Brain Network Dynamics: Brain as Anti-Algorithm

Cognitive scientists are uncovering more secrets of the brain every day. One fascinating line of brain research involves how the brain forms categories and metaphors.

At the IMP in Vienna, neurobiologist Simon Rumpel and his post-doc Brice Bathellier have been able to show that certain properties of neuronal networks in the brain are responsible for the formation of categories. In experiments with mice, the researchers produced an array of sounds and monitored the activity of nerve cell-clusters in the auditory cortex. They found that groups of 50 to 100 neurons displayed only a limited number of different activity-patterns in response to the different sounds.

The scientists then selected two basis sounds that produced different response patterns and constructed linear mixtures from them. When the mixture ratio was varied continuously, the answer was not a continuous change in the activity patters of the nerve cells, but rather an abrupt transition. Such dynamic behavior is reminiscent of the behavior of artificial attractor-networks that have been suggested by computer scientists as a solution to the categorization problem. _SD

Here is the study abstract from Neuron:

The ability to group stimuli into perceptual categories is essential for efficient interaction with the environment. Discrete dynamics that emerge in brain networks are believed to be the neuronal correlate of category formation. Observations of such dynamics have recently been made; however, it is still unresolved if they actually match perceptual categories. Using in vivo two-photon calcium imaging in the auditory cortex of mice, we show that local network activity evoked by sounds is constrained to few response modes. Transitions between response modes are characterized by an abrupt switch, indicating attractor-like, discrete dynamics. Moreover, we show that local cortical responses quantitatively predict discrimination performance and spontaneous categorization of sounds in behaving mice. Our results therefore demonstrate that local nonlinear dynamics in the auditory cortex generate spontaneous sound categories which can be selected for behavioral or perceptual decisions. _Neuron Article Abstract

Here is a broader look at brain network dynamics in the context of decision making:

Cortical network dynamics of perceptual decision-making in the human brain

Brain cells work together in groups, in a dynamic fashion.

Spontaneous rhythmical activity occurs in groups of neurons -- whether artificially cultured in the lab, or in self-selected groups within a living brain.

When separated groups of neurons communicate with each other over a distance in the brain, they utilise a method of synchronous oscillations -- a language that scientists have just begun to understand.

Billions of dollars are spent every year on the quest to achieve human level artificial intelligence. Most of this research is based upon algorithmic design, utilising digital computers. But as anyone can see from looking over recent findings in the neuroscience of cognition, the brain is more of an anti-algorithm. The logic of brain network dynamics has almost nothing in common, conceptually, with the algorithmic basis of digital computing.

AI researchers have attempted to narrow the conceptual gap by utilising "neural net computing," "fuzzy logic computing," and "genetic algorithmic computing," to name three alternative approaches. And these alternative approaches are likely to be very helpful in both applied and theoretical computing and information science. But do they get AI researchers closer to the goal of human-level machine intelligence?

Probably not. Not even the startling potential of memristors and similar semiconductor devices are likely to close that gap appreciably.

As discussed recently in an article quoting quantum physicist David Deutsch, artificial intelligence research is desperately in need of better supporting philosophical structures.

Until then, it is likely that artificial intelligence research will continue to spin its wheels pursuing better algorithms to emulate the brain, without a good understanding of what the brain does.

It is possible to emulate the human brain, using an approach that depends to a limited extent upon algorithmic control, in conjunction with other conceptual methods. But not before researchers learn to approach the problem in entirely new ways, on new logical levels..

Introduction to brain oscillations video

Rhythmic Learning: Our Oscillating Reality

Researchers at the Max Planck Institute of Psychiatry in Munich, have discovered an important mechanism for long term learning in the hippocampus of the mouse brain.

Max Planck Institute

The hippocampus represents an important brain structure for learning. Scientists at the Max Planck Institute of Psychiatry in Munich discovered how it filters electrical neuronal signals through an input and output control, thus regulating learning and memory processes. Accordingly, effective signal transmission needs so-called theta-frequency impulses of the cerebral cortex. With a frequency of three to eight hertz, these impulses generate waves of electrical activity that propagate through the hippocampus. Impulses of a different frequency evoke no transmission, or only a much weaker one. Moreover, signal transmission in other areas of the brain through long-term potentiation (LTP), which is essential for learning, occurs only when the activity waves take place for a certain while.

...Jens Stepan, a junior scientist at the Max Planck Institute of Psychiatry in Munich, stimulated the input region of the hippocampus the first time that specifically theta-frequency stimulations produce an effective impulse transmission across the hippocampal CA3/CA1 region. This finding is very important, as it is known from previous studies that theta-rhythmical neuronal activity in the entorhinal cortex always occurs when new information is taken up in a focused manner. With this finding, the researchers demonstrate that the hippocampus highly selectively reacts to the entorhinal signals.

...One possible reaction is the formation of the so-called long-term potentiation (LTP) of signal transmission at CA3-CA1 synapses, which is often essential for learning and memory. The present study documents that this CA1-LTP occurs only when the activity waves through the hippocampus take place for a certain time. Translating this to our learning behavior, to commit for instance an image to memory, we should intently view it for a while, as only then we produce the activity waves described long enough to store the image in our brain. With this study, Matthias Eder and colleagues succeeded in closing a knowledge gap. "Our investigation on neuronal communication via the hippocampal trisynaptic circuit provides us with a new understanding of learning in the living organism. We are the first to show that long-term potentiation depends on the frequency and persistency of incoming sensory signals in the hippocampus," says Matthias Eder.

More information: Jens Stepan, Julien Dine, Thomas Fenzl, Stephanie A. Polta, Gregor von Wolff, Carsten T. Wotjak and Matthias Eder (2012) Entorhinal theta-frequency input to the dentate gyrus trisynaptically evokes hippocampal CA1 LTP, Frontiers in Neural Circuits, Volume 6, Article 64, 1-13. _MedXPress_via_Max Planck

The translation from the German is a bit sloppy, but the basic idea is that in order for the brain to remember something, the information must be encoded properly and presented for a minimum time period.

We have looked at the importance of brain oscillations in consciousness and cognition, previously. Specifically, oscillations in the gamma band -- modulated by theta frequency oscillations -- appear to facilitate communication between different brain centres in an intermittently synchronous manner.

In other words, unlike digital computers, the brain does not have a central clock to maintain synchronous transfer of data. But by using specific modulated brain frequencies, the brain can apparently synchronise information transfer between different parts of the brain for short periods of time.

This is a key concept in understanding how sophisticated biological brains function. An incredible amount of insight can be derived from that simple concept, and Al Fin cognitivists anticipate that both neuroscience and the broader field of cognitive science will benefit immensely, once the insights are more broadly propagated within the various fields of study involved.

Rhythmic Nature of Brain Activity Slowly Unfolds

In monitoring electrical brain activity of motor-cortex neurons, researchers found that they typically exhibit a brief oscillatory response. These responses are not independent from neuron to neuron. Instead, the entire neural population oscillates as one in a beautiful and lawfully coordinated way.

The electrical signal that drives a given movement is therefore an amalgam—a summation—of the rhythms of all the motor neurons firing at a given moment.

"Under this new way of looking at things, the inscrutable becomes predictable," said Churchland. "Each neuron behaves like a player in a band. When the rhythms of all the players are summed over the whole band, a cascade of fluid and accurate motion results."

...electrical engineering associate professor Krishna Shenoy and postdoctoral researchers Mark Churchland, now a professor at Columbia University, and John Cunningham of Cambridge University, now a professor at Washington University in Saint Louis, have shown that the brain activity controlling arm movement does not encode external spatial information—such as direction, distance, and speed—but is instead rhythmic in nature. _R&D Mag

This finding, (Nature, June 3, 2012, doi:10.1038/nature11129), is apparently quite startling to a select group of cognitive psychologists who are not well grounded in biology. For Al Fin cognitive scientists, on the other hand, the rhythmic nature of brain activity -- motor, sensory, and higher order activity -- has always been axiomatic.

The specific rhythms associated with particular brain functions will need to be worked out. The Stanford study helps in this regard, with respect to motor cortex rhythms controlling body motion.

In a series of striking graphs, the Stanford team plotted the signals from individual neurons in the motor-cortex as monkeys completed a series of reaches. The reaching motions are shown by the starburst patterns at the top left of each graph. The neuronal patterns are then plotted atop one another for the entire series of reaches, clearly establishing the rhythmic nature of the brain activity. Credit: Mark Churchland, Stanford School of Engineering

More:

When they monitored the electrical activity of motor-cortex neurons, researchers found that they typically oscillate briefly, not independently as single neurons, but as an entire neural population in one beautifully coordinated way. "Each neuron behaves like a player in a band," said Churchland. "When the rhythms of all the players are summed over the whole band, a cascade of fluid and accurate motion results." The electrical signal for a movement is the sum of the rhythms of all the motor neurons firing at a given moment.

Rhythmic neural activity has been known for a while. It is present in the swimming motion of leeches and the gait of a walking monkey, for instance.

The engineers studied the brain activity of monkeys reaching to touch a target. The pattern of shoulder-muscle behavior could always be described by the sum of two underlying rhythms. "Say you're throwing a ball. Beneath it all is a pattern. Maybe your shoulder muscle contracts, relaxes slightly, contracts again, and then relaxes completely, all in short order," explained Churchland. "That activity may not be exactly rhythmic, but it can be created by adding together two or three other rhythms. Our data argue that this may be how the brain solves the problem of creating the pattern of movement." _Atlantic

The dominant, dysfunctional perspective of machine-oriented cognitivists, who attempt to think of the brain as a mechanical construct -- instead of the evolved organ within a larger organism that it is -- has slowed discovery in cognitive science unnecessarily.

Here is another look at this research:

In the new model, a few relatively simple rhythms explain neural features that had confounded science earlier.

"Many of the most-baffling aspects of motor-cortex neurons seem natural and straightforward in light of this model," said Cunnigham.

The team studied non-rhythmic reaching movements, which made the presence of rhythmic neural activity a surprise even though, the team notes, rhythmic neural activity has a long precedence in nature. Such rhythms are present in the swimming motion of leeches and the gait of a walking monkey, for instance.

"The brain has had an evolutionary goal to drive movements that help us survive. The primary motor cortex is key to these functions. The patterns of activity it displays presumably derive from evolutionarily older rhythmic motions such as swimming and walking. Rhythm is a basic building block of movement," explained Churchland. _PO

Perhaps now, more cognitivists will take this hard-earned insight, and apply it to associative and higher order brain activity. For too long, graduate students and their advisors have struggled under the "single neuron" and "segregated parameters and functions" delusion of brain activity, memory, and control -- perhaps as a confused, erroneous subconscious analogy to computer memory and computer processors.

It is somewhat paradoxical that this important clarification would be made, at least in part, by electrical engineers. It illustrates the importance of multi-disciplinary cooperation in complex fields of research. The real world, it seems, has not been informed that it should hold all its parts within the artificial confines of human academic fields.

Brain Waves and the Limits of Short Term Memory

Capacity of short-term memory impacts the effects of reasoning -- the greater the capacity, the better the effects. Currently researchers conduct studies on developing the most effective ways of training short-term memory.

...In 1995 researchers from Brandeis University in Waltham suggested that the capacity of short-term memory could depend on two bands of brain's electric activity: theta and gamma waves..."The hypothesis formulated by Lisman and Idiart in 1995 assumes that we are able to memorise as many 'bites' of information, as there are gamma cycles for one theta cycle. Research to date provided only indirect support for this hypothesis," say psychologist Jan Kamiński, PhD student from the Nencki Institute... _SD

Most neurocognitivists are slowly coming around to the idea that the "language of mind" is carried via modulated brain waves. The interaction of gamma waves (30Hz and above) and theta waves (4 to 7 Hz) is a particularly intense focus of research into neurocognition.

Recent research at the Nencki Institute in Warsaw suggests that a person's crucial short term memory capacity may be limited by the number of gamma cycles which fall on each theta cycle.

A 'bite' of information refers to its portion in memory. A 'bite' may be a number, letter, idea, situation, picture or smell. "Designing experiments on the capacity of memory one needs to be very careful not to make it too easy for the subject to group many 'bites' into one," stresses Kamiński and as an example gives the following sequence of letters: 2, 0, 1, 1. "Such four 'bites' of information are easy to group into the number corresponding to current year. Instead of four bites of information we are left with just one."

Interpreting the length of theta and gamma waves from EEG recording is not easy either. These waves are not directly visible in the EEG signal. Kamiński proposed a new method of determining them. Researchers recorded brain's electric activity in seventeen volunteers resting with closed eyes for five minutes. Next they filtered the signals and analysed not the cycles themselves but their correlations. Only based on discovered correlations the ratio of the length of theta wave to gamma wave was determined and the likely capacity of verbal short-term memory was determined.

Following the EEG recording, the volunteers, were subjected to classic short-term memory capacity test. It consisted of repeated display of longer and longer sequences of numbers. Each number was presented for one second. Then volunteers had to reconstruct the sequence from memory. At first the sequence consisted of three numbers but at the end of the exam of as many as nine. "We have observed that the longer the theta cycles, the more information 'bites' the subject was able to remember; the longer the gamma cycle, the less the subject remembered. Next we determined the correlation between the results of the tests and estimates from the EEG measurements. Just as expected the correlation turned out to be very high and it confirmed the hypothesis of Lisman and Idiart," says Kamiński._SD

Article abstract...Neurobiology of Learning and Memory

A look at Beta band oscillations and alertness, by the same authors... International Journal of Psychophysiology

And then there is this research into a "master control gene" of memory which may control as many as hundreds of other genes in the brain -- particularly the hippocampus -- which provides a much lower-level glimpse into the machinery of learning and behaviour. All levels of brain function, from the molecular to the behavioural, are important -- although it can be difficult to focus on all of them at the same time.

Our unconscious minds -- including the underpinnings of our short term memories -- function on a parallel basis. Our conscious minds tend to function on a serial -- one thing at a time -- basis. Better educational methods might take those differences into account.

Syncopated Ghost Whispers Haunt Your Internal Web

Long Distance Brain Network Macaque PNAS
Above, you see an early depiction of the "long distance network" of the brain, connecting different brain centres with each other. The complex visualisation was compiled using information obtained from the study of the macaque brain.

The Human Connectome Project is hard at work producing images such as this, using an MRI technique known as diffusion tensor imaging.

With 100 billion neurons, each with around 10,000 connections, mapping the human brain will be no easy feat, and charting every single connection could take decades. The HCP will tackle the lowest hanging fruit first: charting the major highways between different brain regions, and showing how these connections vary between individuals. To do this they will combine several imaging tools including something called diffusion MRI, which maps the structure of the white matter that insulates the "wires" of the brain, and also resting-state MRI, which measures how brain regions oscillate in unison as a result of shared connections. _NewScientist

Cortical parcellations (PDF) such as the above, use another MRI technique. This method of brain visualisation separates different cortical domains which serve particular functions.

These brain images are presented for purposes of orientation and grounding. They may help to picture the various nodes and connections presented in the abstracted images and schematics.

Above, you see some of the brain areas involved in three important brain networks: Default Mode, Salience, and Central Executive. When viewing such fMRI "activation" images, it is helpful to mentally superimpose the connections between the activated brain centres. More on the three pictured networks:

The default mode (DMN) or default brain network (shown in blue) is what your brain does when not engaged in specific tasks. It is the busy or active part of your brain when you are mentally passive. According to Bresslor and Brennon the “DMN is seen to collectively comprise an integrated system for autobiographical, self-monitoring and social cognitive functions.” It has also been characterized as responsible for REST (rapid episodic spontaneous thinking). In other words, this is the spontaneous mind wandering and internal self-talk and thinking we engage in when not working on a specific task or, when completing a task that is so automatized (e.g., driving a car) that our mind starts to wander and generate spontaneous thoughts.

...The salience network (shown in yellow) is a controllor or network switcher. It monitors information from within (internal input) and from the external world arounding us, which is constantly bombarding us with information. Think of the salience network as the air traffic controllor of the brain. Its job is to scan all information bombarding us from the outside world and also that from within our own brains. This controller decides which information is most urgent, task relevant, and which should receive priority in the que of sending brain signals to areas of the brain for processing.

...Finally, the central-executive network (CEN; shown in red) “is engaged in higher-order cognitive and attentional control.” In other words, when you must engage your concious brain to work on a problem, place information in your working memory as you think, focus your attention on a task or problem, etc., you are “thinking” and must focus your controlled attention. _BrainClockBlog

We have talked about the default mode network previously, and will devote future time to the integration of various overlapping -- as well as mutually exclusive -- networks.

Now, we are getting close to the "brass tacks" of how separate brain nodes communicate synchronously with each other via the connectome. The brain functions as a hierarchical network, and depends upon analogous -- but different -- mechanisms of ensemble activity at different levels of the hierarchy.

... when multiple neurons spread all over the brain are tuned in to a specific pattern of electrical activity at a specific frequency, then whenever that global activity pattern occurs, those neurons can act as a coordinated assembly."
The researchers pointed out that this mechanism of cell assembly formation via oscillatory phase coupling is selective. Two neurons that are sensitive to different frequencies or to different spatial coupling patterns will exhibit independent activity, no matter how close they are spatially, and will not be part of the same assembly. Conversely, two neurons that prefer a similar pattern of coupling will exhibit similar spiking activity over time, even if they are widely separated or in different brain areas. _SD

One of the many things that makes understanding the brain so difficult, is the fact that so many things are happening all at once, on so many different levels -- both in serial and parallel format. Almost all of the things that go on in the brain occur on the unconscious or subconscious levels. Consciousness, as we know it, is something of an over-rated evolutionary accident.

Video via Kevin at Brain Clock Blog

Finally, watch ghostly whispers moving through the human brain as it is put through its paces.

More information on brain networks at The Brain Clock Blog: The brain as a set of networks: Fine tunning your networks

Can You Intensify Experience by "Overclocking" Your Brain?

Different parts of the brain communicate with each other via frequency modulation, involving high frequency gamma waves modulated by low frequency theta waves. Presumably, the higher the frequency of the gamma carrier wave, the greater the possible communication bandwidth and the faster a high fidelity signal could be passed accurately. Far more easily said than done, of course.

Neurons recruited for local computations exhibit rhythmic activity at gamma frequencies. The amplitude and frequency of these oscillations are continuously modulated depending on stimulus and behavioral state. This modulation is believed to crucially control information flow across cortical areas....by rapidly balancing excitation with inhibition, the hippocampal network is able to swiftly modulate gamma oscillations over a wide band of frequencies. _ScienceDirect

SPIE
Besides finding ways to prolong one's life, it would be worthwhile to find ways to live one's life more intensely. In earlier postings on Al Fin Longevity, we have discussed ways in which we might reduce the amount of time spent in sleep, without suffering from diminished mental or physical health. There are also everyday ways in which a person can intensify his experience of his waking time. Some examples are listed at the end of this piece.

From the neurocognitive standpoint, the concept of the controlled "overclocking" of the brain -- speeding up the functioning of brain processes so that more can be experienced and accomplished in less time -- is just coming into the realm of possiblity. The concept, once developed, will rest upon a sound understanding of brain processing and inter-brain communications.

Brain activity changes between different brain states, whether awake, asleep, drugged, etc. Besides the activation of different centers in the brain according to brain state, the actual speed (frequency) of brain activity varies with different brain states.

It is thought that synchronous oscillations involving gamma carrier waves (30 to 100 Hz) modulated by theta frequencies (4 to 8 Hz) allow multiple brain processes to occur, including the transfer of working memory to long-term memory, and the binding of different sensory or other inputs into a coherent mental image of an object or idea. In other words, the way the oscillations of the brain are organised on a moment to moment basis, is what allows us to "think" and remember. (see Working Memory: The Importance of Theta and Gamma Oscillations, Lisman, Current Biology Vol 20 No 11)

Gamma oscillations are thought to transiently link distributed cell assemblies that are processing related information1, 2, a function that is probably important for network processes such as perception1, 2, 3, attentional selection4 and memory5, 6. This 'binding' mechanism requires that spatially distributed cells fire together with millisecond range precision7, 8; _Nature

A microcomputer has a synchronous clock that controls the speed of the processes being run. A brain has no such central clock controller, but higher brain function does involve transient locked synchrony between different parts of the brain. How could we speed up this synchrony as it spontaneously occurs and disappears across the cortex?

We know that the top end of the gamma "carrier wave" frequency can vary between types of animals. Some kinds of insects, for example, exhibit brain synchrony at frequencies up to 200 Hz in certain circuits. (Kirschfeld PNAS USA Vol. 89, pp. 4764-4768, May 1992 Neurobiology)

Different frequencies of gamma oscillation serve to connect different brain centers, in practise. This allows for simultaneous parallel activity between multiple circuits. Therefore, when "overclocking," one must be sure not to "step on" the frequencies used by different brain circuits.

There are a number of other cautions, assuming that one had a good idea how to begin to go about ramping up gamma oscillation carrier wave frequencies in the first place. The intricacy of neuronal signaling of brain circuits should discourage any attempts to permanently alter neuronal oscillatory activity. For example, gamma frequencies are closely controlled and modulated by inhibitory interneurons. You cannot change the timing of just one type of cell and expect to maintain a system of smooth communication between brain nuclei. Rather, multiple keys that control the timing of networks across the brain will have to be discovered and mastered.

Al Fin neuroscientists believe that the key to fruitful research along these lines will be found in the field of optogenomics. But Al Fin neuropharmacologists are convinced that they can develop a drug medley which could accomplish the same thing. The biological limits of cognitive functioning will not be easily transcended by just one breakthrough from one particular field of research.

Why should we bother to attempt something which will require so much work? It is possible, after all, to intensify the experience of everyday life without resorting to the extremes of genetic modification of the brain. Below are some of the everyday means by which some persons provide themselves with temporary experiences of high intensity consciousness:

Pharmacological brain stimulants have been used for this purpose for centuries, but in general they extract a steep price from the user who does not exercise prudence. Veterans of combat can attest to the consciousness-intensifying effect of the life-or-death experience. But we are looking for something more sustainable and less risky. Sky-diving, hang gliding, scuba diving, whitewater kayaking, etc. are less risky than combat, but provide a temporary aura of intensity which lingers after the experience. In occupational settings, life or death emergencies attended to by firefighters, police officers, EMS personnel, medical personnel in hospitals, etc. provide temporary "fixes" of intensity. And under the category of "not to be recommended," the commission of a crime and the attendant risk of being caught supplies the outlaw with a feeling of intensity which can become addictive to some. Similarly, committing acts which may be legal but which are socially or occupationally frowned upon, can sometimes provide a touch of that "outlaw intensity," that accompanies risk.

Perhaps the most dangerous method of intensifying experience is to fall in love. The fallout from such a turn is apt to be fatal to any number of persons involved, or in the immediate vicinity. ;-)

Still, the challenges of the modern day world require a significantly higher level of insight and invention than is typically found within populations at large -- even within high IQ populations made up of largely European or East Asian peoples. The inertia of the monkey mind is difficult to overcome. And still we keep trying.

Adapted from an earlier article at Al Fin Longevity

Brain from the Bottom Up: Spontaneous Birth of Synchrony in Small Neuronal Networks

More 13 July 2011: Brian Wang looks at the same research, with an emphasis on the hardware (electronic) aspect. It is fitting to look at both the neurons and the electronics, since the coming cybernetic biosingularity will be dependent upon both.

Human intelligence and consciousness are poorly understood, even by cognitive scientists, neuroscientists, and consciousness specialists. No one understands how to build a human intelligence from scratch, much less how to build a non-human intelligence capable of interacting with humans and the outside world on its own terms. But researchers at Tel Aviv University from the departments of Electrical Engineering and Physics, have taken a fascinating approach to building the basic components of brains: networks of biological neurons. Something wonderful happened when enough cultured neurons linked together in network: They spontaneously "synched up."

Background


Information processing in neuronal networks relies on the network's ability to generate temporal patterns of action potentials. Although the nature of neuronal network activity has been intensively investigated in the past several decades at the individual neuron level, the underlying principles of the collective network activity, such as the synchronization and coordination between neurons, are largely unknown. Here we focus on isolated neuronal clusters in culture and address the following simple, yet fundamental questions: What is the minimal number of cells needed to exhibit collective dynamics? What are the internal temporal characteristics of such dynamics and how do the temporal features of network activity alternate upon crossover from minimal networks to large networks?


Methodology/Principal Findings


We used network engineering techniques to induce self-organization of cultured networks into neuronal clusters of different sizes. We found that small clusters made of as few as 40 cells already exhibit spontaneous collective events characterized by innate synchronous network oscillations in the range of 25 to 100 Hz. The oscillation frequency of each network appeared to be independent of cluster size. The duration and rate of the network events scale with cluster size but converge to that of large uniform networks. Finally, the investigation of two coupled clusters revealed clear activity propagation with master/slave asymmetry.

Conclusions/Significance


The nature of the activity patterns observed in small networks, namely the consistent emergence of similar activity across networks of different size and morphology, suggests that neuronal clusters self-regulate their activity to sustain network bursts with internal oscillatory features. We therefore suggest that clusters of as few as tens of cells can serve as a minimal but sufficient functional network, capable of sustaining oscillatory activity. Interestingly, the frequencies of these oscillations are similar those observed in vivo. _PLoS

More papers by Mark Shein Idelson

Brain synchrony is an important topic of study, linked to consciousness, memory, learning, and normal function of general human brain activity. But synchronous oscillations are also programmed into the neurons themselves, at the smallest level of neuronal organisation. The challenge now, is to build "networks of networks", to discover the communications strategies which interconnected networks will evolve.

Contrast such a biological, bottom up approach with complex machine models of brain function such as the SpiNNaker project out of the University of Manchester, or the Human Brain Project (HBP) led by Henry Markram at Ecole Polytechnique de Lausanne.

Both of the above brain modeling approaches using computers, are based upon bottom-up theories of how brains work. The Lausanne project (HBP) is far more detailed -- going down to the ion channel level of neurons. The Manchester approach is impressive in its parallel computing ambitions, but it begins at the individual "neuronal spiking" level. SpiNNaker is more of a hybrid CompSci:Neurosci approach, than an actual model of the brain like the HBP.

Conventional artificial intelligence approaches do not mimic brain function closely, and are generally more "top-down" approaches, utilising conventional algorithmic concepts of mainstream computer science. Such approaches are doomed to failure before they even begin, as the last 70 years of conventional AI attempts continue to demonstrate.

In reality, brains must be grown. And new types of brains have to be evolved. Not necessarily from biological materials, but up until now the only working brains we know are biological. The first successful autonomous brains are likely to be evolved either from biological materials, or using ingenious abstractions of processes which emerge from biological mechanisms.

Al Fin cognitive scientists suggest that both the Lausanne approach and the Manchester approach are abstracted at the wrong level, if they wish to provide rapid paths to evolved intelligences. Creative human beings will have to discover the appropriate balance, but they will certainly be aided by computing systems in doing so. This is not gobbledygook nor is it AI-psychobabble. It is the genuine crux and pivot point of the problem.

What are the implications for the singularity? There will be no "uploading of consciousness" for the foreseeable future. The cyborg biosingularity is still on schedule for the decade between 2020 and 2030, if humans can avoid an extended Obama Dark Ages. The main question is how many of the cyborg components will be biological in origin, and how many will be non-biological (probably utilising nanotechnology).

In the Brain, Inhibition Sets Us Free

Our brains would not be able to function without inhibitory inter-neurons. The best description that I have read describing how interneurons control brain activity comes from Gyorgy Buzsaki's excellent book, "Rhythms of the Brain."

Scholarpedia presents a nice, brief description of inhibitory interneurons:

The importance of inhibition in the brain is aptly illustrated by the fact that in addition to excitatory principal cells, the brain contains diverse classes of specialized inhibitory interneurons that selectively innervate specific parts of the somatodendritic surfaces of principal cells and other interneurons. In the cortex, axon terminals of interneurons release gamma amino butyric acid (GABA) onto their synaptic targets, where the inhibitory action can compete with the excitatory forces brought about by the principal cells. However, inhibitory interneurons do much more than just provide stop signals for excitation. Proper dynamics in neuronal networks can only be maintained if the excitatory forces are counteracted by effective inhibitory forces. With only excitatory cells, it would be difficult to create form or order or secure some autonomy for transiently active groups, the hypothetical "cell assemblies", because in interconnected networks, excitation begets more excitation. Interneurons, by way of their inhibitory actions, provide the necessary autonomy and independence to neighboring principal cells. The functional diversity of principal cells can also be enhanced by the membrane domain-specific actions of GABAergic interneurons. Additionally, the opposing actions of excitation and inhibition often give rise to membrane and network oscillations which, in turn, provide temporal coordination of the messages conveyed by principal cells. _Scholarpedia

The image above and to the right illustrates a simple 2 neuron oscillator composed of an excitatory neuron and an inhibitory (inter) neuron. Input from the outside is always excitatory, and it is the turning on and off of the inhibitory neuron which accounts for the assembly's oscillation. The image below illustrates a 3 neuron oscillator, with the assembly on the left oscillating at 40 Hz and the assembly on the right oscillating at 30 Hz. The input from the NMDA neuron at the upper left determines which of the two oscillators is operating.

Image SourceReal neuronal assemblies in the brain are far more complex than these simple oscillators. But it helps to picture something simple before thinking about more complex and realistic assemblies -- which have a lot more things that can go wrong. Researchers at Baylor University have recently discovered a genetic variation that leads to dysfunction of inhibitory interneurons in Rett Syndrome -- a devastating neurologic disease of early childhood leading to severe problems of intellectual and motor development.

Children, mostly girls, born with Rett syndrome, appear normal at first, but stop or slow intellectual and motor development between three months and three years of age, losing speech, developing learning and gait problems. Some of their symptoms resemble those of autism.

These inhibitory (gamma-amino-butyric-acid [GABA]-ergic) neurons make up only 15 to 20 percent of the total number of neurons in the brain. Loss of MeCP2 causes a 30 to 40 percent reduction in the amount of GABA, the specific signaling chemical made by these neurons. This loss impairs how these neurons communicate with other neurons in the brain. These inhibitory neurons keep the brakes on the communication system, enabling proper transfer of information.

"In effect, the lack of MeCP2 impairs the GABAergic neurons that are key regulators governing the transfer of information in the brain," said Dr. Hsiao-Tuan Chao, an M.D./Ph.D student in Zoghbi's laboratory and first author of the report.

..."This study taught us that an alteration in the signal from GABAergic neurons is sufficient to produce features of autism and other neuropsychiatric disorders," said Zoghbi, a Howard Hughes Medical Institute investigator and director of the Jan and Dan Duncan Neurological Research Institute at Texas Children's Hospital. _SD

It does not require much interference in the normal operation of the molecular biology of the brain to cause severe dysfunction. The pathway of the dysfunction -- from molecule to synapse to cell assembly to developmental and behavioural dysfunction -- is intriguingly complex on many levels.

My main interest in this regard, is the transient long distance synchrony of cell assembly oscillatory activity in different parts of the brain. It would take several lifetimes to understand such phenomena in all their variation, origination, and modification. The implications of such understanding to human learning, creativity, health and disease, personality, and so on, are profound.


Inhibitory Interneurons and Network Oscillations

Some background reading on the phase-locking of neural populations via inhibitory interneurons PDF [Notice: Opening PDF documents can tie up a browser for several moments. If you think you want to download a PDF document, you may want to right click and select "save linked content as" option.]

Human Oscillatory Brain Activity near 40 Hz Correlates with Cognitive Temporal Binding PDF

The Self-Organising Brain

...the brain of a newborn itself seems to organise its own development. The electrical activity of the waking brain triggers the series of events... _SD

We are not born knowing how to improvise jazz riffs on the saxophone. A newborn child is not able to prove mathematical theorems or argue political economics. But the infant child is born with the potential to do those things. How does the brain bootstrap upon itself to create the sophisticated organ of thought, action, and communication that adult humans possess?

Chaos brews in the brains of newborns: the nerve cells are still bound only loosely to each other. Under the leadership of Academy Research Fellow Sari Lauri, a team of researchers at the University of Helsinki has been studying for years how a neural network capable of processing information effectively is created out of chaos. The team has now found a new kind of mechanism that adjusts the functional development of nerve cell contacts.

The results were published in early September as the leading article of the Journal of Neuroscience.

The work carried out by Lauri's team and its partners at the Viikki campus sheds light on a development path that results in some of the large number of early synapses becoming stronger. The researchers found out that the BDNF growth factor of nerve cells triggers a functional chain which promotes the release of the neurotransmitter glutamate. BDNF enables the release of glutamate by prohibiting the function of kainate receptors which slow down the development of the preforms of the synapses. The activity of the kainate receptors restricts the release of glutamate and the development of synapses into functional nerve cell contacts. _SD

Until brain cells begin communicating via bioelectric signalling, true organisation of the brain cannot take place. The "waking brain" triggers the bioelectric signalling. What creates the "waking brain?"

During early phases of brain development gene expression and postranslational modifications of gene expression are controlled by biochemical signals which are produced in a cellular microenvironment. Later in brain development there is a difference from the development of other organs because electrical signals are added to biochemical messengers as a further signaling in the self-organizing between genes and their respective environments. It must be considered that these electrical signals are capable in influencing gene expression and postranslational modifications. Electrical signals are transported by neuronal processes over distances and with highly topological selectively. This enlarges the range and complexity of the "environment" available to self-organization process. The "environment" relevant to brain self-organization includes all domains with which the evolving brain is capable to interact and from which it receives messages. The same electrical signals which convey messages are used by the brain as information carriers for computational process .... results in the replacement of sensory stimuli by self-organization activity patterns that are contingent on past experience, present motivational state and expectancy of the future. _Turbes1993

Something has to "wake up" the developing brain to initiate this highly complex process of autopoietic development. While the child remains warm and wet within the womb, it has little reason to wake. But when squeezed and thrust into the cold and open brightness, little eyes instinctively open in reaction to the disturbance. This tiny opening sets up a chain of events leading to the jazz riff or the discovery of a new scientific principle.

Far from being a mere local phenomenon, clusters of nerves organise themselves both locally and over distances within the brain, and between the brain and peripheral organs of the body. How is it done?

A considerable amount of evidence and theory suggests that transient, short-lived phase-coupled oscillations within and between specialized areas of the brain provide a mechanism for neural integration. The idea is that these oscillations are coupled or “bound” together into a coherent network when people attend to a stimulus, perceive, remember, think and act. __ScottKelso

In the brain, traditional biochemical signaling and transduction is combined with neuroelectrical stimulation in a complex brew of cognitive magic. Current understanding of the brain is not yet at a fine enough resolution to trace back to first principles of how mind emerges from brain. But we are getting there.

Rhythms of the Brain

All the things that humans have done since the dawn of civilisation have come about because of the rhythmic electro-pulsing inside the cranial vault. Scientific knowledge of the oscillatory language of the brain is accruing rapidly. In Rhythms of the Brain:

Gyorgy Buzsaki guides the reader from the physics of oscillations through neuronal assembly organization to complex cognitive processing and memory storage. His clear, fluid writing accessible to any reader with some scientific knowledge is supplemented by extensive footnotes and references that make it just as gratifying and instructive a read for the specialist. The coherent view of a single author who has been at the forefront of research in this exciting field, this volume is essential reading for anyone interested in our rapidly evolving understanding of the brain.

Source.

Chris at the excellent neuroblog Develintel gives a very favorable review:

"Rhythms of the Brain" begins with the premise that "structure defines function," and then outlines how the architectural principles of neural networks can give rise to neural oscillations. In the process, he meticulously covers topics like the complex, small-world, scale-free connectivity of cortex without resorting to complicated equations - the concepts are carefully grounded in real-world analogies and lay terms.

Buzsáki introduces several other topics that are usually found only in mathematically sophisticated academic works on the brain: for example, how "neural noise" can actually enhance processing through stochastic resonance and the 1/f or "pink noise" signature of EEG, mechanisms of "phase precession" and "phase reset" within nested oscillations, and the difference between relaxation and harmonic oscillators.

Source.

This appears to be a very important book, on a topic that bears on recent discussions in comments here. Although I have not yet read the book, I intend to, and based upon Chris Chatham's recommendation, if you are interested in how the brain works, you should consider reading it as well.

Neural Oscillations and the Virtues of Automaticity

Two good postings on brain function today. One from our old friend Chris at Develintel, titled Models of Active Maintenance as Oscillation. This excellent post continues the series on neural oscillation and synchrony.

Lisman and Idiart, Luck and Vogel, and Nelson Cowan have all suggested that working memory could be the result of the multiplexing of gamma oscillations (20-60 Hz) by theta oscillations (5-10 Hz) in the prefrontal cortex, such that capacity is determined by the number of gamma cycles that can occur within a single theta cycle.

Supporting this highly reductionistic claim are the observations that gamma oscillations are made more prominent by focused attention, that gamma oscillations are known to be important for transmitting information across large cortical distances and for visual binding of features into singular objects. Gamma synchrony is also known to increase performance in target detection as well as recall. Further, by playing auditory "clicks'" at near-gamma frequencies, it is possible to upwardly or downwardly entrain gamma rhythms and directly observe their effects on working memory span - exactly this was done by Burle and Bonnet.

Read the entire post here.

The second post on brain function is from Eide Neurolearning Blog, titled In Praise of Automaticity. Automaticity is when subconscious brain assets take over many tasks for the conscious brain. The Eides point out that conversion of conscious tasks into automaticity saves a great deal of work, and allows for natural progression of learning.

When academic or motor skills don't become automatic, a whole host of problems present themselves. Dyslexic students who have trouble remembering how to form letters automatically, can overload with essay writing, taking notes, or math problem sets (dysgraphia). If math facts, spelling or grammar conventions aren't known to the point of automaticity, then even very intelligent students can find themselves overwhelmed by higher order activities based on these building block skills. As a result, if we don't look for opportunities to accommodate, we may never discover a student's creative or critical thinking strengths.

For us adults, automatic expertise helps us carry out most of our activities of daily living and multi-tasking. It's a beautiful system because it allows us to rest while still getting plenty of work done.

Hat tip to Kevin at Intelligence Testing Blog.

It is helpful to look at brain activity from the different perspectives of neural networks, and developing human organisms.

Brain Synchrony, the Eureka! Moment, and Information Integration

Chris at Develintel Blog gives us two interesting posts titled Neural Correlates of Insight, and Entangled Oscillations.

Both postings concern the phenomenon of synchronous oscillatory neural activity. Brain synchrony is a fascinating topic that Chris posts on fairly often. Many neuroscientists believe that the understanding of brain synchrony holds the key to the decoding of the language of mind.

Chris provides links to excellent papers here, here, and here, to help understand the phenomenon that he describes.