Biomedical Engineering Reference
In-Depth Information
of information between the cerebral cortex, the basal ganglia/thalamus, and the cere-
bellum. A growing body of evidence has been accumulated in recent years that
challenges the conventional view of segregated processing of perceptual, motor and
cognitive information [61]. For example, it was usually considered that basal gan-
glia and cerebellum were specialized for motor control and different cortical areas
were devoted to specific functionalities, with a clear separation of sensory, motor
and cognitive areas. This is not anymore the conventional wisdom and the emerg-
ing picture is that the three main computational sites for adaptive behaviour are all
concerned with processing sensorimotor patterns in a cognitive-sensitive way but are
specialized as regards the learning paradigms and the types of representation:
1. The
cerebral cortex
appears to be characterized by a process of unsupervised
learning that affects its basic computational modules (the micro-columns that
are known to have massive recurrent connections). The function of these com-
putations might be the representation of non-linear manifolds, such as a
body
schema
in the posterior parietal cortex [47]. This view seems to be contra-
dicted by the
fractured shape
of cortical maps [58], but the apparent contra-
diction may only be a side effect of the basic paradox faced by the cortical
areas: how to fit higher dimensional manifolds (such as a proprioceptive body
schema) onto a physically flat surface;
2. The
Cerebellum
is plausibly specialized in the kind of
supervised learning
exemplified by the feedback error learning model. Moreover, the cerebellar
hardware (characterised by a large number of micro-zones, comprising mossy-
fiber input, bundles of parallel fibers, output Purkinje cells, with teaching sig-
nals via climbing fibers) is well designed for the representation of time series,
according to a sequence-in sequence-out type of operation [12];
3. The
Basal ganglia
are known to be involved in events of reinforcement learn-
ing that are required for the representation of goal-directed sequential be-
haviour [63].
17.6.3
A distributed computational architecture
Figure 17.19
summarizes some of the points outlined in the previous subsection. It
must be emphasized that in spite of its apparent simplicity the underlying model
is extremely complex from many points of view: (i) it is non-linear; (ii) it involves
high-dimensional variables; (iii) it has a coupled dynamics, with internal and external
processes; (iv) it is adaptive, with concurrent learning processes of different types.
No simulation model of this complexity has been constructed so far, also because the
mathematical tools for dominating its design are only partially available. However,
there is a need for improving our current level of understanding in this direction be-
cause this is the only sensible way for interpreting the exponentially growing mass
of data coming from new measurement techniques, such as advanced brain imag-
ing. As a matter of fact, since the time of Marey better measurement techniques of
movement analysis require better and better models of motor control, and vice versa.
Search WWH ::
Custom Search