Abstract
Long-term potentiation (LTP) is a persistent strengthening of synaptic transmission that follows brief, high-frequency stimulation, and it is the leading cellular model of how the brain stores memory. First described by Bliss and Lomo in the rabbit hippocampus in 1973, LTP is Hebbian: it strengthens exactly those connections that are active when the postsynaptic cell fires. This article states its defining properties — input specificity, cooperativity, and associativity — and traces the mechanism from induction to expression: the NMDA receptor acting as a molecular coincidence detector, calcium entry triggering CaMKII, and the insertion of AMPA receptors that strengthens the synapse. It distinguishes short-lived early LTP from protein-synthesis-dependent late LTP, explains synaptic tagging, and reviews the evidence tying LTP to learning. Three interactive demonstrations let the reader operate the coincidence detector, test Hebbian specificity, and follow the time course of potentiation.
Keywords: long-term potentiation, synaptic plasticity, NMDA receptor, Hebbian learning, hippocampus
Long-term potentiation is the long-lasting increase in the strength of a synapse that follows a brief burst of activity across it. Deliver a short train of high-frequency stimulation to a bundle of axons, and the response their target neurons give to a single test pulse grows larger and stays larger — for hours in a slice of tissue, for days or weeks in a living animal (Bliss & Lomo, 1973). That a synapse can record its own recent use, and hold that record, is the property any physical theory of memory needs, and it is why LTP has been the dominant cellular model of learning for half a century (Bliss & Collingridge, 1993).
The idea that memory lives in changeable synapses is older than the finding. Donald Hebb's 1949 postulate — that when one cell repeatedly helps to fire another, the connection between them grows stronger — gave the field its guiding hypothesis decades before anyone could test it. LTP is, in large part, the discovery that the brain actually obeys a Hebbian rule: it strengthens the connections that were active at the moment the receiving cell was driven to fire, and leaves the inactive ones alone (Nicoll, 2017). The rest of this article is an account of how a synapse manages to follow that rule, and of the evidence that following it is what learning is made of.
- Long-term potentiation is a persistent, activity-dependent strengthening of synaptic transmission, first described in the hippocampus by Bliss and Lomo in 1973, and the leading cellular model of memory storage.
- Its three defining properties — input specificity, cooperativity, and associativity — make it a Hebbian mechanism: it strengthens exactly the inputs active when the postsynaptic cell is strongly depolarized.
- Induction depends on the NMDA receptor, which opens only when glutamate binding and postsynaptic depolarization coincide, letting it act as a molecular coincidence detector that admits calcium.
- Expression is largely postsynaptic: calcium activates CaMKII, which drives AMPA receptors into the synapse and converts silent synapses into functional ones.
- Short-lived early LTP does not require new protein; long-lasting late LTP does, and synaptic tagging explains how newly made proteins are captured by the specific synapses that were active.
What Long-Term Potentiation Is
Long-term potentiation is defined operationally, by what is done and what results. An experimenter records the excitatory postsynaptic potential (EPSP) that a test stimulus evokes at a synapse, establishes a stable baseline, then delivers a brief tetanus — a high-frequency train, classically 100 pulses at 100 Hz. When the same test stimulus is resumed, the EPSP is larger and remains larger long after the tetanus ends. The magnitude is reported as the percentage increase in EPSP slope over baseline, and the defining feature is its persistence: minutes at the shortest, and in the intact animal days to weeks (Bliss & Lomo, 1973). The original 1973 recordings were made in the dentate gyrus of the rabbit hippocampus, and the hippocampal slice remains the workhorse preparation, though LTP is now known throughout the brain.
Figure 1
The Canonical Long-Term Potentiation Time Course
Three properties, established early, are what make LTP a plausible memory mechanism rather than a curiosity. The first is input specificity: only the synapses that received the tetanus are strengthened, while other synapses onto the very same neuron are unchanged. A neuron can therefore potentiate one set of inputs without indiscriminately potentiating all of them, which is the storage capacity any useful memory device requires. The second is cooperativity: a stimulus must recruit enough axons together to cross an intensity threshold — a single weak input will not induce LTP, but many acting at once will. The third is associativity: a weak input that alone is below threshold will be potentiated if it is active at the same time as a strong input onto the same cell, so that the weak pathway becomes associated with the strong one (Bliss & Collingridge, 1993).
These three properties are the cellular expression of Hebb's postulate. Associativity in particular — a weak input strengthened because it fired in company with a strong one — is almost a restatement of the rule that cells which fire together wire together. What remained, once the properties were catalogued, was to find the molecule that reads the coincidence of presynaptic activity and postsynaptic depolarization. That molecule is the NMDA receptor, and it is the subject of the next section. Table 1 summarizes the defining properties and their functional significance.
| Property | What it means | Why it matters |
|---|---|---|
| Input specificity | Only the synapses that were active during induction are strengthened; other synapses on the same neuron are unaffected | Gives the neuron the storage capacity to hold many independent memories rather than one global setting |
| Cooperativity | Induction requires enough afferents active together to cross an intensity threshold; one weak input is insufficient | Sets a threshold that prevents trivial or noisy activity from writing to memory |
| Associativity | A weak input is potentiated if it is active at the same time as a strong input onto the same cell | Implements Hebbian association — the cellular basis of linking co-occurring events |
| Persistence | The strengthening outlasts the inducing stimulus by hours in vitro and days to weeks in vivo | Provides the durable trace a memory mechanism must leave |
Induction: The NMDA Receptor as Coincidence Detector
The central mechanistic fact about the commonest form of LTP is that its induction requires the NMDA subtype of glutamate receptor. Graham Collingridge showed that blocking NMDA receptors with the antagonist AP5 prevents LTP in hippocampal area CA1 while leaving ordinary synaptic transmission intact — proof that the receptor is needed to induce the change, not to carry the everyday signal (Bliss & Collingridge, 1993). That dissociation is the key to the whole mechanism, because it separates the molecule that detects the condition for plasticity from the molecules that do the routine work of transmission.
The NMDA receptor detects coincidence because it answers to two conditions at once. Like other glutamate receptors it opens only when the presynaptic terminal releases glutamate and glutamate binds it. But even with glutamate bound, its channel is physically plugged at resting membrane potential by a magnesium ion sitting in the pore. Only when the postsynaptic membrane is also depolarized — driven positive by strong activity at other inputs — is the magnesium expelled and the channel cleared. The receptor therefore passes current only when presynaptic release and postsynaptic depolarization occur together, which is exactly the coincidence Hebb's rule demands. It is, in the standard phrase, a molecular coincidence detector, and it is what makes LTP associative: the weak input supplies the glutamate, the strong input supplies the depolarization, and only the synapse where both meet is marked (Luscher & Malenka, 2012).
What passes through the cleared channel matters as much as that it opens. The NMDA receptor is unusually permeable to calcium, and it is the resulting rise in postsynaptic calcium — sharp, local, and confined to the activated spine — that is the actual trigger for potentiation. A large, brief calcium transient sets in motion the enzymes that strengthen the synapse; a smaller or more prolonged rise triggers the opposite change, weakening it (Malenka & Bear, 2004). Calcium concentration, in other words, is the variable the synapse reads to decide which way to move.
The NMDA coincidence detector
The NMDA receptor answers to two conditions at once. Toggle presynaptic glutamate and postsynaptic depolarization independently: calcium enters only when both are present.
No glutamate and no depolarization: the channel is shut and magnesium plugs the pore.
Figure. Schematic of the NMDA receptor as a two-input logical AND on glutamate binding and membrane depolarization. Illustrative, not to scale; computed locally, not stored. Original illustration.
Expression: AMPA Receptors and Silent Synapses
Induction and expression are separable questions: what triggers LTP is not the same as what sustains the larger response. The calcium that enters through the NMDA receptor activates a cascade of kinases, foremost among them calcium/calmodulin-dependent protein kinase II (CaMKII), which is both necessary for LTP and, once switched on, able to remain active — a candidate molecular switch that holds its state after the calcium has gone (Herring & Nicoll, 2016). The question CaMKII activation then poses is: what does the synapse actually change to become stronger?
The answer, established over two decades of work, is that expression is largely postsynaptic and consists chiefly in the trafficking of AMPA receptors — the glutamate receptors that carry the bulk of fast excitatory transmission. LTP drives additional AMPA receptors into the postsynaptic membrane, so that the same amount of released glutamate produces a larger response (Diering & Huganir, 2018). The most direct evidence comes from silent synapses: connections that contain NMDA receptors but no functional AMPA receptors, and so pass no current at resting potential — they are electrically silent. LTP unsilences them, inserting AMPA receptors where there were none and converting a mute contact into a working one (Herring & Nicoll, 2016). Roger Nicoll's synthesis of this evidence made the postsynaptic insertion of AMPA receptors the consensus account of how LTP is expressed (Nicoll, 2017).
That the same brief calcium signal can move a synapse in either direction is the basis of bidirectional plasticity. Strong, brief calcium influx through NMDA receptors drives LTP by adding AMPA receptors; weaker, more sustained influx drives long-term depression (LTD) by removing them. LTP and LTD are thus mechanistic mirror images, two directions of a single AMPA-receptor-trafficking process, and a synapse's weight at any moment reflects the running balance between them (Citri & Malenka, 2008).
Cooperativity and associativity
Add coactive weak synapses, then try switching on a strong input onto the same neuron. LTP is induced at the weak synapses only when the membrane crosses the −40 mV threshold — from many weak inputs together (cooperativity) or by borrowing a strong input’s depolarization (associativity).
The membrane sits at -62 mV. It is below the −40 mV threshold, so the magnesium block holds and no LTP is induced. Acting alone, at least 8 weak synapses are needed to cross it.
Figure. Membrane potential = −70 mV + 4 mV per weak synapse + 30 mV if the strong input is on; threshold −40 mV. Round illustrative values matching the Worked Example. Computed locally, not stored. Original illustration.
Early LTP, Late LTP, and Synaptic Tagging
LTP is not one phenomenon but at least two, distinguished by how long they last and what they require. Early LTP, produced by a single tetanus, lasts one to three hours and depends only on the modification of proteins already present — the AMPA-receptor trafficking of the previous section, driven by kinases acting on existing molecules. Late LTP, produced by stronger or repeated stimulation, lasts many hours to days and requires new gene expression and protein synthesis; block translation, and the potentiation still appears but decays back to baseline within a few hours, its early phase intact and its late phase abolished (Kandel, 2001). The persistence that makes LTP a memory mechanism is therefore a built, protein-dependent thing, not merely a lingering chemical change.
This division raises a problem of address. New proteins made in response to strong activity are synthesized in the cell body and distributed widely, yet late LTP retains the input specificity of early LTP — only the synapses that were active become durably strengthened, not every synapse on the neuron. How do cell-wide protein products reach only the correct synapses? The answer is synaptic tagging and capture, proposed by Uwe Frey and Richard Morris: an activated synapse sets a transient, local molecular tag, and the diffusely delivered plasticity proteins are captured only at tagged synapses, which is where they stabilize the change (Frey & Morris, 1997). The tag explains how a global signal produces a specific, lasting outcome, and it predicts a striking result — a weak event that would normally yield only early LTP can be converted to late LTP if a strong event occurs nearby in time, because the weak synapse's tag captures the proteins the strong event caused to be made.
The time course: early vs late LTP
Choose the induction protocol and whether protein synthesis is blocked, then read the EPSP slope over six hours. Late LTP’s persistence is a built, protein-dependent thing — block translation and only the early phase survives.
Repeated tetani produce late LTP: the slope decays toward a maintained plateau and is still 150% of baseline at six hours. The persistence that makes LTP a memory mechanism requires new protein.
Figure. Illustrative EPSP-slope kinetics: early LTP as exponential decay to baseline, late LTP as decay to a maintained plateau, with the protein-synthesis inhibitor collapsing late kinetics onto early. Schematic, not measured data; computed locally, not stored. Original illustration.
Long-Term Potentiation and Memory
The reason LTP commands the attention it does is the claim that it is the physical substrate of learning. That claim needs evidence of two kinds: that interfering with LTP interferes with memory, and that learning itself produces LTP. Both now exist. Richard Morris and colleagues provided the first when they showed that infusing the NMDA-receptor blocker AP5 into a rat's brain, at a dose that prevents hippocampal LTP, also prevents the animal from learning the location of a hidden platform in a water maze — a spatial task that depends on the hippocampus (Morris et al., 1986). Blocking the synaptic mechanism blocked the learning; the correspondence was the strongest link yet between a cellular process and a behavioral one, and it made LTP central to the study of spatial learning and memory consolidation.
The converse evidence — that learning drives LTP — was harder and came later. If experience uses LTP to store memories, then an animal that has just learned something should show, at the relevant synapses, the same changes that artificial tetanus produces. Whitlock and colleagues demonstrated exactly this: a single training episode in an avoidance task produced LTP-like potentiation at hippocampal synapses, and occluded further LTP there, as if learning had already used up the mechanism (Whitlock et al., 2006). Together the two lines of evidence make LTP not merely analogous to memory but, at least in the hippocampus, a mechanism learning actually recruits — the cellular counterpart of forming an episodic or spatial memory (Kandel, 2001).
Worked Example
Consider how cooperativity and associativity fall out of the NMDA receptor's voltage dependence, using round numbers. Suppose the postsynaptic neuron rests at −70 mV, and that substantial relief of the magnesium block — enough NMDA current to induce LTP — requires the membrane to be depolarized to about −40 mV, a swing of +30 mV. Suppose each coactive weak synapse contributes an EPSP of +4 mV at the postsynaptic site.
Cooperativity now follows arithmetically. To reach the −40 mV threshold from rest requires 30 ÷ 4 = 7.5, so at least 8 weak synapses must be active together. Two weak synapses acting alone depolarize the cell by only 2 × 4 = 8 mV, reaching just −62 mV — far below threshold, so the magnesium block stays in place, no calcium enters, and no LTP is induced however faithfully those two synapses fire. This is why a single weak input cannot potentiate itself: it cannot supply the depolarization its own NMDA receptors need.
Associativity is the same arithmetic seen from the other side. Let a strong input, active at the same moment, depolarize the cell by the full +30 mV on its own. The two weak synapses still release glutamate onto their NMDA receptors, but now the membrane is at −40 mV because of the strong input, the magnesium is expelled, calcium enters at the weak synapses, and they are potentiated — even though 8 mV of their own depolarization would never have sufficed. The weak pathway is strengthened precisely because it was active in company with the strong one. Cooperativity and associativity are thus not separate rules but two consequences of a single fact: the coincidence detector reads the total depolarization at the synapse, whatever its source.
Discussion
The deepest open question about LTP is also the oldest: exactly where the strengthened synapse changes. The account above is resolutely postsynaptic — AMPA receptors added on the receiving side — and for hippocampal CA1 LTP that account now commands broad agreement (Nicoll, 2017). But a long and unresolved debate concerned whether expression is instead, or also, presynaptic, with the potentiated terminal releasing more glutamate. The dispute was sharp for years, partly because different forms of LTP at different synapses genuinely differ, and the lesson that survived it is that LTP is a family, not a single mechanism: mossy-fiber LTP in the hippocampus is NMDA-receptor-independent and presynaptically expressed, quite unlike the canonical CA1 form this article has described (Malenka & Bear, 2004).
A second problem is the inferential gap between the slice and the memory. Most of what is known about LTP comes from artificial stimulation of tissue, and the tetanus that induces it — a hundred pulses a second — is nothing a synapse encounters in natural behavior. The demonstration that learning produces LTP-like changes (Whitlock et al., 2006) narrows that gap but does not close it, and the question of whether the LTP studied in vitro is quantitatively the same process that stores a natural memory remains genuinely open. A related worry is sufficiency: showing that LTP is necessary for learning, as the AP5 experiments do, is not showing that it is sufficient to create a memory, and the direct test — writing a specific memory by artificially inducing LTP at chosen synapses — has only recently become approachable with optogenetic tools (Nicoll, 2017). The closest demonstration to date is that of Nabavi and colleagues, who conditioned a fear memory in the rat, erased it by delivering a low-frequency protocol that induced LTD at the auditory inputs carrying the conditioned stimulus, and then reinstated the memory by potentiating those same inputs — optically driving a specific memory out of and back into existence by moving synaptic strength down and up (Nabavi et al., 2014).
Finally, LTP does not act alone. A synapse's strength reflects the running balance of potentiation and depression, and stability over a lifetime requires that this balance be held in check by homeostatic mechanisms that scale synaptic weights up or down to keep neurons in their operating range. How Hebbian plasticity, which is inherently destabilizing — strong synapses get stronger — coexists with the homeostatic control that prevents runaway strengthening is one of the central theoretical problems the field now works on (Citri & Malenka, 2008).
Current Directions
Contemporary work has widened the picture in two directions at once. The first is a broadening of the induction rule itself. The classical account pairs presynaptic input with postsynaptic depolarization on a timescale of milliseconds, but recent work in behaving animals has identified forms of plasticity that operate on the far longer timescale of behavior — seconds — in which a single dendritic calcium event, driven by input that arrived up to several seconds earlier, can potentiate synapses and rapidly establish a place field in the hippocampus. This behavioral-timescale synaptic plasticity does not fit the narrow coincidence window of the classical Hebbian rule and suggests the brain uses a richer family of learning rules than LTP alone (Magee & Grienberger, 2020).
The second direction is molecular and mechanistic: the recognition that AMPA-receptor number is set by a detailed code of receptor subunit composition, phosphorylation, and auxiliary proteins that together determine how receptors are trafficked, anchored, and removed. Understanding that code is turning the qualitative story of receptor insertion into a quantitative account of how much a synapse changes and for how long (Diering & Huganir, 2018). Both directions share a methodological engine — the ability to record and manipulate identified synapses in awake, behaving animals — which is steadily replacing the slice as the setting in which the relationship between synaptic plasticity and learning is tested (Magee & Grienberger, 2020).
Common Misconceptions
- LTP is memory.
- LTP is a synaptic mechanism that memory appears to use, not memory itself. The evidence shows it is necessary for certain kinds of learning and that learning recruits it, but a memory is a distributed pattern across many synapses, not a single potentiated connection (Whitlock et al., 2006).
- All LTP works through NMDA receptors.
- The canonical CA1 form does, but LTP is a family. Mossy-fiber LTP in the hippocampus is NMDA-receptor-independent and expressed presynaptically, so no single mechanism covers every case (Malenka & Bear, 2004).
- Potentiating a synapse means the presynaptic cell releases more transmitter.
- For the common CA1 form, expression is chiefly postsynaptic: the receiving side inserts more AMPA receptors so that the same released glutamate produces a larger response (Nicoll, 2017).
Glossary
- AMPA receptor.
- The glutamate receptor that carries most fast excitatory transmission; LTP strengthens a synapse largely by inserting more of these into the postsynaptic membrane.
- Associativity.
- The property whereby a weak input, itself below threshold, is potentiated if it is active at the same time as a strong input onto the same neuron; the cellular form of Hebbian association.
- CaMKII.
- Calcium/calmodulin-dependent protein kinase II; the enzyme activated by calcium entry during induction that drives AMPA-receptor trafficking and can remain active as a molecular switch.
- Coincidence detection.
- The NMDA receptor's ability to pass current only when presynaptic glutamate release and postsynaptic depolarization occur together, implementing the Hebbian condition for plasticity.
- Cooperativity.
- The requirement that enough afferents be active together to cross an intensity threshold for induction; a single weak input cannot induce LTP on its own.
- Dentate gyrus.
- The hippocampal region where Bliss and Lomo first recorded LTP in 1973; the target of the perforant-path input from entorhinal cortex.
- Early LTP.
- The initial one-to-three-hour phase of potentiation, produced by a single tetanus, that modifies existing proteins and does not require new protein synthesis.
- EPSP.
- Excitatory postsynaptic potential; the depolarizing response a synapse produces in its target cell, whose slope or amplitude is the standard measure of synaptic strength in LTP experiments.
- Hebbian plasticity.
- Change in synaptic strength governed by Hebb's rule that a connection strengthens when the presynaptic cell repeatedly helps to fire the postsynaptic cell; LTP is its leading physical realization.
- Hippocampus.
- The medial-temporal-lobe structure central to memory in which LTP was discovered and is most studied; its CA1 region is the source of the canonical NMDA-receptor-dependent form.
- Input specificity.
- The property whereby only the synapses that received the inducing stimulation are strengthened, leaving other synapses on the same neuron unchanged; the basis of storage capacity.
- Late LTP.
- The durable phase of potentiation, lasting many hours to days, that requires new gene expression and protein synthesis; abolished by translation blockers that spare early LTP.
- Long-term depression (LTD).
- The persistent weakening of a synapse, driven by weaker or more sustained calcium influx that removes AMPA receptors; the mechanistic mirror image of LTP.
- NMDA receptor.
- The glutamate receptor whose channel is blocked by magnesium until the membrane depolarizes; its calcium permeability and voltage dependence make it the coincidence detector that induces LTP.
- Silent synapse.
- A connection containing NMDA but no functional AMPA receptors, and so passing no current at rest; LTP unsilences it by inserting AMPA receptors, converting a mute contact into a working one.
- Synaptic tagging and capture.
- The mechanism by which an active synapse sets a transient local tag that captures diffusely delivered plasticity proteins, explaining how late LTP retains input specificity.
- Tetanus.
- A brief burst of high-frequency stimulation, classically 100 pulses at 100 Hz, used to induce LTP experimentally.
Key Researchers
Timothy V. P. Bliss
(contemporary). Group leader emeritus at the Francis Crick Institute; with Terje Lomo gave the first detailed description of LTP in 1973 and shared the 2016 Brain Prize for establishing it as the leading synaptic model of memory. Wikipedia - Royal Society
Graham L. Collingridge
(contemporary). Professor at the University of Toronto and the Lunenfeld-Tanenbaum Research Institute; showed that LTP induction in area CA1 requires the NMDA receptor, supplying the molecular basis for Hebbian coincidence detection. Shared the 2016 Brain Prize. Faculty Page
Donald O. Hebb
(1904-1985). Professor at McGill University; his 1949 postulate that neurons which fire together wire together gave LTP the theory it would later be found to implement. Wikipedia
Terje Lomo
(contemporary). Professor emeritus at the University of Oslo; first observed the potentiating effect in Per Andersen's laboratory in 1966 and, with Bliss, published its first systematic characterization, coining the term long-term potentiation. Wikipedia - Faculty Page
Robert C. Malenka
(contemporary). Professor at Stanford University; dissected the postsynaptic induction and expression mechanisms of LTP and LTD, including the central role of calcium and AMPA-receptor regulation. Wikipedia - Google Scholar - Faculty Page
Richard G. M. Morris
(contemporary). Professor at the University of Edinburgh; used the NMDA antagonist AP5 to show that blocking LTP impairs spatial learning, and proposed synaptic tagging and capture. Shared the 2016 Brain Prize. Wikipedia - ORCID
Roger A. Nicoll
(contemporary). Professor at the University of California, San Francisco; established that LTP is expressed postsynaptically through AMPA-receptor insertion and the unsilencing of silent synapses, and wrote the definitive short history of the field. Wikipedia - Faculty Page
Frequently Asked Questions
What is long-term potentiation in simple terms?
Long-term potentiation is a lasting increase in the strength of a synapse after it has been used intensely for a brief moment. A short burst of high-frequency activity leaves the connection stronger for hours or days, so the synapse effectively records its own recent use (Bliss & Lomo, 1973).
Why is LTP considered the basis of memory?
Because it has the properties a memory device needs and because interfering with it interferes with learning. Blocking LTP prevents spatial learning, and learning itself produces LTP-like changes at hippocampal synapses, tying the cellular process to behavior (Morris et al., 1986).
What is the role of the NMDA receptor in LTP?
The NMDA receptor is the coincidence detector that induces the common form of LTP. It opens only when glutamate is present and the membrane is depolarized at the same time, and the calcium it then admits triggers the strengthening (Luscher & Malenka, 2012).
What is the difference between early and late LTP?
Early LTP lasts a few hours and modifies proteins already present in the synapse. Late LTP lasts many hours to days and requires new protein synthesis; blocking translation abolishes the late phase while leaving the early phase intact (Kandel, 2001).
What does it mean that LTP is Hebbian?
It means LTP strengthens exactly the inputs that were active when the postsynaptic cell was strongly driven, matching Hebb's rule that connections grow when one cell helps to fire another. Associativity, in which a weak input is strengthened in company with a strong one, is the clearest expression of this (Nicoll, 2017).
Is LTP the same thing as a memory?
No. LTP is a synaptic mechanism that memory appears to use, not memory itself. A memory is a distributed pattern across many synapses; LTP is the change at each one, necessary for certain learning but not identical to the stored trace (Whitlock et al., 2006).
What is long-term depression?
Long-term depression is the persistent weakening of a synapse, the mirror image of LTP. Weaker or more sustained calcium entry removes AMPA receptors rather than adding them, so a single trafficking process can move a synapse in either direction (Citri & Malenka, 2008).
Where in the brain is LTP studied?
LTP was discovered in the dentate gyrus of the hippocampus and is most studied in the hippocampal slice, especially area CA1, but it occurs throughout the brain and is now recorded in awake, behaving animals (Magee & Grienberger, 2020).
References
Bliss, T. V. P., & Lomo, T. (1973). Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. The Journal of Physiology, 232(2), 331-356. https://doi.org/10.1113/jphysiol.1973.sp010273
Bliss, T. V. P., & Collingridge, G. L. (1993). A synaptic model of memory: Long-term potentiation in the hippocampus. Nature, 361(6407), 31-39. https://doi.org/10.1038/361031a0
Citri, A., & Malenka, R. C. (2008). Synaptic plasticity: Multiple forms, functions, and mechanisms. Neuropsychopharmacology, 33(1), 18-41. https://doi.org/10.1038/sj.npp.1301559
Diering, G. H., & Huganir, R. L. (2018). The AMPA receptor code of synaptic plasticity. Neuron, 100(2), 314-329. https://doi.org/10.1016/j.neuron.2018.10.018
Frey, U., & Morris, R. G. M. (1997). Synaptic tagging and long-term potentiation. Nature, 385(6616), 533-536. https://doi.org/10.1038/385533a0
Herring, B. E., & Nicoll, R. A. (2016). Long-term potentiation: From CaMKII to AMPA receptor trafficking. Annual Review of Physiology, 78, 351-365. https://doi.org/10.1146/annurev-physiol-021014-071753
Kandel, E. R. (2001). The molecular biology of memory storage: A dialogue between genes and synapses. Science, 294(5544), 1030-1038. https://doi.org/10.1126/science.1067020
Luscher, C., & Malenka, R. C. (2012). NMDA receptor-dependent long-term potentiation and long-term depression (LTP/LTD). Cold Spring Harbor Perspectives in Biology, 4(6), a005710. https://doi.org/10.1101/cshperspect.a005710
Malenka, R. C., & Bear, M. F. (2004). LTP and LTD: An embarrassment of riches. Neuron, 44(1), 5-21. https://doi.org/10.1016/j.neuron.2004.09.012
Magee, J. C., & Grienberger, C. (2020). Synaptic plasticity forms and functions. Annual Review of Neuroscience, 43, 95-117. https://doi.org/10.1146/annurev-neuro-090919-022842
Morris, R. G. M., Anderson, E., Lynch, G. S., & Baudry, M. (1986). Selective impairment of learning and blockade of long-term potentiation by an N-methyl-D-aspartate receptor antagonist, AP5. Nature, 319(6056), 774-776. https://doi.org/10.1038/319774a0
Nabavi, S., Fox, R., Proulx, C. D., Lin, J. Y., Tsien, R. Y., & Malinow, R. (2014). Engineering a memory with LTD and LTP. Nature, 511(7509), 348-352. https://doi.org/10.1038/nature13294
Nicoll, R. A. (2017). A brief history of long-term potentiation. Neuron, 93(2), 281-290. https://doi.org/10.1016/j.neuron.2016.12.015
Whitlock, J. R., Heynen, A. J., Shuler, M. G., & Bear, M. F. (2006). Learning induces long-term potentiation in the hippocampus. Science, 313(5790), 1093-1097. https://doi.org/10.1126/science.1128134