Memory consolidation

Memory consolidation is a category of processes that stabilize a memory trace after the initial acquisition. Consolidation is distinguished into two specific processes, synaptic consolidation, which occurs within the first few hours after learning, and system consolidation, where hippocampus-dependent memories become independent of the hippocampus over a period of weeks to years. Recently, a third process has become the focus of research, reconsolidation, in which previously consolidated memories can be made labile again through reactivation of the memory trace.

History
Memory consolidation was first referred to in the writings of the renowned Roman teacher of rhetoric Quintillian. He noted the “curious fact… that the interval of a single night will greatly increase the strength of the memory,” and presented the possibility that “… the power of recollection .. undergoes a process of ripening and maturing during the time which intervenes.” The process of consolidation was later proposed based on clinical data illustrated in 1882 by Ribot’s Law of Regression, “progressive destruction advances progressively from the unstable to the stable”. This idea was elaborated on by William H. Burnham a few years later in a paper on amnesia integrating findings from experimental psychology and neurology. Coining of the term “consolidation” is credited to the German researchers Müller and Alfons Pilzecker who rediscovered the concept that memory takes time to fixate or undergo “Konsolidierung” in their studies conducted between 1892 and 1900.

Systematic studies of retrograde amnesia started to emerge in the 1960s and 1970s. These were accompanied by the creation of animal models of human amnesia in an effort to identify brain substrates critical for slow consolidation. Meanwhile, neuropharmacological studies of selected brain areas began to shed light on the molecules possibly responsible for fast consolidation. In recent decades, advancements in cellular preparations, molecular biology, and neurogenetics have revolutionized the study of consolidation.

Synaptic Consolidation
Synaptic consolidation is one form of memory consolidation seen across all species and long-term memory tasks. Long-term memory, when discussed in the arena of synaptic consolidation, is memory that lasts for at least 24 hours. An exception to this 24-hour rule is long-term potentiation, or LTP, a model of synaptic plasticity related to learning, in which an hour is thought to be sufficient. Synaptic consolidation is achieved faster than systems consolidation, within only minutes to hours of learning. LTP, one of the best understood forms of synaptic plasticity, is thought to be a possible underlying process in synaptic consolidation.

Standard Model
The standard model of synaptic consolidation suggests that alterations of synaptic protein synthesis and changes in membrane potential are achieved through activating intracellular transduction cascades. These molecular cascades trigger transcription factors that lead to changes in gene expression. The result of the gene expression is the lasting alteration of synaptic proteins, as well as synaptic remodeling and growth. In a short time-frame immediately following learning, the molecular cascade, expression and process of both transcription factors and immediate early genes, are susceptible to disruptions. Disruptions caused by specific drugs, antibodies and gross physical trauma can block the effects of synaptic consolidation.

Long-term Potentiation
LTP can be thought of as the prolonged strengthening of synaptic transmission, and is known to produce increases in the neurotransmitter production and receptor sensitivity, lasting minutes to even days. The process of LTP is regarded as a contributing factor to synaptic plasticity and in the growth of synaptic strength, which are suggested to underlie memory formation. LTP is also considered to be an important mechanism in terms of maintaining memories within brain regions, and therefore is thought to be involved in learning. There is compelling evidence that LTP is critical for Pavlovian fear conditioning in rats suggesting that it mediates learning and memory in mammals. Specifically, NMDA-receptor antagonists appear to block the induction of both LTP and fear conditioning and that fear conditioning increases amygdaloidal synaptic transmission that would result in LTP.

Timeline of Consolidation
Synaptic consolidation, when compared to systems consolidation (which is said to take weeks to months to years to be accomplished), is considerably faster. There is evidence to suggest that synaptic consolidation takes place within minutes to hours of memory encoding or learning, and as such is considered the ‘fast’ type of consolidation. As soon as six hours after training, memories become impervious to interferences that disrupt synaptic consolidation and the formation of long-term memory.

Spacing Effect
Distributed learning has been found to enhance memory consolidation, specifically for relational memory. Experimental results suggest that distributing learning over the course of 24 hours decreases the rate of forgetting compared to massed learning, and enhances relational memory consolidation. When interpreted in the context of synaptic consolidation, mechanisms of synaptic strengthening may depend on the spacing of memory reactivation to allow sufficient time for protein synthesis to occur, and thereby strengthen long-term memory.

Protein Synthesis
Protein synthesis has been suggested to play a critical role in the formation of new memories. Studies have shown that protein synthesis inhibitors administered after learning, weaken memory, suggesting that protein synthesis is required for memory consolidation. Additionally, reports have suggested that the effects of protein synthesis inhibitors also inhibit LTP. However, it should be noted that other results have shown that protein synthesis may not in fact be necessary for memory consolidation, as it has been found that the formation of memories can withstand vast amounts of protein synthesis inhibition, suggesting that this criterion of protein synthesis as necessary for memory consolidation is not unconditional.

Dietary Flavanoids
There is evidence to suggest that dietary flavanoids have effects on encouraging LTP and synaptic plasticity, therefore affecting memory. Specifically, it was found that dietary-derived flavanoids might protect neurons, enhance neuronal function, and stimulate neuronal regeneration. Additionally, these dietary phytochemicals interact with several neuronal signaling cascade pathways that are responsible for alterations in LTP, and consequently, learning and human memory. Flavanoids may trigger certain events, including the activation of the CREB transcription factor, which is important to the enhancement of short-term and long-term memory. This activation then triggers the synthesis of important proteins related to LTP, ultimately leading to synapse growth and eventually long-term memory.

System Consolidation
System Consolidation is the second form of memory consolidation. It is a reorganization process in which memories from the hippocampal region where memories are first encoded are moved to the neo-cortex in a more permanent form of storage. System consolidation is a slow dynamic process that can take from one to two decades to be fully formed in humans, unlike synaptic consolidation that only takes minutes to hours for new information to stabilize into memories.

Standard Model
The Standard model of systems consolidation has been summarized by Squire and Alvarez (1995) ; it states that when novel information is originally encoded and registered, memory of these new stimuli becomes retained in both the hippocampus and cortical regions. Later the hippocampus’ representations of this information become active in explicit (conscious) recall or implicit (unconscious) recall like in sleep and ‘offline’ processes.

Memory is retained in the hippocampus for up to one week after initial learning, representing the hippocampus-dependent stage. During this stage the hippocampus is ‘teaching’ the cortex more and more about the information and when the information is recalled it strengthens the cortico-cortical connection thus making the memory hippocampus-independent. Therefore from one week and beyond the initial training experience, the memory is slowly transferred to the neo-cortex where it becomes permanently stored.

Semantic vs. Episodic Memory
Nadel and Moscovitch argued that when studying the structures and systems involved in memory consolidation, semantic memory and episodic memory need to be treated as different types of memory. This additional distinction expands the Standard Model by Frankland, which does not consider the types of memory as separate. Evidence from extensive neuro-imaging research on the different function of cortical and hippocampus memory traces have found that the hippocampus provides temporal and spatial context, whereas the cortical traces are primarily context-free. Episodic memory, no matter whether new or old, relies on hippocampus-cortical networks whereas remote semantic memories can be retrieved independent of the hippocampus.

Multiple Trace Theory
Multiple Trace Theory (MTT) builds on the distinction between semantic memory and episodic memory, arguing that semantic memories are created by multiple traces left in the neo-cortex in the process of consolidation, separate from the hippocampus. Hence, while proper hippocampus functioning is necessary for the retention and retrieval of episodic memories, it is less necessary for semantic memories.

REM Sleep
Rapid eye movement (REM) sleep has been implicated in the overnight learning in humans by the re-organization of novel information in the hippocampal and cortical regions of the brain. REM sleep elicits an increase in neuronal activity following an enriched or novel waking experience, thus increasing neuronal plasticity and therefore playing an essential role in the consolidation of memories. In particular studies have been done on sensory and motor related tasks. In one study testing finger-tapping, people were split into two groups and tested post-training with or without intervening sleep; results concluded that sleep post-training increases both speed and accuracy in this particular task, while increasing the activation of both cortical and hippocampal regions; whereas the post-training awake group had no such improvements.

Zif268 & REM Sleep
Zif268 is an Immediate Early Gene (IEG) thought to be involved in neuroplasticity by an up-regulation of the transcription factor during REM sleep after pre-exposure to an enriched environment. Results from studies testing the effects of zif268 on mice brains postmortem, suggest that a waking experience prior to sleep can have an enduring effect in the brain, due to an increase of neuroplasticity.

Reconsolidation
Memory reconsolidation is the process of previously consolidated memories being recalled and actively consolidated. It is a distinct process that serves to maintain, strengthen and modify memories that are already stored in the long-term memory. Once memories undergo the process of consolidation and become part of long-term memory, they are thought of as stable. However, the retrieval of a memory trace can cause another labile phase that then requires an active process to make the memory stable after retrieval is complete. It is believed that post-retrieval stabilization is different and distinct from consolidation, despite its overlap in function (e.g. storage) and its mechanisms (e.g. protein synthesis). Memory modification needs to be demonstrated in the retrieval in order for this independent process to be valid.

History
The theory of reconsolidation has been debated for many years and has become quite controversial. Reconsolidation was first conceptualized after studies were done on elimination of phobias with electroconvulsive shock therapy; the disruption of the consolidated fear memory after shock administration led to further investigation into the concept. In many early studies electroconvulsive shock therapy was used to test for reconsolidation, as it was a known amnesic agent, and lead to memory loss if administered directly after the retrieval of a memory. Later research using Pavlovian fear conditioning on rats found that a consolidated fear memory can return to a labile state, with immediate amygdala infusions of the protein synthesis inhibitor anisomycin, but not infusions made six hours afterwards. It was concluded that consolidated fear memory, when reactivated, enters a changeable state that requires de novo protein synthesis for new consolidation or reconsolidation of the old memory. Since these break through studies many more have been done testing the theory of reconsolidation. Studies have been done on numerous subjects including; crabs, chicks, honeybees, medaka fish, lymnaea, humans and rodents.

Criticisms
Some studies have supported this theory, while others have failed to demonstrate disruption of consolidated memory after retrieval. It is important to note that negative results may be examples of conditions where memories are not susceptible to a permanent disruption, thus a determining factor of reconsolidation. After much debate and a detailed review of this field it had been concluded that reconsolidation was a real phenomenon. More recently Tronson and Taylor compiled a lengthy summary of multiple reconsolidation studies, noting a number of studies were unable to show memory impairments due to blocked reconsolidation. However the need for standardized methods was underscored as in some learning tasks such as fear conditioning, certain forms of memory reactivation could actually represent new extinction learning rather than activation of an old memory trace. Under this possibility, traditional disruptions of reconsolidation might actually maintain the original memory trace but preventing the consolidation of extinction learning.

Reconsolidation experiments are more difficult to run than typical consolidation experiments as disruption of a previously consolidated memory must be shown to be specific to the reactivation of the original memory trace. Furthermore, it is important to demonstrate that the vulnerability of reactivation occurs in a limited time frame, which can be assessed by delaying infusion till six hours after reactivation. It is also useful to show that the behavioral measure used to assess disruption of memory is not just due to task impairment caused by the procedure, which can be demonstrated by testing control groups in absence of the original learning. Finally, it is important to rule out alternative explanations, such as extinction learning by lengthening the reactivation phase.

Distinctions from Consolidation
Questions arose if reconsolidation was a unique process or merely another phase of consolidation. Both consolidation and reconsolidation can be disrupted by pharmacological agents (e.g. the protein synthesis inhibitor anisomycin) and both require the transcription factor CREB. However, recent amygdala research suggests that BDNF is required for consolidation (but not reconsolidation) whereas the transcription factor and immediate early gene Zif268 is required for reconsolidation but not consolidation. A similar double dissociation between Zif268 for reconsolidation and BDNF for consolidation was found in the hippocampus for fear conditioning. However not all memory tasks show this double dissociation, such as object recognition memory.