Re: [Comp-neuro] Post-Doctoral position at Blue Brain Project, EPFL, in Mesoscale Cortical Synthesis

As Background (including experimental evidence or the lack there of):

The original proposal that the cerebral cortex is constructed of a series of vertical columns was made by Vernon Mountcastle in a largely theoretical paper published in 1957.  "Modality and Topographic Properties of Single Neurons of Cat's Somatic Sensory Cortex"  J. Neurophysiology 20: 408-434 .  This paper was actually written as a summary of a series of single electrode receptive field mapping studies performed in the Mountcastle laboratory at Johns Hopkins over the previous years but was specifically paired with a data rich paper:  Mountcastle, V.B., Davis, P.W. and Berman, A.L. (1957) Response Properties of Neurons of Cat's Somatic Sensory Cortex to Peripheral Stimuli.  J. Neurophysiology, 20: 374-407.

The introduction of the theoretical paper states very clearly that Mountcastle is proposing:

 "An Hypothesis of the functional organization of this cortical area."

 And more specifically:

 "This is that the neurons which lie in narrow vertical columns, or cylinders, extending from layer II through Layer VI make up an elementary unit of organization, for they are activated by stimulation of the same single class of peripheral receptors, from almost identical peripheral receptive fields at latencies which are not significantly different for the cells of the various layers."

 

Over the years, I have taught a graduate course called, "the History of Your Science" in which students pick a classical paper in their fields, read it and then trace the idea presented to modern times.  I also require them to provide evidence that the paper they present is actually "classical".

 To start the course, I almost always present the theoretical paper by Mountcastle – with the proof of its classical status being the fact that its citations start out much lower than the paper with the data, but today essentially, is the only paper of the two referenced.  The reason for this is that subsequent experimental studies of the responses of neurons in somatosensory cortex, especially in unanethetized animals, have essentially shown that all the experimental claims made by Mountcastle as demonstrating cortical columns were wrong. (Those interested in digging deeper into the issue, might be interested in the defense of cortical columns Mountcastle felt it necessary to publish in 1997:  http://brain.oxfordjournals.org/content/120/4/701.short)

 As the students and I go through the results of subsequent experiments, not only in somatosensory cortex, but also in other regions of neo-cortex including especially the visual cortex, which was largely responsible for the success of the cortical column idea (see Mountcastle 1997), they become increasingly amazed at the extent to which the "idea" of cortical columns stands in contrast to the actual data.

 Of course, the reason this is the case is because "the idea of a cortical column is appealing", as neuroscientists especially of the theoretical type are always looking for ways to reduce the complexity of the real nervous system.

 In the case of the cortical column, however, I would claim that its proposed existence has had many profound and unfortunate consequences for thinking about how the cerebral cortex works.  As just one example, "cortical column" thinking meant for many years that neurobiologists both experimental and theoretical tended to under consider or under value the extensive set of horizontal connections between cortical neurons.  In fact, neurobiologists continued to be surprised that "extra columnar" horizontal connections are so predominant. 

 I would (and have) also claimed that the interest in cortical columns reflects yet another underlying bias or assumption made by many theoretical and experimental neurobiologists, and that is that the visual system (the most studied region of cerebral neo cortex) is a good model for understanding the larger question of cerebral cortical organization.  I would (and have) claimed that, in fact, the visual system, and especially the primary visual cortical areas actually represent a structurally and evolutionarily quite eccentric form of cerebral cortical organization, and certainly should not be considered a prototype for cerebral cortex as a whole.  In fact, from an evolutionary point of view, it is far more likely that the olfactory system and olfactory cortex represent the fundamental computational structure of cerebral cortical circuits, and there is no evidence of any columnar structure in those networks.  In those networks "horizontal" projections dominate.

 Of course, that makes things much more complex, and I probably should also say, harder to understand and also to sell.

These things matter.

Jim Bower

On Jul 5, 2013, at 5:16 PM, ramana dodla <ramana.dodla@gmail.com> wrote:

I don't know the answers to Jim's questions, but they touch upon issues that are
of importance to (neuro-) scientists. But as a matter of scientific quest, and
from a physics point of view, the idea of a cortical column is appealing. I could think of them
as a set of eigen-shapes that could interact with one another, and define the cortical
communication. For this approach, in fact any other shape would be equally good/bad.

This fits into the idea of methodological reductionism. Physics is a nice example.
The human body itself is another best example. Who knows what is best for the brain,
but I don't know why reductionism shouldn't work in the brain. Isn't reductionism the idea
behind electrophysiology?

rmn

On Thu, Jul 4, 2013 at 9:24 AM, james bower <bower@uthscsa.edu> wrote:

Why not, it's the summer, CNS meeting is coming up, and it has been a long time since our previous discussion of 'noise in the nervous system'.

So, why not:

Serious  question:  do cortical columns actually exist? Is this the right way to think about cortical processing?

More formally, is the idea of a cortical column, originally proposed by Mountcastle in the 1950s as a "computational" (in current lexicon)  building block for cerebral cortex:

1) supported by the data

2) the right way to think about cortical structure?  computational or otherwise.

In case anyone is wondering, I think not - and have found it mildly amusing that this project is devoted to reconstructing something that probably doesn't exist.

:-)

Jim Bower

On Jul 4, 2013, at 7:43 AM, Shillcock Julian Charles <julian.shillcock@epfl.ch> wrote:

The Blue Brain Project has modelling infrastructure for constructing in silico neocortical columns containing about 30,000 neurons, distributing the cells through the column and forming synapses. Several such columns have already been assembled into a planar hexagonal mosaic containing up to 1,000,000 neurons. However, axons grow long distances along tracts within the brain's white matter, and the next stage of development is to populate three-dimensional mesh models of brain regions with appropriately-shaped mesocircuits, adjusting their dimensions and shape to match rodent brain anatomy, and connecting the circuits according to known large-scale connectomics data,  yielding a complete rat brain model containing on the order of tens of millions of neurons.

Please see this page for more deals: http://emploi.epfl.ch/page-94325-en.html

Julian Shillcock

_______________________________________________
Comp-neuro mailing list
Comp-neuro@neuroinf.org
http://www.neuroinf.org/mailman/listinfo/comp-neuro

 

 

Dr. James M. Bower Ph.D.

Professor of Computational Neurobiology

Barshop Institute for Longevity and Aging Studies.

15355 Lambda Drive

University of Texas Health Science Center 

San Antonio, Texas  78245

 

Phone:  210 382 0553

Email: bower@uthscsa.edu

Web: http://www.bower-lab.org

twitter: superid101

linkedin: Jim Bower

 

CONFIDENTIAL NOTICE:

The contents of this email and any attachments to it may be privileged or contain privileged and confidential information. This information is only for the viewing or use of the intended recipient. If you have received this e-mail in error or are not the intended recipient, you are hereby notified that any disclosure, copying, distribution or use of, or the taking of any action in reliance upon, any of the information contained in this e-mail, or

any of the attachments to this e-mail, is strictly prohibited and that this e-mail and all of the attachments to this e-mail, if any, must be

immediately returned to the sender or destroyed and, in either case, this e-mail and all attachments to this e-mail must be immediately deleted from your computer without making any copies hereof and any and all hard copies made must be destroyed. If you have received this e-mail in error, please notify the sender by e-mail immediately.

 

_______________________________________________
Comp-neuro mailing list
Comp-neuro@neuroinf.org
http://www.neuroinf.org/mailman/listinfo/comp-neuro

_______________________________________________
Comp-neuro mailing list
Comp-neuro@neuroinf.org
http://www.neuroinf.org/mailman/listinfo/comp-neuro

 

 

Dr. James M. Bower Ph.D.

Professor of Computational Neurobiology

Barshop Institute for Longevity and Aging Studies.

15355 Lambda Drive

University of Texas Health Science Center 

San Antonio, Texas  78245

 

Phone:  210 382 0553

Email: bower@uthscsa.edu

Web: http://www.bower-lab.org

twitter: superid101

linkedin: Jim Bower

 

CONFIDENTIAL NOTICE:

The contents of this email and any attachments to it may be privileged or contain privileged and confidential information. This information is only for the viewing or use of the intended recipient. If you have received this e-mail in error or are not the intended recipient, you are hereby notified that any disclosure, copying, distribution or use of, or the taking of any action in reliance upon, any of the information contained in this e-mail, or

any of the attachments to this e-mail, is strictly prohibited and that this e-mail and all of the attachments to this e-mail, if any, must be

immediately returned to the sender or destroyed and, in either case, this e-mail and all attachments to this e-mail must be immediately deleted from your computer without making any copies hereof and any and all hard copies made must be destroyed. If you have received this e-mail in error, please notify the sender by e-mail immediately.