Large-Scale Brain Area Functional Organization - Computational Explorations in Cognitive Neuroscience - page 209

the thalamus and its role in visual processing in chap-
ter 8. Although once thought to be just a relay sta-
tion, the thalamus is increasingly being appreciated as
an active processing system, contributing for example
to attentional processing. Several pathways exist within
the thalamus that could subserve communication among
disparate cortical areas, providing a potential role in co-
ordinating processing in these different areas, for exam-
ple. However, not much is known at this time about the
true extent of thalamic involvement in cortical process-
ing.

complements the frontal cortex, which is also a motor
control structure. In short, there is a tradeoff between
the kind of continuous, constraint-satisfaction process-
ing that is thought to occur in the cortex, and the need
for a more discrete, high-threshold system for deciding
when the accumulated evidence warrants some kind of
action. Thus, the basal ganglia might be specialized for
performing this high-threshold decision-making pro-
cess. So, whereas cortical neurons are constantly firing
and shaping the flow and integration of information, the
basal ganglia neurons are quietly waiting until they re-
ceive just the right pattern of input activation from the
cortex to trigger an action.
The continuous versus discrete tradeoff idea is con-
sistent with the oft-discussed notion that the basal gan-
glia are important for action selection — choosing an
appropriate action given the current sensory-motor con-
text. Furthermore, neural recordings of neurons in the
striatum, which is the principal area where action selec-
tion is thought to occur, show that the neurons there are
mostly completely silent, which is consistent with the
high-threshold notion. This idea is also consistent with
one of the main effects of Parkinson's disease, which
affects the basal ganglia, where patients are unable to
initiate movements — this is just what one would ex-
pect if the system that is responsible for detecting when
to make an action is impaired. Interestingly, these pa-
tients are otherwise capable of making relatively normal
movements once they have been induced, which would
suggest that the basal ganglia are really just specialized
for the initiation of action, but not its execution. Prob-
ably the execution depends more on the frontal corti-
cal areas, which are bidirectionally connected with the
basal ganglia.
It should be clear that a high-threshold detection sys-
tem is appropriate for initiating motor outputs — mov-
ing muscles uses a lot of energy, and should not be done
needlessly. By the same token, it should also be clear
that such a high-threshold system would be useful for
controlling the initiation of purely cognitive “actions.”
In chapters 9 and 11, we consider the possibility that the
symbiotic relationship between the frontal cortex and
basal ganglia that likely has evolved for motor control
could have been coopted for controlling the updating
and storage of activation-based memories in the frontal

7.3.4 The Basal Ganglia, Cerebellum, and Motor

Control

For obvious reasons, the brain has devoted a consider-
able amount of neural resources to motor control. Two
brain areas that are known to be very important for mo-
tor control are the basal ganglia and the cerebellum .
The basal ganglia starts just below the medial area of the
cortex surrounding the third ventricle, and comprises
a set of interconnected brain areas including the cau-
date and putamen (which are collectively known as
the striatum ), globus pallidus , substantia nigra , sub-
thalamic nucleus ,and nucleus accumbens . The cere-
bellum is the large structure that looks like cauliflower
at the base of the brain, and is composed of the lateral
cerebellar hemispheres and more medial nuclei.
Because of the cognitive focus of this topic, we will
largely ignore the motor functions of the basal ganglia
and cerebellum. However, it has recently become popu-
lar to emphasize the cognitive roles of these areas (e.g.,
Brown & Marsden, 1990; Brown, Schneider, & Lidsky,
1997; Gao, Parsons, & Fox, 1996), and numerous com-
putational models have been developed (e.g., Beiser,
Hua, & Houk, 1997; Wickens, 1997). We acknowledge
that these areas do probably make an important cogni-
tive contribution, but it is likely to be of a somewhat
ancillary or specialized nature that would complement,
but not replace, the cortically mediated processing that
is the focus of this topic.
In keeping with the emphasis in this chapter on func-
tional tradeoffs for understanding brain area specializa-
tions, we offer the following idea as to what the cog-
nitive role of the basal ganglia might be, and how this