Biology Reference
In-Depth Information
Entorhinal cortex also participates in path integration through a reciprocal connection
with the subiculum (McNaughton
et al.
, 1996; Sharp, 1999). Entorhinal grid cells (Hafting
et
al.
, 2005; Sargolini
et al.
, 2006) are accompanied by head direction cells and grid by direction
cells in the entorhinal cortex (Hafting
et al.
, 2005; Sargolini
et al.
, 2006), suggesting that
entorhinal cortex shares similar path integration functions with the anatomically-adjacent
subiculum (McNaughton
et al.
, 2006).
In path integration, the updated information is a continuous variable representing position
or head direction. A continuum of cell assemblies, or a continuous attractor (Amari, 1977;
Droulez & Berthoz, 1991; Tsodyks, 1999; Stringer
et al.
, 2004), is therefore required to
encode position or head direction. In such an attractor the strength of the excitatory
connections between two cells could decrease with the distance between their respective
preferred directions (Redish
et al.
, 1996; Zhang, 1996), which would result in a focused
activity related to a particular direction (McNaughton
et al.
, 2006). A recurrent synaptic
matrix with such architecture will ensure the strength of the excitatory connections between
two cells decreases in proportion to the physical distance between the cells' respective place
fields.
Cells in such a direction-specific network will encode, conjointly, the rat's position
and velocity vectors (McNaughton
et al.
, 1996; Zhang, 1996; Samsonovich & McNaughton,
1997), therefore, they would combine head direction and running speed inputs with location
information from the attractor layer.
Finally, path integration research reveals that hippocampal place fields represent the
changes of current location, environmental context, current and recent environmental sensory
stimuli under the continuous reference of the idiothetic experience. Network connectivity
alterations for path integration obey mechanisms of experience-dependent synaptic plasticity.
As these mechanisms are to certain degree universal for all brain regions, a useful approach in
understanding hippocampal learning is to compare medial temporal lobe with other regions
known to undergo experience-dependent learning processes.
8.
P
LASTICITY IN
O
THER
S
YSTEMS
8.1. Cerebellar synaptic plasticity
The two main memory research directions, one involving declarative memory and the
other - procedural memory reveal plasticity rules common for both explicit and implicit
learning. Therefore comparison between the mechanisms that underline experience-dependent
plasticity in both memory systems will give us better understanding of how brain networks
are modified through experience. A central problem in the procedural memory studies is to
demonstrate, both experimentally and theoretically, how neuronal networks of the cerebellum
undergo synaptic plasticity after error-driven, LTD-based learning (Ito, 2001). Modern
control system theories have been useful in accurately defining roles played by a
microcomplex in motor control. In the usual design of a control system, precise control is
secured by feedback (Fig 9). A two-degrees-of-freedom adaptive control system for voluntary
movement proposed to combine feedback control by the cerebral cortex with feed-forward
control by the cerebellum (Kawato
et al.
, 1987; Gomi & Kawato, 1992). In the cerebellum,
there is a regularly organized circuit that delivers relatively unprocessed somatosensory and
