Can We Grasp The Brain’s Complexity?

By Neuroskeptic | July 14, 2014 1:48 pm

An entertaining paper just out in Frontiers in Systems Neuroscience offers a panoramic view of the whole of neuroscience: Enlarging the scope: grasping brain complexity

The paper is remarkable not just for its content but also for its style. Some examples:

How does the brain work? This nagging question is an habitué from the top ten lists of enduring problems in Science’s grand challenges. Grasp this paradox: how is one human brain – a chef d’oeuvre of complexity honed by Nature – ever to reach such a feast as to understand itself? Where one brain may fail at this notorious philosophical riddle, may be a strong and diversely-skilled army of brains may come closer.

Or

It remains an uneasy feeling that so much of Brain Science is built upon the foundation of a pair of neurons, outside the context of their networks, and with two open-ended areas of darkness at either of their extremities that must be thought of as the entire remainder of the organism’s brain (and body).

And my favorite:

As humans tend to agree, increased size makes up for smarter brains (disclosure: both authors are human)

I love it. I’m not sure I understand it, though.

The authors, Tognoli and Kelso, begin by framing a fundamental tension between directed information transfer and neural synchrony, pointing out that neurons firing perfectly in synch with each other could not transfer information between themselves.

They say that the ‘transfer’ model originated in the physiology of the synapse, while the ‘synchrony’ model grew out of electrophysiology – and both are valid at different scales. But how are we to reconcile them? How can information flow through a brain where everything is locked in synch?

Here’s how they depict the issue:

fnsys-08-00122-g002

What’s in the big black uncharted zone? It’s a great question, and Tognoli and Kelso frame it very well. However, I’m not sure if they answer it. In a rather abstract way, they point to the notion of metastability as the master key to unlocking the brain, saying that

A new paradigm would help to integrate principles that seem contradictory in their radical form: transfer and synchronization, as well as integration and segregation. Those pairs of concepts are reconciled under the dynamical regime of metastability…

Under a metastable regime, information is continuously created, preserved and annihilated by spatiotemporally changing coalitions among parts and processes. This is a source of dynamic complexity, and the likely origin of the human brain’s many prowesses.

OK. But it would be nice to see some examples of metastability in neuroscience, yet none are detailed. Then again, the paper does include references to the authors’ previous papers on this topic, such as this one. Perhaps there is more to be found there.

ResearchBlogging.orgTognoli E, & Kelso JA (2014). Enlarging the scope: grasping brain complexity. Frontiers in Systems Neuroscience, 8 PMID: 25009476

CATEGORIZED UNDER: funny, papers, science, select, Top Posts
ADVERTISEMENT
  • FH

    Great to read you are excited about this topic, but a bit surprised you call for examples of metastability in neuroscience.

    There are plenty around, ranging from indirect evidence of metastable states via (multi-)fractal characteristics of oscillations said to indicate self-organised critical (multi-stable) states (e.g., Sporns, Linkenkaer-Hansen), to neural avalanches in single cell recordings and spatiotemporal dynamics (e.g. Freeman) and models of (recurrent) neural computation (Hopfield (1984); Langton(1990); MacLennan (1999) ;Kello (critical branching), Kaiser (hierarchical topology, extended criticality) ).

    And let’s not forget Ashby who wrote several decades ago:

    “As a particular case, a type of system that deserves much more thorough investigation is the large system that is built of parts that have many states of equilibrium. Such systems are extremely common in the terrestrial world; they exist all around us, and in fact, intelligence as we know it would be almost impossible otherwise” (Ashby, 1962).

    A selection:

    Ashby, W. R. (1947). Principles of the self-organizing dynamic system. The Journal of General Psychology, 37(2), 37–41. doi:10.1080/00221309.1947.9918144

    Langton, C. (1990). Computation at the edge of chaos: Phase transitions and emergent computation. Physica D: Nonlinear Phenomena, 42, 12–37.

    Friston, K. J. (1997). Transients, metastability, and neuronal dynamics. NeuroImage, 5(2), 164–71. doi:10.1006/nimg.1997.0259

    MacLennan, B. (1999). Field Computation in Natural and Artificial Intelligence Extended Version. Information Sciences, 119(1-2), 73–89. doi:10.1016/S0020-0255(99)00053-5

    Bressler, S. L., & Kelso, J. (2001). Cortical coordination dynamics and cognition. Trends in Cognitive Sciences, 5(1), 26–36.

    Kaiser, M., Görner, M., & Hilgetag, C. C. (2007). Criticality of spreading dynamics in hierarchical cluster networks without inhibition. New Journal of Physics, 9(5), 110–110. doi:10.1088/1367-2630/9/5/110

    Freeman, W. J. (2008). A pseudo-equilibrium thermodynamic model of information processing in nonlinear brain dynamics. Neural Networks : The Official Journal of the International Neural Network Society, 21(2-3), 257–65. doi:10.1016/j.neunet.2007.12.011

    Palva, J. M., Zhigalov, A., Hirvonen, J., Korhonen, O., Linkenkaer-Hansen, K., & Palva, S. (2013). Neuronal long-range temporal correlations and avalanche dynamics are correlated with behavioral scaling laws. PNAS. doi:10.1073/pnas.1216855110

    • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic

      Thanks – I see there is tons of stuff out there!

      • Sys Best

        Yes, plenty of stuff just like those excerpts. As I said, difficult to demonstrate them they are wrong, although they can’t demonstrate they are right.

    • Sys Best

      Just because the shoes fit her feet doesn’t mean she’s Cinderella.

      All those references you provide are more or less opinion articles and speculation. Still, the evidence is not there. The difficulty is to tell someone that their opinion is not correct when there is no evidence, for or against it.

      • Fred Hasselman

        I can see why you would call simulation studies “opinions”, but I wouldn’t call analyses of properties of observed physiological time series that point to the existence of multiplicative interactions and self-organised critical avalanche dynamics “speculation”.

        I think what you describe is called “science”. There never is a definitive answer and the goal is to evaluate both theoretical, formal and empirical arguments in order to decide which account is more plausible.

        I could have also focused on the abundance of evidence against the validity of the Neuron Doctrine and the “input-output” information processing model of the neuron and the brain as a whole, but that was not the topic of the post.

        • Sys Best

          Here’s a nice discussion on the topic, not that it brings any hard evidence for the neuromodel

          Being Critical of Criticality in the Brain
          http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3369250/

          From Merriam-Webster
          ‘speculation: ideas or guesses about something that is not known’

          Which cherry-picked features of which physiological signals do you specifically refer to?

          • Fred Hasselman

            Yes, that’s a great article!
            Does it provide evidence that the brain is a simple linear input-output information processing machine? That we should use factorial designs and average brain activity across subjects in order to find independent components we can think of as independent functions/modules that are the efficient causes of our behaviour and experience of the internal and external world? No.

            Whether those studies showing long range dependence actually are evidence of SOC can be evaluated empirically, for example using the SOC formalism describing the necessary and sufficient conditions for self-organized critical processes, see Aschwanden 2012 http://arxiv.org/abs/1204.5119 (this will actually be the next project to embark on for me)

            Cherry picking is one thing, but completely ignoring anomalies to theories / ontology is problematic for a science (e.g. http://www.ingentaconnect.com/content/imp/jcs/2012/00000019/F0020003/art00005)

            So, decide for yourself if these are all cherries or may contain anomalies:

            Some of the critiques against the Neuron Doctrine and “path/trace theories” by Ashby (1931) http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1038822/pdf/jnpsycho00022-0052.pdf and Lashley (1950) http://gureckislab.org/courses/fall13/learnmem/papers/Lashley1950.pdf remain unresolved (also by complexity perspective)

            More recent empirical evidence against the doctrines 4 postulates (some of these examples are from https://www.theguardian.com/science/neurophilosophy/2013/aug/09/a-new-way-of-thinking-about-how-the-brain-works):

            (1) the neuron is the fundamental structural and functional unit of the nervous system;

            Glia cells have signalling functions too and they are abundant depending inn cortex/cerebellum, total brain: 50/50.

            (2) neurons are discrete cells which are not continuous with other cells;

            Nope.

            (3) the neuron is composed of 3 parts – the dendrites, axon and cell body; and (4) information flows along the neuron in one direction (from the dendrites to the axon, via the cell body).

            Antidromic firing

            Impulses can originate anywhere

            Role of dendritic spines

            Glia cells play a role in conducting signals (possibly control large populations of neurons over long ranges: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4426493/)

          • Sys Best

            Critical Truths About Power Laws

            MATHEMATICS

            Michael P. H. Stumpf 1 and Mason A. Porter

            Most reported power laws lack statistical support and mechanistic backing.

            SCIENCE VOL 335 10 FEBRUARY 2012

          • Fred Hasselman

            Yes. There are several of those articles, e.g. from Santa Fé institute.

            So the key is to decide whether claims of power laws in neuroscienc are statistically reliable and the data reveal more characteristics of SOC than just the power law.

            I can only speak for such studies on time and trial series of human performance and physiology:

            In section 4 of this http://didier.delignieres.perso.sfr.fr/Publis-docs/Diniz%20et%20al%20_2010_%20HMS.pdf we discuss some additional conditions that need to be met in order to pass for a signature of complexity.

            One importan issue is experimental control over appearance and disappesrance of the scaling laws, this has been shown and replicated in fitts tasks and word naming
            In addition, but this is less strong as evidence of SOC also between populations of experts/novice cycling, rowing, skiing pathological/healthy ADHD, dyslexia, parkinsons, alzheimer, heart disease

            Another is the strong prediction that longer timeseries should produce more variance (lower freqs on the power law) and not less variance (prediction if noise is random). This has been shown in several studies, most dramatically a 6hr word naming study. https://www.cs.colorado.edu/~mozer/Teaching/syllabi/7782/readings/van%20orden%20holden%20turvey.pdf

            On the other hand, I can return the questions about reliability and validity of claims of meaningful “average activity” with respect to the use of inferential statistics based on the classical ergodic theorems

  • daniel8

    this has also been explored with the concept of criticality (or balanced state), by two main actors, Chialvo and Deco. They both demonstrated quite neatly (starting from Chialvo I would say) that even simple models, tuned around criticality or metastability best explain real brain dynamics at large scale.

    This is definitely an important step, and Tognoli and Kelso frame it correctly in terms of information transfer.

    Of course we have to be careful in not taking the explanation as the reason. Models tuned around criticality are able to explain the richest repertoire of states, and furthermore the conservation laws of physics that are connected to the critical state are the ones allowing for long range correlations as the ones found in the brain. I am afraid that sometimes this is forgotten and the criticality or metastability are taken as the cause of, and not the most convenient way to model brain function.

    Brain organization into resting state networks emerges at criticality on a model of the human connectome.
    Haimovici A, Tagliazucchi E, Balenzuela P, Chialvo DR.
    Phys Rev Lett. 2013 Apr 26;110(17):178101.

    Ongoing cortical activity at rest: criticality, multistability, and ghost attractors.
    Deco G, Jirsa VK.
    J Neurosci. 2012 Mar 7;32(10):3366-75.

    Spike avalanches in vivo suggest a driven, slightly subcritical brain state.
    Priesemann V, Wibral M, Valderrama M, Pröpper R, Le Van Quyen M, Geisel T, Triesch J, Nikolić D, Munk MH.
    Front Syst Neurosci. 2014 Jun 24;8:108. doi: 10.3389/fnsys.2014.00108

    and my 50p, connecting criticality and information transfer

    Information transfer and criticality in the Ising model on the human connectome.
    Marinazzo D, Pellicoro M, Wu G, Angelini L, Cortés JM, Stramaglia S.
    PLoS One. 2014 Apr 4;9(4):e93616

  • Deneb Needs Memes

    ALife guru Steve Grand has been working on something he has cryptically termed yin and yang pathways which the very vague descriptions on his weblog make sound maddeningly close to a candidate model for this sort of thing. Grand has been very tight-lipped about the details; hopefully if he really is on to something big he won’t sit on it for years the way Darwin did.

  • Jonathan

    Annihilated or consolidated, is the question that comes to my mind. I doubt the brain could ever know what information to destroy; but I can see the brain constantly optimizing. Would it make sense to suppose that information which has been learned must adjust to information which is entirely new?

  • Pingback: Can We Grasp The Brain's Complexity? | Dr. Goul...()

  • http://petrossa.me/ petrossa

    To me it’s simply the result of the evidently needed regulatory mechanism evolved to integrate all the various bits starting to integrate itself in the process.

    “There was once a mammal. It needed a lot of little bits of operating systems in order to let all components of its body function properly. Over time they became so numerous that it needed a system to coordinate the other bits . That system became so complex that it was capable to reprogram itself in order to be able to assimilate the ever increasing flow of information. It called itself: conscience.
    Objectively impossible to determine if it exists, since conscience itself determines what are the criteria defining conscience.

    That conscience, in an attempt to preprogram future acts of the body, starts tell a tale to itself.
    A continuous flowchart enabling it by correlating previous events and by means of extrapolation to arrive at a predefined future action.

    The conscience calls that tale: reality. Again objectively impossible to determine if it exists, the conscience stipulates what is reality.”

    http://petrossa.me/2010/04/16/the-brain-believes-do-you/

  • dante chialvo

    Of course, the issues of metastability/flexibility/criticality discussed by our friends Tognoli & Kelso are all fundamental for our understanding of brain and mind dynamics. These views are receiving increasing attention lately, thanks for spreading the word!. Our own contributions to this long-standing questions are listed in our webpage (www.chialvo.net). Cheers , Dante

  • Pingback: The beauty of brain science | neuroecology()

  • Pingback: Improve Brain Memory IQ Mind Brain News and Informative Articles | Brain Training » Can We Grasp The Brain’s Complexity?()

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Neuroskeptic

No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.

ADVERTISEMENT

See More

@Neuro_Skeptic on Twitter

ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar
+