Episodic Memory

Abstract

Episodic memory is a type of memory for personally experienced events, bound to the time and place of their occurrence and re-experienced through what Tulving called autonoetic consciousness. Tulving distinguished it from semantic memory, the store of context-free general knowledge, and argued that only episodic remembering involves mental time travel into one's own past. Its neural basis lies in the medial temporal lobe, established when bilateral hippocampal resection left a profoundly amnesic patient unable to form new lasting memories of events. A constructive view now holds that the system reconstructing the past also simulates the future, so that remembering and imagining draw on a common network. This article develops the episodic-semantic distinction, autonoetic consciousness, the hippocampal substrate, encoding specificity, constructive simulation, animal episodic-like memory, and consolidation.

Keywords: episodic memory, semantic memory, autonoetic consciousness, hippocampus, mental time travel

Episodic memory is the capacity to recollect specific events from one's own past, complete with the details of what happened, where it happened, and when, and to re-experience them as events that once occurred to oneself (Tulving, 2002). It is distinguished from semantic memory, the vast store of facts and general knowledge that carries no record of when or where it was acquired: a person can know that Paris is the capital of France without remembering the occasion of learning it, and can recall a particular dinner in Paris as an event lived through. What makes episodic memory a single, definable system rather than a loose label for autobiographical detail is the kind of awareness that accompanies it, a first-person re-living of the original experience that Tulving named autonoetic consciousness (Tulving, 1985b). This subjective mark separates genuine remembering from merely knowing that something happened, and it is the feature that later work in neuropsychology, neuroimaging, and comparative cognition has tried to pin to a mechanism. The system is fragile in a revealing way: damage to the medial temporal lobe abolishes the formation of new episodic memories while sparing general knowledge and skills, which is the clinical fact that first showed episodic memory to be neurally distinct (Scoville & Milner, 1957). The sections below develop the episodic-semantic distinction, the autonoetic awareness that defines the system, its hippocampal substrate, the principle that governs retrieval, the constructive view that ties remembering to imagining the future, the question of whether animals have episodic memory, and the consolidation of episodic traces over time.

What Episodic Memory Is

Episodic memory is defined by three properties that together set it apart from every other memory system: its content is a specific event rather than a general fact, that content is bound to a unique spatial and temporal context, and its retrieval is accompanied by the sense of re-living the original experience. Tulving introduced the term in 1972 to name a system that receives and stores information about temporally dated episodes and the relations among them, in explicit contrast to semantic memory, which holds organized knowledge about words, concepts, and their referents without any record of the personal occasions on which that knowledge was gained (Tulving, 2002). The distinction is not merely one of content but of the kind of information represented: episodic memory always carries a self-referential when and where, a pointer back to a particular moment in the rememberer's own life, whereas semantic memory has been stripped of that context and holds only the fact. A single experience typically lays down both kinds of trace at once. Reading a news report deposits semantic knowledge of the event and, separately, an episodic memory of the act of reading it, and the two can dissociate: the fact can survive after the memory of acquiring it has faded, or, in amnesia, the memory of the event can fail to form while the fact is nonetheless learned. This dissociability is what justifies treating episodic memory as a system in its own right rather than as a richly detailed corner of general memory, and it is the property that the rest of this article traces from Tulving's original proposal through its neural, computational, and comparative elaborations (Tulving, 1985a).

The Episodic-Semantic Distinction

The division of declarative memory into episodic and semantic systems is the organizing claim of the field, and Tulving pressed it as more than a convenient labelling of content (Tulving, 1985a). His question, in the title of the paper that made the case, was how many memory systems there are, and his answer was that episodic and semantic memory differ not only in what they store but in how they operate, how they develop, and how they fail. Semantic memory is the general-purpose knowledge system, holding the meanings of words, the properties of objects, and the facts of the world in a form that is abstracted away from any particular encounter; episodic memory is the record of the encounters themselves. Squire embedded this pair within a broader taxonomy that separates declarative memory, the conscious memory for facts and events served by the medial temporal lobe, from a family of nondeclarative systems, the skills, habits, priming, and conditioning that operate without conscious recollection and survive the amnesia that devastates declarative memory (Squire, 1992). Within the declarative branch, episodic and semantic memory are strongly interdependent even as they dissociate. New semantic knowledge is often acquired through episodes and may depend on the episodic system during learning, while episodic memories are built out of semantic elements, since to remember a dinner one must recognize the food, the room, and the faces. Recent analysis has argued that the sharp dichotomy is best replaced by a continuum, on which memories vary in how much episodic and semantic content they carry and can be transformed from one toward the other over time, so that the distinction marks the poles of a graded representational space rather than two sealed stores (Renoult et al., 2019).

Autonoetic Consciousness

What ultimately individuates episodic memory, in Tulving's mature account, is not its content but the kind of consciousness it entails (Tulving, 1985b). He distinguished three levels of awareness tied to three memory systems: anoetic consciousness, a non-knowing awareness accompanying procedural memory; noetic consciousness, the knowing awareness of semantic memory, in which one is aware of a fact without any sense of the personal past; and autonoetic consciousness, the self-knowing awareness of episodic memory, in which one is aware of an event as something that happened to oneself at a particular time. Autonoetic consciousness is what makes episodic retrieval a re-experiencing rather than a mere retrieval of stored information, and it is the psychological content of mental time travel, the capacity to project oneself backward into the personal past and, symmetrically, forward into the personal future. Wheeler, Stuss, and Tulving argued that this self-aware recollection depends specifically on the frontal lobes, so that autonoetic consciousness has an anatomy distinct from the medial-temporal machinery of memory storage, and that frontal damage can impair the subjective sense of remembering while leaving the ability to retrieve information relatively intact (Wheeler et al., 1997). The subjective states are measured experimentally through the remember-know procedure, in which a person who recognizes an item reports whether they remember it, recollecting the episode of encountering it, or merely know it is familiar without any recollection of the encounter. The two responses dissociate under manipulations of attention, depth of processing, and brain damage, giving autonoetic recollection an operational handle and showing that recognition can proceed either through genuine episodic remembering or through a semantic feeling of familiarity. The procedure was given its experimental form by Gardiner, who showed that a levels-of-processing manipulation that improved recognition acted almost entirely on remember responses and left know responses untouched, giving the autonoetic-noetic distinction its first experimental handle in healthy participants and establishing that the two states are functionally separable rather than points on a single continuum of memory strength (Gardiner, 1988).

The Medial Temporal Lobe and Amnesia

The demonstration that episodic memory has a dedicated neural substrate came from the surgery performed on the patient known as H.M., whose intractable epilepsy was treated by bilateral removal of the medial temporal lobes, including the greater part of the hippocampus (Scoville & Milner, 1957). The seizures abated, but the patient was left with a dense anterograde amnesia: he could no longer form new lasting memories of the events of his life, forgetting conversations and meals within minutes, while his intelligence, language, and personality were preserved and his memories from well before the surgery remained. Milner's study of the case established two facts that founded the cognitive neuroscience of memory. The first is that the capacity to form new declarative memories is localized, dependent on the medial temporal lobe rather than distributed across the cortex, so that a circumscribed lesion can abolish new learning while sparing everything else. The second, uncovered as the case was studied over years, is that memory is not a single faculty: the same patient who could form no new episodic memories could nonetheless acquire new perceptual-motor skills, improving on a mirror-drawing task across days while denying each session that he had ever attempted it, which showed that skill learning is served by a system outside the medial temporal lobe. Squire's synthesis of the human, monkey, and rodent evidence generalized the lesion findings into an anatomy of declarative memory, identifying the hippocampus and the adjacent entorhinal, perirhinal, and parahippocampal cortices as the structures whose integrity is required to convert experience into lasting, consciously accessible memory (Squire, 1992). The hippocampus is not a fixed structure but is reshaped by the demands placed on it: London taxi drivers, who hold an elaborate spatial and event memory of the city, have an enlarged posterior hippocampus that grows with their years of navigating experience, direct evidence that the structure central to episodic memory is plastic and use-dependent (Maguire et al., 2000). The pattern of preserved and impaired abilities in medial-temporal amnesia became the template against which every subsequent theory of memory systems was tested.

Encoding Specificity and Retrieval

Episodic memory is not a passive archive from which items are read out on demand; whether a stored event can be retrieved depends on the relation between the conditions of encoding and the conditions of retrieval. Tulving and Thomson formalized this in the encoding specificity principle: a retrieval cue is effective to the extent that the information it carries was encoded together with the target at the time of the original experience (Tulving & Thomson, 1973). The principle overturned the intuitive view that some cues are simply stronger than others. In their experiments, a strongly associated word that would ordinarily be an excellent prompt failed to elicit a target that had been studied in the presence of a different, weakly associated context word, while that weak context word, which shared the encoding episode, succeeded, even though on its own it was a poor associate. What is stored is not the item in isolation but the item as encoded within a particular context, and a cue works by reinstating a portion of that context. This is why the physical or mental setting of learning can act as a cue, so that recall is better when the environment, mood, or internal state at test matches the state at encoding, and why deliberately reinstating the context of an event aids its recollection. Encoding specificity reframes retrieval failure: much of what seems forgotten is not lost from storage but inaccessible for want of the right cue, recoverable when a cue that overlaps the original encoding is supplied. The demonstration below varies the overlap between the encoding context and the cue present at retrieval and shows how the probability of successful recollection rises as that overlap increases.

Constructive Memory and Future Thinking

Episodic memory does not replay a stored recording; it reconstructs an event from distributed elements each time it is retrieved, and this constructive character is both its power and its liability. The same flexibility that lets memory recombine details to represent an event also makes it prone to distortion, so that memories can incorporate information from other sources, drift toward the gist of what happened, and yield confident recollections of events that occurred differently or not at all. The Deese-Roediger-McDermott paradigm demonstrates this cleanly: studying a list of associates such as bed, rest, and tired reliably produces confident false recognition of a related lure, sleep, that was never presented, a gist-consistent intrusion into episodic recollection (Roediger & McDermott, 1995). Schacter and colleagues argued that these constructive properties are not defects of an otherwise faithful system but signatures of its true function, which is not to preserve the past for its own sake but to draw on the past flexibly to anticipate and plan for the future (Schacter et al., 2007). The proposal, the constructive episodic simulation hypothesis, holds that imagining a future event and remembering a past one draw on the same core network and the same stored episodic details, recombined in the one case and reinstated in the other. Addis, Wong, and Schacter provided direct neuroimaging evidence, showing that constructing an imagined future event and recollecting a past one activate a largely overlapping set of regions, with the differences concentrated in the greater constructive demand of assembling a novel scenario (Addis et al., 2007). Hassabis and Maguire converged on the same conclusion from the deficit side, showing that patients with hippocampal amnesia are impaired not only at remembering the past but at imagining new experiences, unable to construct the coherent spatial scenes that both tasks require, which identifies scene construction as a shared process underlying episodic memory and imagination (Hassabis & Maguire, 2007). On this view episodic memory is one expression of a general constructive capacity for mental simulation, and its orientation is as much prospective as retrospective. The demonstration below shows how the amount of episodic detail declines with temporal distance symmetrically into the remembered past and the imagined future.

Episodic-Like Memory in Animals

Whether any non-human animal has episodic memory is contested, because autonoetic consciousness, the defining feature, cannot be reported by a creature without language. Clayton and Dickinson addressed the impasse by setting aside the untestable question of subjective awareness and asking instead whether an animal satisfies the behavioral criteria of episodic memory, integrated knowledge of what happened, where, and when (Clayton & Dickinson, 1998). Working with western scrub jays, birds that cache food and recover it later, they let the birds store two kinds of food, perishable wax worms that the jays prefer and non-perishable peanuts, and then tested recovery after different retention intervals. After a short delay the jays searched preferentially where they had cached the favored worms; after a long delay, over which they had learned the worms would have decayed, they abandoned the worm sites and switched to the peanuts. To do this the jays had to remember what they had cached, where they had cached it, and how long ago, and to integrate the three into a single decision, which is the behavioral signature Tulving's definition demands. Because this meets the operational criteria while leaving the phenomenal question open, the capacity is termed episodic-like memory, a deliberately cautious label. The finding transformed the comparative debate, prompting analogous demonstrations in rodents, corvids, and great apes and forcing a separation between the behavioral content of episodic memory, which can be probed in any species, and the autonoetic awareness that may or may not accompany it. It also sharpened the theoretical stakes of the constructive view, since the same scrub jays that remember specific past caches can also flexibly plan for future needs, extending mental time travel, at least in its behavioral form, beyond the human case. The demonstration below reproduces the design, letting the reader vary the retention interval and watch the jay switch from the perishable to the durable cache as time passes.

Consolidation and Contextual Binding

A newly formed episodic memory is not immediately permanent; it is stabilized over time through consolidation, and how the hippocampus and neocortex divide the labor of long-term storage is among the field's central questions. The standard consolidation account holds that the hippocampus binds the elements of an event at encoding and then, over weeks to years, gradually trains durable representations in the neocortex, after which the memory becomes independent of the hippocampus, which is why amnesia often spares remote memories while destroying recent ones. Moscovitch and colleagues challenged the claim that episodic memories ever leave the hippocampus, proposing in multiple-trace and trace-transformation theory that vivid, detailed recollection of a specific event always depends on the hippocampus no matter how old the memory, while a semanticized gist of the same event can become hippocampus-independent, so that memories transform in character rather than simply migrating out of the medial temporal lobe (Moscovitch et al., 2016). Yonelinas and colleagues reframed the debate around what the hippocampus actually does, arguing in a contextual binding account that it binds items to the context in which they occur and that forgetting and the pattern of amnesia are better explained by interference among overlapping contexts than by a slow systems-level transfer (Yonelinas et al., 2019). Underlying all of these accounts is the physical trace itself, the engram, and work on the cellular basis of memory has begun to identify the specific populations of neurons whose activity encodes an event and whose reactivation constitutes its recall, tying the psychology of consolidation to the sparse ensembles that carry a memory and can be artificially reactivated to evoke it (Josselyn & Tonegawa, 2020). Table 1 summarizes the major theoretical positions developed across this article and the signature evidence for each.

Table 1

Major Theoretical Positions on Episodic Memory

Worked Example

The remember-know procedure yields two quantities that must be kept apart: recollection, the autonoetic re-experiencing of an encounter, and familiarity, the noetic sense that an item was seen without any recollection of seeing it. Because a person reports a bare know only when familiarity is present but recollection is absent, the raw proportion of know responses understates familiarity, and a standard correction recovers its true rate by treating recollection and familiarity as independent. Suppose a recognition test in which studied items draw a remember response with probability R = 0.50 and a know response with probability 0.30. The independence assumption holds that familiarity operates on the items not already recollected, so the reported know rate equals the familiarity probability F acting on the fraction that were not recollected: 0.30 = F · (1 − R) = F · (1 − 0.50). Solving gives F = 0.30 / 0.50 = 0.60, so familiarity is present for 60 percent of items, not the 30 percent the raw know responses suggest, because recollection pre-empted a know response on the many items it captured. The overall probability that an item is recognized through one route or the other is then R + F · (1 − R) = 0.50 + 0.60 · 0.50 = 0.80. The lesson is that recollection and familiarity are separable contributions to recognition, and that the subjective know report, taken at face value, systematically undercounts familiarity, an arithmetic that matters when the two processes are dissociated by amnesia, which spares familiarity while stripping out recollection (Yonelinas et al., 2019).

Discussion

Episodic memory has traveled from a descriptive distinction to a mechanistic theory, and the through-line is Tulving's insistence that a particular kind of consciousness, not merely a kind of content, marks the system out. The episodic-semantic distinction gave memory research its most durable taxonomy and, with Squire's broader declarative-nondeclarative division, organized the lesion evidence into an anatomy in which the medial temporal lobe is the gateway to conscious memory for events (Squire, 1992). The study of the amnesic patient H.M. supplied the founding fact that new episodic learning can be abolished in isolation, and it did so with a clarity that a century of animal work had not achieved (Scoville & Milner, 1957). Encoding specificity then showed that retrieval is a relation, not a property of stored strength, and that much forgetting is a failure of access rather than of storage (Tulving & Thomson, 1973). The most consequential recent turn is the constructive one, which recast episodic memory as a simulation engine oriented as much toward the future as the past, unifying remembering, imagining, and planning under one hippocampally dependent process and explaining memory's characteristic distortions as the cost of that flexibility (Schacter et al., 2007; Hassabis & Maguire, 2007). Comparative work meanwhile showed that the behavioral criteria of episodic memory can be met by a scrub jay, forcing a clean separation between what episodic memory does and the awareness that may accompany it (Clayton & Dickinson, 1998). What remains unsettled is the fate of the trace over time, where trace-transformation and contextual-binding accounts continue to contest how, and whether, an episodic memory ever becomes independent of the hippocampus (Moscovitch et al., 2016; Yonelinas et al., 2019). What began as a way of labelling the difference between remembering a dinner and knowing a fact is now a theory of how a constructive brain builds, stores, and re-uses the record of its own experience.

Current Directions

Contemporary research on episodic memory is concentrated where its psychology meets its cellular biology. The identification of the engram, the sparse ensemble of neurons whose activity encodes a specific memory, has moved from inference to intervention: such ensembles can now be tagged, silenced, and artificially reactivated in animals, so that a memory can be evoked by stimulating its cells and even linked to a false context, which turns questions about consolidation and reconstruction into experiments on identified populations (Josselyn & Tonegawa, 2020). A second front continues to test how far the constructive account reaches, examining whether the same episodic system supports not only remembering and future thinking but also spatial navigation, decision making, and the coherence of the self, and treating its breakdown as a common thread across amnesia, aging, and psychiatric conditions in which past and future thinking are jointly impaired (Schacter et al., 2007). The relationship between episodic and semantic memory is being reconceived as a continuum along which memories are transformed, sharpening the question of how repeated retrieval and consolidation semanticize an initially detailed episode (Renoult et al., 2019). The systems-consolidation debate remains open and has been reframed rather than resolved, with contextual-binding and trace-transformation accounts making competing predictions about when remote memories should engage the hippocampus that high-resolution imaging is beginning to adjudicate (Moscovitch et al., 2016; Yonelinas et al., 2019). Running through all of it is the comparative programme, which continues to probe episodic-like memory and future planning in birds, rodents, and primates, using the behavioral what-where-when criterion to ask which components of human episodic memory are shared and which may be unique (Clayton & Dickinson, 1998). The unifying goal is an account in which the phenomenology Tulving described, the anatomy the lesion studies mapped, and the ensembles the engram work has isolated are understood as levels of one system.

Commonly Confused With

Semantic Memory

Both are declarative systems, and both feel like knowing, which is why they are run together, but they differ in the awareness that accompanies retrieval. Episodic memory carries a when and a where and is re-experienced as one's own past through autonoetic consciousness; semantic memory is general knowledge stripped of its acquisition context, retrieved with noetic knowing and no sense of the occasion on which it was learned. Remembering the dinner at which a fact was mentioned is episodic; knowing the fact afterward, with no recollection of the dinner, is semantic. The two are ordinarily laid down together by a single experience and can dissociate: amnesia can block new episodic memories while leaving the capacity to learn new facts relatively intact (Tulving, 1985a; Renoult et al., 2019).

Common Misconceptions

Episodic memory works like a video recording of the past.

Episodic memory reconstructs an event from distributed elements each time it is retrieved rather than replaying a stored recording. This constructive process is what lets the same system recombine details to imagine future events, and it is also why memory is prone to distortion, incorporating information from other sources and yielding confident recollections of events that happened differently or not at all (Schacter et al., 2007).

Forgetting means a memory has been erased from storage.

The encoding specificity principle shows that much of what seems forgotten is stored but inaccessible for want of a matching cue. A memory that cannot be recalled with one prompt is often recovered when a cue is supplied that reinstates the context present at encoding, so retrieval failure is frequently a failure of access rather than a loss from storage (Tulving & Thomson, 1973).

Episodic memory is purely about the past.

The system that reconstructs past events is the same one that simulates future ones. Imagining a personal future event and remembering a past one draw on a common network, and patients with hippocampal amnesia are impaired at both, so episodic memory is better understood as a constructive capacity for mental time travel in either direction than as a retrospective archive (Addis et al., 2007; Hassabis & Maguire, 2007).

Glossary

Anterograde amnesia.

The inability to form new lasting memories of events after the onset of brain damage, with memories from before the damage relatively spared; the defining deficit of medial-temporal amnesia.

Autonoetic consciousness.

The self-knowing awareness that accompanies episodic retrieval, in which an event is re-experienced as having happened to oneself at a particular time; the defining mark of episodic memory.

Consolidation.

The time-dependent process by which a newly encoded memory is stabilized, involving cellular changes and, at the systems level, a changing division of labor between hippocampus and neocortex.

Constructive episodic simulation.

The hypothesis that remembering the past and imagining the future are one constructive process, recombining stored episodic details within a common brain network.

Declarative memory.

The conscious, explicit memory for facts and events, comprising the episodic and semantic systems and dependent on the medial temporal lobe; contrasted with nondeclarative memory.

Deese-Roediger-McDermott (DRM) paradigm.

An experimental method in which studying a list of words all associated with an unpresented lure reliably elicits confident false recognition or recall of that lure; a standard demonstration of memory's constructive, gist-based distortions.

Encoding specificity principle.

The principle that a retrieval cue is effective to the degree that it reinstates information encoded together with the target at the time of the original experience.

Engram.

The physical memory trace, understood as the sparse ensemble of neurons whose activity encodes a specific event and whose reactivation constitutes its recall.

Episodic memory.

The memory system for personally experienced events bound to a specific time and place, retrieved with autonoetic re-experiencing of the original occasion.

Episodic-like memory.

In animals, memory that meets the behavioral what-where-when criteria of episodic memory without a claim about the accompanying awareness; first demonstrated in scrub jays.

Familiarity.

A noetic sense that an item has been encountered before, without recollection of the encounter; one of the two processes supporting recognition alongside recollection.

Hippocampus.

A medial-temporal structure required for forming new episodic memories, for binding the elements of an event to its context, and, on some accounts, for detailed recollection at any age.

Mental time travel.

The projection of the self backward into the personal past or forward into the personal future, the phenomenology of episodic memory and its prospective counterpart.

Noetic consciousness.

The knowing awareness that accompanies semantic retrieval, in which one is aware of a fact without any sense of the personal occasion on which it was learned.

Recollection.

The autonoetic retrieval of a specific past encounter with its contextual detail; the process indexed by a remember response and dissociable from familiarity.

Semantic memory.

The memory system for general knowledge, facts, and word meanings, abstracted away from any record of the personal occasions on which they were acquired.

Trace-transformation theory.

The account that detailed episodic recollection remains hippocampus-dependent no matter how old the memory, while a semanticized gist of the event can become hippocampus-independent.

Key Researchers

Donna Rose Addis (b. 1978). Professor at the Rotman Research Institute, Baycrest and the University of Toronto; with Schacter she formulated the constructive episodic simulation hypothesis and mapped the shared substrates of remembering and future thinking. ORCID - Google Scholar - Faculty Page - Wikipedia

Nicola S. Clayton (b. 1962). Professor at the University of Cambridge; she demonstrated episodic-like what-where-when memory in scrub jays, showing that the behavioral criteria for episodic memory can be met by a non-human animal. ORCID - Google Scholar - Faculty Page - Wikipedia

Demis Hassabis (b. 1976). Cognitive neuroscientist and co-founder of DeepMind; before founding it he showed with Maguire that hippocampal amnesics cannot imagine new experiences, evidence for a common constructive system. ORCID - Google Scholar - Faculty Page - Wikipedia

Eleanor A. Maguire (1970-2025). Professor at University College London; her taxi-driver studies showed experience-dependent hippocampal plasticity, and her later work tied episodic memory to scene construction and imagination. ORCID - Wikipedia

Brenda Milner (b. 1918). Professor at the Montreal Neurological Institute and McGill University; her studies of the patient H.M. established that the medial temporal lobe is necessary for forming new declarative memories. Faculty Page - Wikipedia

Morris Moscovitch (b. 1945). Professor at the University of Toronto and the Rotman Research Institute; he co-developed multiple-trace and trace-transformation theory, holding that detailed episodic recollection stays hippocampus-dependent. ORCID - Google Scholar - Faculty Page - Wikipedia

Daniel L. Schacter (b. 1952). Professor at Harvard University; he developed the constructive view of memory and, with Addis, the constructive episodic simulation hypothesis linking remembering to imagining the future. ORCID - Google Scholar - Faculty Page - Wikipedia

Larry R. Squire (b. 1941). Professor at the University of California, San Diego and the VA San Diego; he formalized the declarative-nondeclarative taxonomy and mapped the hippocampal system's role in declarative memory. Google Scholar - Faculty Page - Wikipedia

Endel Tulving (1927-2023). Professor at the University of Toronto and the Rotman Research Institute; he introduced the episodic-semantic distinction and the concepts of autonoetic consciousness and mental time travel that define the field. Google Scholar - Faculty Page - Wikipedia

Frequently Asked Questions

What is episodic memory?
Episodic memory is the memory system for personally experienced events, bound to the specific time and place of their occurrence and retrieved with a first-person sense of re-living the original experience, which Tulving called autonoetic consciousness (Tulving, 2002).

How does episodic memory differ from semantic memory?
Episodic memory stores events with a when and a where and is re-experienced as one's own past, whereas semantic memory stores general knowledge and facts abstracted away from any record of when or where they were learned; both are declarative, but only episodic retrieval carries autonoetic awareness (Tulving, 1985a).

What is autonoetic consciousness?
Autonoetic consciousness is the self-knowing awareness that accompanies episodic retrieval, in which an event is not merely known to have happened but re-experienced as having happened to oneself at a particular time; it is what distinguishes remembering from knowing (Tulving, 1985b).

Which part of the brain supports episodic memory?
The medial temporal lobe, and the hippocampus in particular, is necessary for forming new episodic memories, established when bilateral removal of these structures in the patient H.M. abolished new declarative learning while sparing intelligence, skills, and old memories (Scoville & Milner, 1957; Squire, 1992).

What is the encoding specificity principle?
The encoding specificity principle holds that a retrieval cue is effective only to the extent that it reinstates information encoded together with the target event, so that a cue sharing the original context can succeed where a normally strong but unrelated cue fails (Tulving & Thomson, 1973).

Why is episodic memory linked to imagining the future?
The constructive episodic simulation hypothesis holds that remembering the past and imagining the future draw on the same core network and stored episodic details; imagined future and recollected past events activate overlapping brain regions, and hippocampal amnesics are impaired at both (Schacter et al., 2007; Addis et al., 2007).

Do animals have episodic memory?
Scrub jays meet the behavioral what-where-when criteria of episodic memory, remembering what food they cached, where, and how long ago, and integrating the three to decide where to search; because the accompanying awareness cannot be verified, this is termed episodic-like memory (Clayton & Dickinson, 1998).

Does an episodic memory ever become independent of the hippocampus?
This is contested: standard consolidation says memories gradually move to the neocortex, but trace-transformation theory holds that detailed recollection stays hippocampus-dependent no matter how old the memory, while a semanticized gist can become independent (Moscovitch et al., 2016; Yonelinas et al., 2019).

References