Biology Reference
In-Depth Information
(i) Stimulus-action coupling
In 1949 Hebb proposed that development of the PFC is particularly important for
developing schemas to solve problems that will be encountered later in life (Hebb 1949).
These are learnt during the relatively late period of development of the PFC, with respect to
other brain structures, and contribute to appropriate behavioural responses that, in humans,
are learnt during early adulthood. Through its anatomical connections with other regions,
including inputs from other cortical regions pertaining to visual, tactile and olfactory cues, as
well as information about the internal milieu of the organism carried from subcortical
structures, the PFC is able integrate an array of information to develop such schemas. This
information can be used to plan the sequence of forthcoming actions according to the current
sensory context and internal state (Shallice 1982; Kolb 1984; Robbins 1996; Shallice and
Burgess 1996; Floresco et al. 1997).
Neurons in the primate dorsolateral PFC increase their firing during the delay period in a
delayed response task, a model for working memory (Fuster and Alexander 1971; Kubota and
Niki 1971; Kojima and Goldman-Rakic 1982). While some neurons fire at the beginning of
the delay period, others fire as the delay progresses (Fuster 1995). The neurons that fire
during and immediately after the cue have been attributed with encoding recent perceptual
stimuli. These neurons are thought to then project to a neighbouring group of neurons that fire
during the delay, encoding projections towards a future action. The latter group only display
enhanced firing following learning of a particular delay-related task (Fuster 1973). These
findings led Fuster to suggest that these neurons form a cortical network that encodes a
stimulus-action memory (Fuster et al. 2000). In keeping with this, during learning of a spatial
navigation task in rats, increased correlated firing was observed in the prelimbic mPFC, and
this firing persisted after learning (Baeg et al. 2007).
As a continuation of the seminal studies by Fuster, Miller and Cohen suggested that the
PFC provides a “bias” signal to the perception-action cycle (Miller and Cohen 2001).
Subsequently, Otani proposed that the mPFC behaves as a “cognitive switch”, coupling a
particular set of stimuli with a particular set of actions, and proposed that synaptic plasticity
in the mPFC is the neural trace underlying the permanent storage of these rules (Otani 2002).
Consistent with this, during repeated training of working memory tasks, an improvement in
performance is observed (McNamara and Scott 2001). This is thought to be mediated by
synaptic plasticity in the mPFC. For example blockade of protein synthesis by infusion of
anisomycin in the mPFC impairs the improvement in learning (Touzani et al. 2007). In this
way declarative memories (involving the hippocampus) are linked with procedural memories
(involving the striatum), thereby contributing to systematic behavioural patterns. Thus
plasticity in the PFC is involved in the laying down of a repertoire of actions that have been
learnt as appropriate ways to respond to a particular set of cues, in a particular behavioural
context. In humans, improvements in working memory with repeated tasks may be of
relevance for learning strategies to deal with complex behaviours.
(ii) Short-term memory
While the cellular basis of working memory is the persistent firing of neurons in the
mPFC over delays up to 20 seconds, delays of minutes in these tasks require short-term
