues in his July 2011, Scientitific American article ''The Limits of Intelligence''. Most tweaks that would
make us smarter would hit limits set by the laws of physics.
Increased brain size, and faster neurons might not be such a good idea (because of the
problems mentioned above). Making wires thinner would hit thermodynamic limitations
similar to those that affect transistors: Communication would get noisy.
The conclusion: Now, higher intelligence comes from technology - Writing, the Internet and other things that enables us
to expand the mind outside the confines of our body.
Specialization of brain areas is also a good ''trick'':
''Researchers have found that as brains gets bigger from species to species,
they are divided into a larger and larger number of distinct areas.
These areas often correspond with specialized functions, say, speech comprehension or
face recognition.
And, as brains gets larger, the specialization unfolds in another dimension: Equivalent areas
in the left and in the right hemisphere take on separate functions - for example, spatial
versus verbal reasoning.
But these areas must still be able to communicate (fast) with each other. Actually,
shorter paths between brain areas correlate with higher IQ:
One study, led in 2009 by Martijn P. van den Heuvel of the University Medical Center Utrecht in the Netherlands,
used functional magnetic resonance imaging to measure how directly
different brain areas talk to each other - that is, whether they talk via
a large or a small number of intermediadiary areas.
Van den Heuvel found that shorter paths between brain areas correlated with higher IQ.
Edward Bullmore has obtained similiar results for working memory.
People with the most direct communication and the fastest neural chatter had the best working
memory (e.g. could hold the most numbers in memory at once).
The rare, nonstop neural connections in the brain have a disproportional influence on smarts.
Cutting just a few of them makes a brain do noticeably worse.
Sure, as brains gets larger, they save space and energy by limiting the number of direct
connections between regions. But some connections should be kept. The right ones ....
In the end, greater individual (brain) smarts might not be so easy to achieve here and now. But then, luckily,
there is technology, culture and computers...
-Simon
Simon Laub
www.simonlaub.net
Rat Cyborgs.
NewScientist article, September 24th 2011.
A very interesting article in NewScientist reports that an artificial
cerebellum has
restored lost brain function in rats.
Apparently, a team under Matti Mintz has
analysed brainstem signals feeding into a real cerebellum, and the output the rat cerebellum generated in response.
One of the functions of the cerebellum is to help coordinate and time movements.
This and the fact that it has a relatively straightforward neuronal architecture,
make it a good region of the (rat) brain to synthesise.
A synthetic chip was then built that could replicate what the real
cerebellum would output on a certain input.
Mintz' team then anaesthetised a rat, and disabled its biological cerebellum.
Finally the team hooked up the rat, using electrodes, to the synthetic
chip.
To test the chip. They then tried to teach the anaesthesised animal
a conditioned motor reflex - a blink - by combining an
auditory tone with a puff of air on the eye, until the animal
blinked on hearing the tone alone.
They first tried this without the chip connected (and without the real
cerebellum), and found the rat was unable to learn the motor reflex.
But, once the artificial cerebellum was connected, the rat behaved as
a normal animal would, learning to connect the sound with the need to blink.
Definitely interesting!
Sure, Mintz' circuitry mimics very basic functionality.
But, it is certainly not difficult to imagine more advanced circuitry.
According to Robert Prueckl: Ultimately, the goal is to build chips that can
replicate complex areas of the brain....
Francesco Sepulveda (University of Essex, UK) takes it one step further:
Well-organised brain parts, usch as the hippocampus or the visual cortex
will have synthetic correlates before the end of the century....
NewScientist concludes the article by stating that ''protheses
might one day be used to enhance brain function in healthy people -
to speed up learning, or enhance memory... I.e. you could speed up learning by adding an artificial learning network in parallel
to the biological one...''
Exciting times ahead for sure!
-Simon
Simon Laub
www.simonlaub.net
