Jessie Buchanan 
Ryan Scofield
Mike Moran 
Alicia Jones
Dave Reis

Humboldt State University

MOTOR LEARNING AND MEMORY
by
Alicia Jones

INTRODUCTION

     There are three major motor areas of the brain, the
cerebellum, the cerebral motor cortex, and the basal
ganglia.  All of these areas play a part in controlling
various motor functions, but it appears that the
cerebellum is the principle area where motor learning and
memory occur.  
     The cerebellum is the largest part of the motor
system and works with the cerebrum, spinal cord and other
structures, to regulate the rate, range, and force of
movements(Mallonee 13).  The cerebellum specifically
controls elementary reflexes, posture and locomotion, and
voluntary movements(Ito 353).
     The cerebellum is enveloped by the cerebellar
cortex, which consists of three layers.  Only five types
of neurons are present in these layers, Purkinje cells,
granule, Golgi, stellate, and basket neurons.  The
Purkinje cells, which are inhibitory, provide the sole
output from the cerebellum.  Input reaches the cerebellum
from two excitatory fibre systems, the mossy and climbing
fibres.  Each Purkinje cell receives converging input
from a large number of mossy fibres by way of parallel
fibres that are sent by the granule cells.  Input is also
received by one climbing fibre for each Purkinje
cell(Dudai 180).  
     The focus of this paper is to discuss the type of
synaptic changes that occur in the cerebellar cortex that
allow learning and memory to transpire.  We will also
discuss the chemical aspects of these changes, the
synaptic mechanisms, neural systems, and synaptic
specializations that underlie the morphology of the
synapses.

NEURAL PHENOMENA AND MOTOR LEARNING
     
     There has been much discussion regarding the
phenomena of long-term potentiation(LTP), a theory
proposed by D.O. Hebb.  According to this theory, the
co-occurence of pre-and postsynaptic activity can result
in an increase in synaptic efficacy due to modification. 
This phenomena is thought to be a key element in learning
and memory, and has been extensively studied in the
hippocampus.  However, there is a notable lack of recent
studies of this particular neuronal change occurring in
the motor areas of the brain, including the cerebellar
cortex. 
     Instead, scientists such as Masao Ito and others
have evidence that the synaptic changes indicating
learning in the cerebellum are a result of long-term
depression(LTD), which is triggered by a short-lived
afferent activation and leads to a long-term suppression
of postsynaptic activity(Klimesch 195).  
     A study of the function of LTD took place with mice. 
Long-term depression is absent in mutant mice that lack a
specific metabotropic glutamate-receptor subtype.  Such
mutants can learn the classical conditioning of the eye
blink reflex as well as normal animals can during the
first three days of training but on the fourth and fifth
day, the mutant animals learn significantly less.  Some
scientists consider this to be a demonstration of the
role of cerebellar LTD in associative motor learning,
although others may argue that synaptic-plasticity
deficits in other brain structures may account for the
late learning deficit of the mice.  It was noted that the
motor control deficits of the mice were similar to the
clinical symptoms of cerebellar lesions and were probably
caused by the absence of cerebellar LTD.  The mice were
ataxic, had impaired walking, and performed badly on
tests of balance, although they groomed normally and
could swim(Schutter TINS 292).
     Ito's study of the vestibulo-ocular reflex(VOR) also
shows the role of cerebellar LTD in learning.  This
reflex stabilizes the eye against changes in head
position.  The signals that initiate this reflex
originate in the inner ear.  He observed that the
parallel fibres mediate vestibular signals, and the
climbing fibres mediate ocular signals.  He suggests that
concurrent activation of parallel fibres and climbing
fibres induces long-term depression in the parallel
fibres synapting on Purkinje cell dendrites, and
contributes to experience-dependent modification of the
reflex.  He further postulates that LTD probably involves
glutamate receptor desensitization(Dudai 181-2).
     While long-term depression appears to be the most
likely neural phenomena that occurs in the cerebellum,
there are other types of phenomena that are possible
mechanisms for learning and memory in the motor areas of
the brain.  Sprouting and neurogenesis are two examples
of neural phenomena that may be involved.  Sprouting is
the growth of new axons that replace damaged or destroyed
axons, and has been observed at neuromuscular junctions. 
One study involved the staining of rat soleus muscles and
observing the course of denervation and reinnervation. 
"Following denervation...regenerating axons grow back
into the muscle by following the endoneurial tubes
previously occupied by the axons that innervated the
muscle, and thus are led to denervated endplates.  Motor
axons arriving at these endplates proceed to re-occupy
the endplate site"(Son TINS 281).  This new growth may
have some bearing on learning and memory.  
     Neurogenesis is another phenomena that may occur
during the process of learning.  A series of studies with
canary song was conducted by F. Nottebohm and his
colleagues on the question of neurogenesis.  Normal
female canaries will sing when treated with androgens as
adults.  This appearance of song is accompanied by a
marked expansion of the HVc(hyperstriatum ventrale, a
component of the song system).  The scientists used
autoradiographic monitoring of the incorporation of
radioactively labelled thymidine to detect neurogenesis. 
One-year-old female canaries were injected with
testosterone, or cholesterol as a control, and then
injected with radioactively labelled thymidine.  About
five weeks later, the birds were sacrificed and their
brains sectioned.  Results showed that in both
experimental and control birds, neurons were added to the
HVc, and there was massive cell death to accommodate the
new neural growth.  If neurogenesis is connected with
learning, it may occur in other areas of the brain
besides the forebrain, including the motor areas.
     Neural phenomena is still an area that needs further
investigation before any concrete theories can be
offered.   The theories of long-term potentiation and
depression have been documented as being possible
mechanisms of learning and memory, especially in the
hippocampus and the cerebellar cortex.  However,
sprouting and neurogenesis are mysteries that may or may
not be influential in learning.  

BIBLIOGRAPHY

De Schutter, Erik.  Cerebellar Long-Term Depression Might
Normalize Excitation of Purkinje Cells: a Hypothesis. 
Trends in Neuroscience 18, 1995: 291-295.

Dudai, Yadin.  The Neurobiology of Memory.  New York:
Oxford UP, 1989.

Ito, Masao.  The Cerebellum and Neural Control.  New
York: Raven, 1984.

Klimesch, Wolfgang.  The Structure of Long-Term Memory. 
Hillsdale: Lawrence Erlbaum Assoc., 1994.

Mallonee, Jay.  The Effect of Learning and Memory on the
Cortical Cells. Thesis.  Humboldt State University,
Arcata, 1983.

Son, Young-Jin, Joshua T. Trachtenberg, and Wesley J.
Thompson.  Schwann Cells Induce and Guide Sprouting and
Reinnervation of Neuromuscular Junctions.  Trends in
Neuroscience 19, 1996: 280-284.





Intracellular Mechanisms That Allow Learning and Memory.

By Jessie Buchanan

    Motor learning can be described as the establishment
of changes within the motor system.  The more complex and
original the behavior, the more the neural circuits in
the motor systems of the brain must be modified (435). 
Although skill learning primarily involves changes in the
neural circuits which involve control movement, it cannot
take  place without sensory guidance from the surrounding
environment.  When performing motor tasks, one receives
feedback from muscles, joints, vestibular apparatus,
eyes, and interaction with objects.
    But what exactly is occurring?  Researchers have
found that memory formation is registered in the form of
cell biological change in various neurones in the brain
(Rose 211).  There are many hypotheses as to what exactly
occurs but it is generally believed that the change
involves synaptic remodeling to reorganize pathways and
connectivity (211).  Because motor learning also involves
sensory and perceptual learning, it would follow that key
changes take place within the hippocampal formation, as
well as in other parts of the brain.  However, most
research has been focused on changes within the
hippocampal formation.
    The hippocampal formation is a region of the limbic
cortex found in the temporal lobe and includes the
entorhinal cortex, the subicular complex, the
hippocampus, the dentate gyrus, the perforant path, areas
CA3 and CA1, the Schaffler collateral axon, and the
Schaffer commissural hippocampal formation lead to long
term physiological changes that seem to be partly
responsible for learning (Carlson 447).Lomo (1966) found
that intense electrical stimulation of axons staring in
the entorhinal cortex and leading to the dentate gyrus
causes a long term increase in EPSPs n the postsynaptic
cells (447).  This increase is termed long-term
Potentiation.     The occurrence of long term
potentiation requires two simultaneous events: activation
of synapses and depolarization of the postsynaptic neuron
(452).  It appears to be triggered by an influx of
calcium through a type of glutamate receptor called NMDA. 
However, this is not the only mechanism by which the
concentration of calcium within the cell can be raised.  
It can also be raised by influx through voltage activated
calcium channels, though the activity of pumps and
exchangers, or calcium can be released from intracellular
storage sites through increases in concentration of
sodium, calcium, and IP3 (Martin, Nicholls, Wallace
(265-266).  However, it seems that the influx of calcium
is critical for long-term potentiation normally enters
through the NMDA receptor.  This is supported by research
with drugs that block the NMDA receptors.  Research has
found that the drug AP5, which blocks NMDA receptors,
prevents the establishment of long-term potentiation in
area CA1 and the dentate gyrus (Carlson 452).     NMDA is
found within the hippocampal formation, especially in CA1
and controls the calcium ion channel.  It is normally
blocked by a magnesium ion which does not allow for the
influx of calcium.  In order for NMDA to allow calcium to
enter the cell, glutamate must bind to the NMDA receptor,
and the cell must be depolarized so magnesium is ejected
(Carlson 452).  If the postsynaptic membrane is
depolarized causing magnesium to be ejected, and
glutamate is binded to the NMDA receptor, calcium is free
to enter.
    Once the influx of calcium occurs, it triggers a
second messenger system, which is a series of molecular
events leading to a functional response such as the
closing or opening of membrane channels (Martin,
Nicholls, Wallace 726).  Because of this, calcium is
often called a second messenger.Some membrane channels
(ie specific potassium, cation selective, and chloride
channels) are directly regulated by the intracellular
connections of calcium (266).  Calcium activates the
persistent enhancement of synaptic transmission by
activating three major targets: Protein kineses C,
calmodulin, and calpain (266).  Calpains are a group of
proteins that have been  implicated in the regulation of
cytoskeleton as well as a number of membrane" proteins
(266).  Calmodulin is a ubiquitous protein that has four
calcium-binding sites.  When these sites are occupies, a
variety of enzymes including calcium calmodulin dependent
protein kineses, calsineurn, and nitric oxide (266). 
Once activated, protein kinase C becomes persistently
active.
    There is evidence that the calcium activated second
messenger, or even calcium acting directly, may cause the
release of "one or more retrograde messengers from the
dendritic spines of the active postsynaptic cell"
(Kandell 684).  The retrograde messengers diffuse to the
presynaptic terminal to activate one or more second
messengers that enhance the transmitter release, thereby
maintaining long-term potentiation (684).  The retrograde
messengers are not known but it has been speculated that
they may be nitric oxide, or carbon monoxide working
alone or jointly with other molecules (684).
    Some research has been done that does suggests that
nitric oxide may indeed be a retrograde messenger
involved in long-term potentiation (Carlson 458).  It has
been reported that drugs that block nitric oxide
synthesis prevents long-term potentiation establishment
in hippocampal slices (Haley, Wilcox, Chapman, 1992). 
They also reported that long-term potentiation was
blocked by a chemical which destroys the nitric oxide
interstitial fluid.
    There are many hypotheses as to what is exactly
occurring intracellularly during learning and memory. 
However, although the research seems to establish the
role of NMDA receptors in long-term potentiation, there
is just as much research which shows that its role is nog
as broad as many people think.  The only conclusion which
can be made are that NMDA plays at least some role in the
induction of long-term potentiation , and that the influx
of calcium is critical for long-term potentiation.  NMDA
may not play such a large role in long-term potentiation
in all parts of the brain involved in learning, but does
play a major role in area CA1 of the hippocampal
formation.

Carlson, N.  (1994).  Physiology of behavior.  Boston:
Allyn and Bacon. 
Haley J.E., Wilcox, G.L., and Chapman, P.F.  The role of
nitric oxide in hippocampal long term potentiation. 
Neuron: 1992, 211-216

Jessel, T.M., Kandell, G.R., and Schwartz, J.H.(EDS),
Essentials of Neuroscience and behavior (pp.  667-694)
Connecticut: Appleton and Lange. 
Martin, A.R., Nicholls, J.G., and Wallace, B.G.(1992). 
From Neuron to Brain (3rd ed.) Massachusetts: Sinquer
Associates, INC.

Rose, S.P.R, (1983).  Toward a biochemistry of memory
formation.   
Caputto, R, and Mason, C.A., MD.  (Eds.), Neural
Transmission, Learning, and Memory (Vol.  10, pp. 
211-220).  New York: Raven Press.


Neuronal Structural Changes

by Mike Moran

In both developing and adult animals the same components
of the cell are changed in response to repetitive
activation, above routine levels (skill learning):
synaptic number, size, shape and possibly perforation;
dendritic spine number and spine head and neck shape; and
dendritic branch length.  These alterations have been
observed in different protocols such as normal
development, neuronal stimulation in the adult,
behavioral training, long term potentiation, imprinting,
environmental enrichment, different states of hydration,
epileptic discharges, etc...  This consistency in
neuronal response to repetitive activation above routine
level suggests that these aspects of cellular plasticity
are not simply the result of maturation during
development or some aberrant response to researchers
protocols.  Rather, the universal nature of these
findings suggests that these cellular components are
plastic during adulthood and are capable of responding to
external events with an alteration in their structure. 
Further, and most important, the observed changes in
neural anatomy form a series of events.  This series of
events appears to begin with changes in the shape of the
synapse and increases in the size of the pre or post
synaptic thickening, as well as increase in the size of
the spine head and changes in the shape of the spine
neck.  This appears to be followed by the formation of
new synapses, perhaps a division (initially seen as
synaptic perforations?) of existing synapses.  Finally
new dendritic spines and new dendritic branch length are
added to the cell.  Given the universal agreement of the
data collected in this area, it would seem reasonable to
conclude that this series of events constitutes the
neurons structural response to repetitive activation
above routine levels, and that such a series of events
likely forms the basis of long term skill learning and
memory processes.
    The question remains as to what these structural
changes mean to neuronal functioning.  Would they, as
Hebb suggested, increase the power of synaptic
transmission from the pre to post synaptic neuron?  It
seems reasonable to conclude that increases in the number
of synapses would increase the efficacy and power of
transmission between interconnected neurons and, as such,
could form a basis fora learning and memory.  Also, an
increase in the size of the synapse, particularly the pre
synaptic terminal, would lead to an increase in the
number of available calcium channels, which would in
turin increase the magnitude of the vesicular release. 
These structural changes in the synapse,along with the
increased number of synapses, would result in an increase
in the total number of synaptic quanta available for
release per impulse in affected neurons.  The alteration
in the curvature of the synapse may allow new areas of
the membrane and their associated transmitter receptors,
previously hidden within the membrane to now become
exposed; such an increase has been observed in glutamate
receptors following long term potentiation (Baudry et
al., 1980).  Finally, alterations in the shape of the
dendritic spines and spine necks would alter their cable
properties and increase the power of synapses located on
them (Fifkova, 1985).
    All of the above research points to a mechanism by
which the repetitive use of neurons can cause alterations
in their shape.  These alterations can be rapid and
underlie not only the electrophysiologiclal consequences
of repetitive activation such as long term skill memory
and learning, but also would be a reasonable substrate
for the formation of all kinds of learning and memory. 
This model depends on repeated activation of neurons
above routine levels and allows the neuron a mechanism
for turning transient signals into relatively long term
changes in structure.  These morphologic alterations, in
turn, create more contact points, allowing more complex
networks and greater integration and storage of
information.     In summary, these findings indicate that
both synaptic size and number are very quickly altered
with neuronal use.  The number of synapses rapidly
increase and there may be changes in the shape and
proportion of different types of synapses.  The
morphology of the synapse and dendritic spine also
changes, with large increase in the size of the pre or
post synaptic terminal in certain synaptic types,
depending on protocol, species and brian area examined. 
Change in the number of dendritic spines and dendritic
length appear to occur over a longer period of time. 
Bailey, Chen (1983).  Morphological basis of long term
habituation and sensitization.  Science 220: 91-93

Baudry et al., (1980).  Increase in glutamate receptors
following repetitive electrical stimulation in
hippocampal slices.  Life Science 27: 325-330 
Chang, Greenough (1984) Transient and enduring
morphological correlates of synaptic activity and
efficacy change in the rat hippocampal slice.  Brain Res.  309:
35-46

Duffy, Rakic (1983).  Differentiation of granule cell
dendrites in the dentate gyrus of the rhesus monkey: A
quantitative golgi study.  Journ Comp Neurol 214: 224-237  Dyson,
Jones (1984) Synaptic Remodeling during
development and maturation: Junctional differentiation
and splitting as mechanisms for modifying connectivity. 
Dev Brain Res 13: 125-137

Lynch (1985) What memories are made of.  The Sciences 25:
38-43 
Tweedle, Hatton (1984).  Synapse formation and
disappearances in adult rat supraoptic nucleus during
different hydration states.  Brain Res 309: 373-376 
  

     Dendritic Spines  in LongTerm Potentiation 

                 by Ryan Scofield 

    First discovered nearly a hundred years ago, the role
of dendritic spines in neural functions still remains a
mystery.  Anatomist Cajal first saw them in 1888,
describing them as  small twiglike appendages arising
from the branchlets of Purkinje cell dendrites" (Cajal,
1911).  Until recently the small size of the dendritic
spines has thwarted any direct measurements of its
functional properties.Regardless, many attempts at
theorizing possible explanations have long been
investigated.  The combination of these theories with
recent experimental and computational data has provided
few clear answers, but have brought modern research to
believe the assumption that the spines work biochemically
in conjunction with neurons rather than electrically. 
This assumption has been well studied and is generally
agreed upon, but the physiological and behavioral
implications of the dendritic spines can only be answered
theoretically and hypothetically, with no real concrete
answers.        DS's are large in number, with  as many
as 15,000 spines, at a density of two spines per
micrometer of dendritic length, cover(ing) the surface of
a layer V pyramidal cell in the visual cortex.  In
Purkinje cells, the number can be as high as 200,000"
(Koch, Zador, Brown, 1992).The exact dimensions of the
spine varies, with a neck length of 0.08 to 1.58 um and
with a diameter of 0.04 to 0.46 um. The volume of the
spine neck varies from 0.004 to 0.56 um.  "Spines are so
small that at a resting calcium concentration of 80 nM
only about three free calcium ions would be found in a
spine with the average head volume of 0.051 um (Koch,
Zador, Brown).     The shape of dendritic spines,
specifically the length and diameter of the spine neck,
may change during neuronal development or in response to
behavioral significant stimuli.  "High frequency
electrical stimulation of specific hippocampal
pathways--have also
been reported to alter spine heads, changes in the shape
of the spine stem,...and more synapses on the shaft. 
However, it is unclear what direct role...these changes
have in causing changes in synaptic efficiency" (Koch,
Zador, Brown).     A popular hypothesi concerning the
function of the 
dendritic spine is that the spine acts as a precursor to
brain disorders.  Several pathological conditions have
been found to greatly effect the DS.  The 
..deafferentation of the visual pathway causes
degeneration of spines of lateral geniculate neurons and
spiny stellate cells in layer 4C of the visual cortex. 
Spines on cortical pyramidal cells degenerate in several
types of mental disorder.  Spines are thin and tiny in
Down's syndrome; fewer, elongated, and irregular in form
a Patua's syndrome; and long and tortuous in mental
retardation" (Shepherd, 1992).  The nature of the spine
changes in these disorders is not presently known.  They
may possible result from the primary disorders of the
neuron or spine, or the possibility of a disuse or
overuse of the spine synapses.
    A second view of the dendritic spine is that it acts
as a local dendritic input-output unit.  A classical look
at the DS is that it is purely a postsynaptic structure,
but in fact many types of spines are also presynaptic. 
The spines of granule cells within the olfactory bulb
receive type 1 synapses from mitral cell dendrites, while
at the same time reciprocal type 2 synapses are connected
to those same dendrites.  These particular findings
provide examples of several principles of the
neuro-organizational pattern within the spines.  "They
show, first, that a single spine can serve as a synaptic
input-output unit, and additionally, the spine is the
smallest neuronal compartment capable of performing a
complete input-output operation of a single synapse. 
Second, (they show that) a single neuron can contain many
of these input-output units, thereby greatly increasing
the computational capacity of the neuron.  Third, because
these units are distributed widely on a dendritic tree,
it implies that individual units can function
independently of each other", and lastly,  these synaptic
input-output functions can take place without the need
for generation of impulses" (Shepherd, 1992).
    Lastly, in connection to long-term potentiation, it
is commonly believed that the morphological changes in
the spine underlie structural changes believed to be the
physical basis of behavior.  These changes have been
proposed to underlie both behavioral memory and long-term
potentiation of synaptic responses.  Spine changes are
basically the only reported morphological changes seen in
conditions related to memory, with long-term memory
reflecting these morphological changes in neurons.
However, in the spirit of the enigma that is the
dendritic spine, brain  slices taken from 15 day old
pups, where spines are not yet fully developed, express
far larger LTP than slices taken from adult rats,
indicating that spines are not essential for expression
of LTP" (Segal, 1995).

Bibliography

Koch, Cristoff (1992).  Dendritic Spines: Convergence of
Theory and Experiment.Science 256: 973-989

Segal, Menathem (1995).  Dendritic Spines for
Neuroprotection: A Hypothesis.  Trends in Neurosciences
18: 458-470

Shephard, Gordon M (1996).  The Dendritic Spine: A
Multifunctional Integrative Unit.  The American
Psychological Society 75: 2197-2207

Yuste, Rafael (1995).  Dendritic Spines as Basic
Functional Units of neuronal integration.  Nature 375:
682-684.





The Motor Neural Hierarchy:
An Overview and Its Relevance
To Skill Formation and Memory.







David Reis
Psychobiology
Dr. Morgan



The purpose of this paper is to provide a brief overview
of the neural systems responsible for the formation of
motor skill learning and memory.  The major neural
structures will be addressed including a description of
the motor neural system as a whole and their relevance to
skill memory and learning.

First, a definition of skill is necessary.  Motor skill
behavior acquisition involves classifying and actively
selecting motor acts from prior experience (motor memory)
that are conducive to a specific overall goal, and
combining them with other stimuli.  Skill behavior
formation varies widely in complexity with respect to two
elements; degree of input from multiple sources (sensory,
cognitive, spatial), and the degree to which the involved
motor acts are voluntary.  Thus, a motor behavior that
involves a single motor movement with a high degree of
autonomy is a simple skill.  One that requires a higher
degree of voluntary motor action and receives many inputs
in its formation is a more complex skill.

Attempting to assign memory into clear-cut categories and
analogous neural systems, while being a gross
oversimplification, provides a baseline working neural
model for memory function.  For the purpose of this
paper, it will be accepted that the formation of motor
behavior involves an independent (non hippocampal)
hierarchy of motor behaviors (Fuster, 1995).  This set of
structures ascends from the spinal cord to the cerebral
cortex.  Keep in mind that the formation of a skill
involves combining an old skill memory with a short term
motor memory or program.  Thus, any programs of behavior
combining to form a complex skill (goal) require a series
of intermediate steps to reach that goal.  It follows
that each of these steps consist of smaller sub-steps,
and so forth.  Each of these steps, sub-steps, etc. vary
in complexity and are formed in a neural structure of
corresponding complexity.

With regards to specific neural structures and to the
formation of overall behavior, it is difficult to neatly
attribute specific hierarchal functions to specific
neural structures.  Rather, the overall hierarchy of
structures is like a symphony orchestra; if one
instrument is missing, others miss their cue (and so
forth), and the overall system suffers.  In addition, the
fact that one instrument may be more difficult to play
(more complex) does not make it more indispensable than
simpler elements in the overall "symphony".

Just as it is difficult to isolate and analyze specific
neural structures from the whole, it is just as difficult
to separate skill from other motor behaviors (are all
learned motor behaviors skills?).  Furthermore, skill
learning and skill memory go hand in hand.  For this
reason, the terms motor memory and skill memory will be
assumed to mean the same thing.

The first structures in the motor hierarchy are the
spinal cord and the cerebellum.  These structures play an
essential role in motor action, but because of their
place in the motor hierarchy (bottom), they will not be
covered in this paper.

Next in the hierarchy are the basal ganglia consisting of
(among other structures), the caudate nucleus, the globus
pallidus an the putamen.  Lesions of the basal ganglia in
rats produce serious motor and behavioral deficits
(Packard, 1990).  Caudate nucleus lesion studies suggest
the caudate may mediate the initial integration of motor
information distributed in cortical areas.  Rats in this
study exhibited a deficit in acquiring what would be
analogous to a learned skill task.

A study of basal ganglia lesions in humans (Bhatia, 1994)
showed motor and behavioral deficiencies associated with
these lesions.  Specifically, caudate nucleus lesions
produced mostly behavioral problems such as loss of
mental and motor initiative.  Lesions of the lentiform
nucleus (Globus pallidus, putamen) produced mostly motor
disorders such as parkinsonism, ballism and chorea. 
Numerous other studies have been done on the basal
ganglia, most of them fairly consistent in suggesting
both a motor and behavioral integrative rule in motor
formation.  What is clear is the basal ganglia, with its
massive amount of connective loops to other areas of the
brain, plays several crucial roles in the motor
development of the animal.  What exactly those roles are,
however, is the subject of much debate.

From the basal ganglia the hierarchy extends upward to
the prefrontal cortex, premotor cortex and motor cortex. 
The cortical structures will be described from a "bottom
down" perspective, dealing with the prefrontal (highest)
cortex first and then down to the motor cortex (lowest).

The prefrontal cortex at the top of the motor hierarchy,
is the center of planning, execution and initiation of
voluntary action.  A patient with prefrontal damage has
difficulty spontaneously initiating new sequences of
motor behavior.

In terms of skill behavior, the short term
representations of voluntary action are formed and stored
in the prefrontal cortex.  The acquisition of a skill,
however, involves the coordination of old and new
memories.  For example, a typical delay task involves two
types of new information; the procedural portion (i.e.
the basic rules of the task) and the trial specific
memory of the cue and of the motor response.  Upon first
learning the delay task, both kinds of memory are new and
the animal needs the prefrontal cortex for both. 
However, once the task has been well learned, the cortex
is no longer needed for the procedural portion of the
task.  It becomes automatic and is relegated to lower,
more involuntary structures of the hierarchy, probably
the premotor cortex or basal ganglia.  However, the
animal still needs the prefrontal cortex for the trial
specific memory.  Simply stated, the development of a
skill would involve constant and incremental drawing of
automatic motor memory from subcortical areas and
combining it with new trial specific stimuli.

The premotor cortex is the first step down in the
cortical hierarchy.  Anatomically and physiologically the
premotor cortex is divided into the premotor cortex (PMC)
proper and the sensorymotor cortex (SMA). Neurons in the
PMC discharge in response to inputs from the prefrontal
cortex.  This response is measured to occur well before
the actual motor action occurs, and is therefore assumed
to be involved in preparation for movement or motor
sequence.  It is believed the PMC neurons encode complex
motor acts (i.e. spatial coordinates, motor sequence and
the goal) rather than simple individual movement.

The more complex SMA is believed to be involved in the
formation of motor memory.  SMA neurons have been shown
to be activated by movement during the learning of a
visuomotor task in monkeys.  After numerous repititions
of the same tasks, the activation of SMA cells disappear. 
Presumably, the encoding of the task migrates to cells in
the PMC and basal ganglia.

The motor cortex is unquestionably the lowest cortical
area in the motor hierarchy.  Learning and conditioning
here have a lesser role, although unit representation is
still subject to some degree of change with learning.

In summary, the motor hierarchal neural modes as well as
some of its important structures have been described
briefly in this paper.  These descriptions have been
integrated with their relevance to the formation and
retention of motor skill behavior.  Although the neural
system model is fairly consistent with anatomical and
lesion studies of the brain, further study is required
before the system can be more fully understood.

References Cited

Bhatia, B.P. and Marsden C.D.  The Behavioral and Motor
Consequences of Focal Lesions of the Basal Ganglia in
Man.  Brain 117 859-874, 1994.

Packard, M.G. and White N.M. Lesions of the Caudate
Nucleus Selectively Impair "Reference Memory" Acquisition
in the Radial Maze.  Behavioral and Neural Biology 53,
39-50, 1990.

(Main source of paper):
Foster, J.M. Memory in the Cerebral Cortex:  An Empirical
Approach to Neural Networks in the Human and Non- human
Primate.  1st Edition.  Cambridge, MA:  MIT Press, 1995
p.161-190.

Copyright © 1997, Dr. John M. Morgan, All rights reserved - This page last edited March 17, 1997
If you have any feedback for the author, E-mail me