Biology Reference
In-Depth Information
(Rawlins, 1985; Kesner, 1998) and also CA3 could bridge the temporal gap required for
hippocampus-dependent temporal associations. The learning process therefore requires CA3
to hold one item active in continuing attractor state until the next item in the sequence arrives
and then when both items are associated by temporally asymmetric synaptic plasticity (Rolls
& Kesner, 2006).
Figure 7. Phase precession of hippocampal place cells and sequences learning. Diagrammatic
illustration of how phase precession occurs as an animal moves through the place field on a well-known
path. On each successive theta cycle, firing occurs with an earlier phase, until the other end of the place
field is reached. Each position (1 to 4, marked with curved solid lines) is defined by the most active cell
assembly firing at each theta cycle (e.g., position 1 by the red assembly). The phase advance is marked
by the curved, dotted arrows. The width of the bars indicates firing rates of the hypothesized assemblies
while the theta time scale temporal differences between assemblies reflect distances of their spatial
representations. Because each assembly contributes to multiple place representations, multiple
assemblies are coactivated in each theta cycle. As a result, the current position is represented by the
maximally active assembly at the cycle trough in CA1.
Assembly sequences within theta cycles could reflect strengthening of connections between adjacent
places (e.g., position 2 - position 3). Cells encoding different items will fire with a temporal separation
of one or several gamma cycles, a time that is within the window of NMDA-dependent LTP. Thus
synaptic modification will occur at recurrent CA3 synapses that connect cells encoding sequential
memory items. For example the place cells indicated with blue and green color will undergo synaptic
potentiation (marked with black curved arrow) in the direction from blue to green as spike-timing
plasticity rules postulate. In the CA3 recurrent system, the temporal differences among assembly
members are assumed to be reflecting synaptic strengths between assembly members, resulting in a
series of related phase precessions (e.g., the phase precession of the blue-marked cell will be followed
by the phase precession of the green one). Similarly CA3 cells which fire earlier than CA1 cells for a
given place field will strengthen the CA3-CA1 connections (marked with straight colored arrows). By
this way is ensured a constant control between the predicted by the CA3 position and updated by the
entorhinal cortex positions. This mechanism may allow distances to be translated into time and time
into synaptic weights [adapted from (Dragoi & Buzsáki, 2006)].
Search WWH ::




Custom Search