A Neural Theory of Mind

Mark Laubach laubach at biogfx.neuro.wfu.edu
Tue Apr 23 07:18:40 EST 1996

paul at phy.ucsf.edu (Paul Bush) wrote:

>Thanks for your very informative post, you obviously know a lot about this system.
>Therefore, could you write a single paragraph that sums up your opinion of the
>function/operation of the basal ganglia? Just a high level idea. Your hunch, gut
>feeling on the subject. Doesn't matter if some bits are wrong - speculate a
>bit, This would be incredibly useful to me.

OK.  I will speculate a bit here.

My idea for how the BG work is based on anatomy and cellular
physiology.  We know that all areas of cortex innervate the striatum
and that these inputs are organized topographically.  So, one way to
view the striatum is that it is a _map_ of the cortex.  We also know
that spiny cells only fire action potentials when many thousands of
active corticostriatal neurons converge on a spiny neuron
synchronously (this is based on work from Charlie Wilson).  This
suggest that spiny neurons are _coincident event detectors_.  The
contact probability of any single cortical afferent on a given spiny
neuron is very low (Tony Kincaid, from Wilson's lab, reported this at
last year's SFN meeting).  We also know that inputs to given region of
the striatum do not just come from one column of cortex or from one
cortical field.  This suggests that the representation of info from
the cortex to striatum is  _sparse_ and _distributed_ .  Finally,
there is a many-to-one mapping between cortex and striatum.  This
suggests that there is a _convergence_ of info to a given region of
striatum, yet overall the topograhic organization maintains a
_divergence_ of info between different regions of striatum.  I view an
activation of a spiny cell as indicating a coincident activation of a
collection of cortical cells distributed in different locations in
cortex and activations at different levels of the striatum  as
representing unique info occuring in relation to a common "event"
(e.g., a behavior, thought).

The higher level view then is that the striatum is one place in the
brain where _binding_ may occur.  It seems that the info that is bound
here is related to motor preparation, motor sequencing, and
sensorimotor intergration.


>Specifics - Please point out where I am wrong here:

>|> >I don't have a good understanding of how the basal ganglia
>|> >work. Basically, though, the striatum appears to be a lateral
>|> >inhibitory network designed for filtering out input patterns
>|> >(compromised in Huntinton's). 
>|> Maybe.  The evidence for an effect of one spiny neuron on its
>|> neighbors is lacking despite a good bit of effort to find it (see
>|> Jaeger and Wilson in J. Neurophysiol last year).  Many modellers have
>|> incorporated the notion of lateral inhibition into their models of the
>|> basal ganglia.  In fact, a paper from my lab argued that this
>|> circuitry gives rise to a form of working memory (Woodward et al. in
>|> the Houk book).  The best evidence for this to date comes from studies
>|> of striatal interneurons (see stuff by Kita, Kawaguchi, Plenz).
>|> Still, the jury's out on this one ...

>Yes, I think working memory is a good label. Interestingly, I now think that
>cortex is also a big laterally inhibitory network.

I agree especially in light of recent paired intracellular recording
studies that have emphasized the interneurons in cortex and striatum

>|> >A projection from the dopaminergic
>|> >midbrain perhaps provides a lower brain 'go' signal to inititate a
>|> >process in frontal cortex (compromised in Parkinson's). 
>|> Probably not.  Schultz has shown that da neurons fire to trigger
>|> stimuli in simple tasks but when sensory events (conditional stimuli)
>|> are added before the trigger cue, the da cells shift to the earlier
>|> events.

>Is this not because the earlier events are important features that must be
>abstracted along with the trigger stimulus to form a higher level (in space/time)
>plan of the situation? The da cells then give a 'go' signal when the stimuli
>first appear and initiate the firing of the plan pattern in frontal cortex.

I think this has more to do with "setting the occasion" for behavior
than giving the go signal.

>|> The exaxt role of the da signal is yet to be clarified.
>|> Also, it should be pointed out that the standard bit about da and the
>|> striatum is based on the rat.  In primates, da neurons innervate all
>|> areas of cortex (see paper from Kaas' lab last year in JCN).

>So the reward signal in primates is more widely distibuted. In rats it just says
>yes/no to the top-level plan.

This assumes that the BG represent the top-level plan.  They may just
be working on figuring out what has happened in cortex.  Without this
(internal) info, planning is not possible.  

>|> >There is probably a role for the two separate patch and matrix
>|> >systems. Interestingly, they receive input from different depths of
>|> >cortical layer 5. This would correspond to different stages of
>|> >prediction of the future in each cortical module. 
>|> This really only holds for the prefrontal cortex and its neighbors.  

>OK, the time difference is only important for the higest level plans.

That would assume that areas like prelimbic are the source for
planing.  Maybe but prelimbic has diverse connections (visceral,
limbic).  These are anything but higher-order.

>This patch/matrix story seems complicated. Any overall idea what's happening here?

Different receptor systems in the two compartments; maybe different
regulation of a common neural circumstance (in the sense of a
coordinated pattern of firing).

The thing about the patches is that they are avoided by some areas
(for example, medial agranular cortex, the homologue of SMA/PMA) and
densely innervated by otheres (e.g., cingulate and prelimbic).  These
areas are not interconnected in cortex and the are not overlapping in
the patches, but they might converge in the matrix.  Also,
intralaminar thalamus, which projects more to striatum than to cortex,
only innervates the matrix.  I think then that the patches are a
further level of order and parcellation of inputs. 

>|> >The basal ganglia assign processes to high level frontal cortex (attention)
>|> >which has reciprocal relationships with the limbic cortex, which as
>|> >described previously is the model the brain makes of itself -
>|> >consciousness. In this way consciousness is integrated with perceptual state.
>|> The best evidence for the role of the bg in the control of behavior
>|> comes from studies of reaction-time performance. 

>From studies of its role in the motor modality, then.

I would say sensorimotor.  Neurons in BG only respond to things motor
or sensory _in the context of some task_ (in the context of some
strucuture).  BTW, sensory responses change with context.  In my work,
a neuron that responds in preparation for movement in a reaction-time
task may also fire to one of two tones in a tone-discrimination task
(in this case, tone-cued treadmill locomotion).

>|> The effect of str
>|> lesions of da depletions is to eliminate a process known as motor
>|> readiness, responding faster with more time to prepare a response.
>|> Carli et al. (1989) showed that the interpretation of a role of the bg
>|> in  motor preparation, and not attention, is valid. 

>In the motor modality, the high level neural equivalent of attention is 'motor 
>readiness', if you think about it. Its the activation of a 'plan' that can cover 
>large distances in space/time.

OK.  We'll learn more about the sensory domain when I start my
post-doc this fall.

>Look at Huntington's - people start with absentmindedness, move to disordered
>thoughts and eventually total dementia - the bg has a role in organizing
>attention in thought as well as the motor system, I think.

You should look up a book in the Progress in Brain Research series
(our libraries copy has been missing now for more than a year).  There
was an article on studies of pallidal lesions where people sat around
for hours not doing anything and then realized that so much time had

I think the term attention is difficult to define and carries a lot of
baggage.  Try to be more specific about what you mean by attention.


Mark Laubach
Dept. of Physiology & Pharmacology
Bowman Gray School of Medicine
Wake Forest University
Winston-Salem, NC 27157
laubach at biogfx.neuro.wfu.edu

More information about the Neur-sci mailing list