Biology Reference
In-Depth Information
Place fields have been shown to become more selective as rats learn a maze (Mizumori &
Kalyani, 1997). Place cells can gradually acquire diverging responses to two sets of similar
but non-identical stimuli, a type of discriminative responding that is commonly known as
“pattern separation” (Bostock
et al.
, 1991; Lever
et al.
, 2002; Hayman
et al.
, 2003). Over
time, the place system considers the two “similar” environments as different and as a result a
divergence of firing patterns in both environments occurs (Fig 3). Such experience-dependent
change in responsiveness to environmental stimuli strongly implies experience-dependent
synaptic plasticity of the place cell connections within the sequence-encoding network. LTP
and LTD of the connections between the contextual inputs are proposed to mediate the long-
term, experience-dependent divergence of place cell representations (Jeffery & Hayman,
2004). The coactivity of contextual inputs that are specific to that environment and the place
cells is able to potentiate with time only those connections, which precisely represent the
environment (Barry
et al.
, 2006). This idea is supported by the ability of place cells to acquire
discriminations between closely similar environments that previously were undiscriminated,
(Bostock
et al.
, 1991; Lever
et al.
, 2002; Hayman
et al.
, 2003; Jeffery & Anderson, 2003).
Discrimination of similar environments appears often to involve the loss of a field in one or
other of the two environments, sometimes with development of a new field (Lever
et al.
,
2002; Wills
et al.
, 2005). The observed phenomenon can be explained with a weakening of
the link between inactive contextual inputs and the field-specifying inputs, by heterosynaptic
LTD, so that this cell comes to be driven only by the relevant contextual elements (Abraham
& Bear, 1996; Fazeli & Collingridge, 1996; Rolls & Deco, 2002). Additionally synaptic
scaling models propose that LTP and LTD are always balanced, suggesting how place cells
can learn to discriminate two environments (Turrigiano & Nelson, 2000): as inputs from the
discriminative stimuli gradually increase in strength, the scaling process weakens the original
inputs so that they are no longer able to induce complex-spiking of the cell (Jeffery &
Hayman, 2004).
The anatomical candidate for pattern separation appears to be the CA3 hippocampal
region. CA3 place cells are able to maintain distinct representations of two visually identical
environments, and selectively reactivate either one of the representation patterns depending
on the experience (Tanila, 1999). When rats experienced a completely different environment,
CA3 place cells developed orthogonal representations of those different environments by
changing their firing rates between the two environments, whereas CA1 place cells
maintained similar responses (Leutgeb
et al.
, 2004).
3.4. Goal- and directionality-related plasticity of place fields
Not only sensory but also motivational factors can induce a reorganization of place fields
(Mizumori, 2006). Changing the reward location within a single session can induce dramatic
place field reorganization (Smith & Mizumori, 2006b, 2006a) (Fig 4). Units in the
hippocampus have been shown to fire not only in relation to spatial location but also in
relation with the different demands of the task (Eichenbaum
et al.
, 1999; Deadwyler &
Hampson, 2004). Recording in a familiar room can induce remapping of the place fields
toward new goal locations (Hollup
et al.
, 2001; Lenck-Santini
et al.
, 2001; Lenck-Santini
et
al.
, 2002). The learning of rewarded locations in the environment can induce strengthening of
