[Comp-neuro] The sniffing brain and free will

james bower bower at uthscsa.edu
Mon Aug 18 20:49:49 CEST 2008

First:  "Greetings, an i´ll keep reading you because you are an  
authority in computational neuroscience."

Don't do that!!!  --  In the interest of full disclosure  perhaps I  
should have explicitly stated some time ago, for those of you who  
don't already know, that my views including those on the evolution of  
cerebral cortex are not shared by the majority of my colleagues :-),  
although also, in full disclosure, I am far from the first person to  
suggest a role for olfaction in cortical evolution (in fact, I would  
say that it is kind of obvious).  So if you want to read an  
"authorized authority on computational neuroscience" you should stop  
reading now!!  There are lots of excellent sources written by people  
who want that kind of authority.   I am less of an authority than  
simply someone who for some reason (low GABA) doesn't mind rocking the  
boat.  :-)

Anyway, if you are still reading, it is not about what came first,  
locomotion or olfaction.  All things I am aware of locomote, olfact,  
sense light, etc.  The argument is that 3 layered cortex came before 6  
layered cortex (i.e. neocortex), and that 3 layered cortex evolved  
originally in the context of the olfactory system.

Movement dictated the organization of various aspects of the spinal  
cord.  Neo-cortical 'movement control centers' if that is what they  
are, where built on top of the more fundamental spinal chord  
mechanisms (I realize that kind of statement is reckless BTW).

To vastly, vastly simplify the evidence and summarize what in fact I  
hope will be a book soon, the fore-brain of the earliest vertebrates  
(at least represented by modern examples -- which does, of course,  
have its own problems) consists of a 3 layered cortex receiving input  
from the olfactory bulb, which then in turn projects to a 'multi- 
modal' cortex mixing up inputs from other sensory systems (vision,  
audition, somatosensation, etc).  As one moves "up" the phylum, what  
one can suggest (if predisposed to be troublesome) are 'pre- 
processing' areas emerge that are (largely) sensory system specific.   
Even these sensory system specific areas, however, in non-mammals,  
share with the olfactory system, the fact that the sensory input  
contacts pyramidal cells in layer I, and spreads across the cortex  
(for example in turtle 'visual' cortex) rather than the pattern in  
neocortex, where the input comes vertically into layer IV.  In 6- 
layered Neocortex, which is largely a mammalian invention, the input  
comes into layer IV although recent studies have suggested that the  
inside to the outside system of layer I efferents in neocortex has a  
pretty dramatic effect on cortical processing.  Our neo-cortical,  
layer IV, feedforward, vision system dominate approach to studying the  
cerebral cortex has meant that this input was not considered for many  
years (gee it's far from the soma too) and is not considered to this  
day in most abstract models of cortical function.  It also posses  
rather significant problems for the idea of the computational  
independence (or I would claim existence) of cortical columns  In my  
view of the world (distorted), this input should actually be seen as  
the historical input to cortical circuitry, So, what does this view of  
cerebral cortical evolution change:

1) it is the highly interconnected 'associative' networks of 3 layered  
cortex (or IT for that matter), and not the much later evolved (pre- 
processing) stages of areas like VI, with their stronger vertical  
organization, that should be considered 'architypical cortex'.  Thus  
the fundamental organizational (computational) structure of cerebral  
cortex is horizontal, not vertical, and thus has nothing to do with  
cortical columns.

2) VI and other primary sensory areas are not the first evolved by the  
last evolved in the inside out evolution of cerebral cortex - thus VI  
of the rat and VI of the monkey are much further removed from each  
other, evolutionarily than is enterhinal cortex in either animal  
(therefore explaining, I think, the rather remarkable differences in  
organization of VI in different mammals).

3) Cerebral cortical evolution has not been dictated by the  
computational problem faced by vision, or any other neocortical  
sensory system -- but instead the 2-D relatively local style of  
computation appropriate for vision has been adapted to the more  
universal (gestault) type processing intrinsic to olfaction - the  
early stages of visual processing in cortex are basically pre- 
processing stages to adapt visual data representation into something  
more appropriate for processing in an olfactory style.  I doubt that  
birds, whose 'higher processing region" is more likely derived from  
the geniculate and thus a visual oriented style of processing, for  
example "think" of trees as being one kind of thing - this kind of  
grouping is the sort of thing that the olfactory system has to do, and  
not the kind of thing that the visual system fundamentally cares about  
(but happy to use once the olfactory system invented it).

4) it is silly to hope to understand 3-layered cortex, like the  
hippocampus or enterhinal cortex, or even IT, using visual stimuli.   
(I have always been amused when I visit most (with only a couple of  
exceptions - Howard Eichenbaum being one) hippocampal labs at the  
graduate students with alcohol rubbing down their mazes between runs,  
to make sure that there are no olfactory cues left by the previous rat.

5) Similarly for frontal neo-cortex.

6) it makes no sense to me at all that for every one modeler,  
experimentalist, anatomist, cognitive type, or brain imager studying  
olfaction (in some cases only 1), there are 100s studying vision.  One  
might almost suspect that we are the odd mammal that believes (note  
believes) that the visual system is really really important.  Why  
would that be?  :-)  I raise horses - you can cover their eyes and  
they don't care - but mess with their nose -- trouble.   (BTW, in our  
textbooks, vision is represented by multiple chapters in the front of  
the book, while the olfactory system usually gets about 5 pages  
somewhere in the back -- odd reflection of the number of cerebral  
cortical areas associated with each).

7) There's lots of stuff we won't ever understand (cortical  
oscillations, slow wave sleep, the functional structure of pyramidal  
cell dendrites,  human cognition and strange stuff like meditation  
(see previous reference to our recent TINS article), without first  
understanding or at least considering the olfactory system.

A practical example: Charlie Grey, whose paper with Wolf Singer was  
one of two responsible for neuroscientists and theorists becoming  
interested in cortical oscillations opened his recording filters to  
look for 40 Hz in visual cortex because he had been a graduate student  
with Walter Freeman studying the olfactory system. Unfortunately most  
thus turned on to oscillations have yet to leave neo-cortex or its  
surrogate, the thalamus (sigh)).


On Aug 16, 2008, at 11:34 AM, anibalmastobiza at terra.es wrote:

> Ok. Dr. Bower.
>  But if we frame the issue  in an evolutionary timescale is it not  
> better  to focus on the machinery responsible for movement and the  
> valuations needed to choose the apropiate course of action  
> available. Because its more pertinent if we would like to talk about  
> the computations underlying "free will" or cognitive freedom.
> I think even before the evolution of our magificent sense of smell  
> is earlier the biological mechanisms for movement (sensory neurons,  
> motoneurons, interneurons...) so it is even better to look at the  
> way the spinal cord computes.
> Greetings, an i´ll keep reading you because you are an authority in  
> computational neuroscience.
> Greetings to all.
> Ahora también puedes acceder a tu correo Terra desde el móvil.
> Infórmate pinchando aquí.
> _______________________________________________
> Comp-neuro mailing list
> Comp-neuro at neuroinf.org
> http://www.neuroinf.org/mailman/listinfo/comp-neuro


Dr. James M. Bower Ph.D.

Professor of Computational Neuroscience

Research Imaging Center
University of Texas Health Science Center -
-  San Antonio
8403 Floyd Curl Drive
San Antonio Texas  78284-6240

Main Number:  210- 567-8100
Fax: 210 567-8152
Mobile:  210-382-0553

The contents of this email and any attachments to it may be privileged  
contain privileged and confidential information. This information is  
for the viewing or use of the intended recipient. If you have received  
e-mail in error or are not the intended recipient, you are hereby  
that any disclosure, copying, distribution or use of, or the taking of  
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  
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,  
e-mail and all attachments to this e-mail must be immediately deleted  
your computer without making any copies hereof and any and all hard  
made must be destroyed. If you have received this e-mail in error,  
notify the sender by e-mail immediately.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080818/9895a636/attachment-0001.html

More information about the Comp-neuro mailing list