Anterograde Amnesia

Abstract

Anterograde amnesia is a form of amnesia in which brain damage leaves a person unable to form new long-term memories for facts and events, while memories laid down before the injury and moment-to-moment awareness remain comparatively intact. It is among the sharpest natural dissociations the mind offers: a person can hold a conversation and reason normally, retaining information while attention is on it, yet retain nothing of it minutes later. The condition made the hippocampus and surrounding medial temporal lobe the anatomical centre of the science of memory, above all through the patient H.M. This article sets out what the condition is, how it separates from retrograde loss, the memory system it exposed, the spared skill learning that divides declarative from nondeclarative memory, consolidation theory, and the causes that converge on the syndrome.

Keywords: anterograde amnesia, medial temporal lobe, memory consolidation

Anterograde amnesia occupies a special position in the study of memory because it is the condition in which the formation of new memory fails while almost everything else about the mind is preserved. A person with dense anterograde amnesia can score in the normal range on an intelligence test, follow and contribute to a discussion, and repeat a telephone number back correctly; what they cannot do is carry any of it forward. Introduced to a visitor, they will greet the same visitor as a stranger a quarter of an hour later. This selective failure — intact perception, intact reasoning, intact working memory, intact old knowledge, but no new lasting record — is what makes the condition so revealing: it shows that laying down new long-term memories is a distinct operation of the brain, separable from thinking, from attending, and from remembering the past (Squire, 2009).

The account below moves from what anterograde amnesia is, through the anterograde–retrograde distinction and the medial temporal lobe system that patient H.M. exposed, to the dissociation of spared and lost forms of learning, the temporal gradient of consolidation, and the range of causes that produce the syndrome. The recurring theme is that a single, circumscribed lesion can subtract one specific capacity — the making of new declarative memories — and by its absence reveal the architecture of a normal memory.

What Anterograde Amnesia Is

Anterograde amnesia is the inability to form new long-term memories for facts and events after the onset of brain injury or disease. The Latin roots are exact: antero- (forward, ahead) and -grade (to go), so the loss runs forward in time from the injury. A person with the condition retains an ordinary short-term hold on information — they can keep a name, a number, or an instruction in mind while they attend to it — but once attention moves away, no durable trace is left behind (Squire, 2009). The defining picture is therefore not confusion or a general loss of intellect, but a specific, repeatable failure: experiences that ought to become memories simply do not.

Two features make the deficit unmistakable in the clinic and instructive in the laboratory. The first is its selectivity. Immediate memory is intact: on a digit-span test, which probes the amount of information working memory can hold at once, a densely amnesic patient performs normally. Language, perception, and general knowledge acquired before the injury are preserved. What fails is the transfer of new experience into long-term memory (Scoville & Milner, 1957). The second is its density: in severe cases the failure is near-total for consciously accessible new information, so that hours of a patient's day leave no recoverable record, even though each moment was lived through with full awareness at the time.

It is important to separate anterograde amnesia from the everyday sense of forgetting. Ordinary forgetting is the decay or interference of memories that were once formed; anterograde amnesia is the failure to form the memory in the first place. The information never enters the store from which it could later be retrieved. This is why cueing, which helps in ordinary forgetting by prompting retrieval of a stored trace, does not restore a memory in dense anterograde amnesia for events the patient cannot recall — there is no consolidated trace for the cue to reach (Squire et al., 2015).

Anterograde versus Retrograde Amnesia

The clearest way to locate anterograde amnesia is against its mirror image, retrograde amnesia. Where anterograde amnesia is the loss of the ability to make memories after an injury, retrograde amnesia is the loss of memories acquired before it. The two are defined relative to the same event — the moment of onset — and they are dissociable: a patient may have severe anterograde amnesia with only mild retrograde loss, as in patient H.M., or the reverse, as in some cases of focal retrograde amnesia (Squire, 2009). The double dissociation is what tells us that forming a memory and retrieving an old one are supported by partly different mechanisms.

Retrograde amnesia is rarely uniform across the past. It characteristically follows a temporal gradient, often called Ribot's law after the nineteenth-century psychologist who first described it: memories from the period just before the injury are the most likely to be lost, and the probability of survival rises the more remote the memory. Childhood memories and long-established knowledge are typically spared even when the months before the injury are gone (Frankland & Bontempi, 2005). This gradient is one of the central pieces of evidence for consolidation theory, discussed below, because it implies that a memory's dependence on the damaged system diminishes with time.

The proportions matter clinically. In the classic medial temporal lobe amnesias, anterograde impairment dominates and retrograde loss, when present, is temporally graded and limited. When retrograde amnesia is extensive and covers decades while new learning is relatively preserved, the picture points either toward damage beyond the medial temporal lobe — to the temporal neocortex where remote memories are thought to reside — or, in some presentations, toward a functional rather than a structural cause (Squire et al., 2015). Reading the balance of the two deficits is therefore a first step in localising the underlying lesion.

The Medial Temporal Lobe Memory System

The modern science of memory begins with a single patient. In 1953, the neurosurgeon William Beecher Scoville removed the medial portions of both temporal lobes of a young man, Henry Molaison — known for decades only as H.M. — in an attempt to control intractable epilepsy. The seizures improved, but the operation produced an unforeseen and profound anterograde amnesia. Brenda Milner's examination revealed a man of normal intelligence and preserved personality who could no longer form new lasting memories of the people he met or the events of his life (Scoville & Milner, 1957). Because the surgery had been bilateral and its extent was documented, H.M. established what no prior case had: that the medial temporal lobe is necessary for forming new declarative memories, and that this function is separable from perception, language, and general intellect (Corkin, 2002).

The critical structures are the hippocampus and the cortex adjacent to it — the entorhinal, perirhinal, and parahippocampal regions — which together form the medial temporal lobe memory system. Converging evidence from human patients and from controlled lesion studies in animals established that damage limited to this system produces anterograde amnesia proportional to its extent, and that the hippocampus proper is especially important for memory of events and their context (Squire et al., 2015). Even damage restricted to a single hippocampal subfield can suffice: bilateral loss of the CA1 pyramidal cells alone, as occurs in certain ischaemic and inflammatory injuries, produces a clinically significant anterograde amnesia (Bartsch, Doehring, Rohr, Jansen, & Deuschl, 2011).

H.M. was studied for more than fifty years, and the follow-up work refined the initial picture in ways that shaped the whole field. Milner, Corkin, and Teuber's fourteen-year follow-up documented both the stability of the amnesia and, crucially, the discovery that H.M. could learn — his performance on a mirror-drawing task improved steadily across days even though he had no memory of ever having done it (Milner et al., 1968). That single observation, elaborated over the following decades, forced the recognition that memory is not one faculty but several, only some of which depend on the medial temporal lobe (Squire, 2009).

Spared Learning: Declarative and Nondeclarative Memory

The most theoretically important fact about anterograde amnesia is what it leaves untouched. Patients with dense amnesia for facts and events can nonetheless acquire, at a normal or near-normal rate, a range of skills and dispositions: motor skills such as mirror tracing and rotary pursuit, perceptual skills such as reading mirror-reversed text, classical conditioning, priming, and simple habits. They improve with practice while consistently denying, on each new session, that they have ever practised (Milner et al., 1968). This is the empirical basis for the central distinction of the field: between declarative memory — the conscious memory for facts and events that anterograde amnesia impairs — and nondeclarative memory, the collection of learning systems that it spares.

Declarative memory itself divides, following Endel Tulving, into episodic memory for personally experienced events, located in time and place, and semantic memory for general facts and knowledge (Tulving, 2002). Anterograde amnesia typically impairs the formation of both, but the impairment of new episodic memory is usually the more devastating and the more complete. Whether densely amnesic adults can acquire any new semantic memory — new facts stripped of the context in which they were learned — has been a productive controversy, with evidence that limited, slow semantic learning is sometimes possible when the hippocampus is damaged but surrounding cortex is spared (Vargha-Khadem et al., 1997).

The nondeclarative systems that survive share a signature: they express themselves through performance rather than through conscious recollection. A patient shows that they have learned a skill by doing it better, not by remembering the training. This is why nondeclarative memory is sometimes called implicit: the knowledge is revealed in behaviour without any accompanying sense of remembering (Squire, 2009). The dissociation is doubly informative. It shows that the medial temporal lobe is specialised for declarative memory in particular, and it shows that the brain solves the problem of learning not once but many times, in parallel systems with different rules, different anatomy, and different relationships to consciousness. Procedural memory for skills, in particular, depends on the basal ganglia and cerebellum rather than the hippocampus, which is why it survives medial temporal damage intact.

Consolidation and the Temporal Gradient

Why should a memory formed a week before an injury be lost while one formed a decade before survives? The answer is consolidation: the process by which a newly encoded memory is gradually transformed from a labile, hippocampus-dependent state into a more stable form that can be retrieved without the hippocampus (Squire et al., 2015). On the standard model of systems consolidation, the hippocampus rapidly binds the elements of a new experience — the what, where, and when — and then, over weeks to years, coordinates the slow strengthening of connections among the neocortical regions that hold the memory's content. As the neocortical representation matures, the memory's dependence on the hippocampus fades. A lesion therefore erodes recent, still-hippocampus-dependent memories while sparing remote ones that have completed the transfer — exactly the temporal gradient that retrograde amnesia displays (Frankland & Bontempi, 2005).

Consolidation operates at two scales. Synaptic (or cellular) consolidation stabilises a trace within hours through molecular changes at the synapse, and depends on the long-term potentiation of hippocampal circuits (Lisman, Buzsaki, Eichenbaum, Nadel, Ranganath, & Redish, 2017). Systems consolidation is the slower, larger-scale reorganisation across brain regions just described. Sleep is thought to be central to the systems process: during slow-wave sleep the hippocampus repeatedly reactivates, or replays, recently encoded patterns, and this replay is believed to drive their gradual integration into neocortical networks (Squire et al., 2015). Anterograde amnesia, on this view, is the failure of the first, hippocampal step: without it, there is no labile trace for consolidation to stabilise, so nothing is carried forward.

The standard model is not unchallenged. On the multiple trace and related accounts, richly detailed episodic memories may remain dependent on the hippocampus indefinitely, and only the semanticised gist becomes hippocampus-independent — a view supported by cases in which extensive hippocampal damage produces retrograde loss of autobiographical detail reaching far back in time (Bartsch et al., 2011). A recent reframing, contextual binding theory, argues that much of what looks like a consolidation gradient is better explained by interference and by the hippocampus's role in binding memories to their context, rather than by a slow transfer of storage out of the hippocampus (Yonelinas, Ranganath, Ekstrom, & Wiltgen, 2019). The debate is unsettled, but all parties agree on the fact anterograde amnesia establishes: the hippocampus is required to form new declarative memories, whatever its longer-term role in storing them.

Causes of Anterograde Amnesia

Anterograde amnesia is a syndrome, not a single disease, and several routes converge on it by damaging the medial temporal lobe or the structures functionally linked to it. Understanding the causes matters both clinically and theoretically, because the pattern of associated deficits helps localise the lesion and because each cause is, in effect, a natural experiment on the memory system.

Medial temporal lobe damage is the prototype. Bilateral surgical resection produced H.M.'s amnesia; the same syndrome follows bilateral hippocampal damage from anoxia or ischaemia (as after cardiac arrest), from herpes simplex encephalitis, and from certain autoimmune encephalitides. Autoimmune limbic encephalitis associated with antibodies to LGI1, for instance, can produce a focal atrophy of the CA3 hippocampal subfield and a corresponding anterograde memory impairment, illustrating how selective the anatomical basis can be (Miller et al., 2017). Diencephalic amnesia arises not from the temporal lobe but from midline structures — the mammillary bodies and the thalamus — damaged most commonly in Korsakoff syndrome, the consequence of thiamine deficiency in chronic alcohol misuse. Diencephalic and medial temporal amnesias are clinically similar, a similarity that maps the extended hippocampal circuit these structures share (Squire et al., 2015).

Pharmacological anterograde amnesia is the most everyday form and the clearest demonstration that memory formation is a discrete, interruptible process. Benzodiazepines and general anaesthetic agents produce a dose-dependent, temporary anterograde amnesia by impairing the encoding of new declarative memories while leaving already-formed memories and immediate attention comparatively intact — which is precisely why they are used to prevent the recall of unpleasant procedures (Ghoneim, 2004). Because the drug is cleared and the deficit lifts, this form is reversible, unlike the structural amnesias. Developmental amnesia results from hippocampal injury early in life, typically from a hypoxic-ischaemic event around birth: children so affected grow up with a severe deficit in episodic memory yet can, remarkably, acquire enough semantic knowledge to attend mainstream school and develop language and literacy — a dissociation that has reshaped thinking about how the two forms of declarative memory relate (Vargha-Khadem et al., 1997). Finally, transient global amnesia is the striking transient form: a sudden, self-limiting episode of dense anterograde amnesia, lasting several hours, in which an otherwise lucid person repeatedly asks the same orienting questions and forms no lasting memory of the episode, before memory function returns to normal. It is associated with transient, reversible signal changes in the CA1 subfield of the hippocampus, and it is the cleanest natural demonstration that the medial temporal memory system can fail and recover as a functional unit without a permanent lesion (Bartsch & Deuschl, 2010).

Worked Example

Consider a standard clinical test that isolates anterograde amnesia: a short story containing a fixed number of scorable idea units is read aloud, the patient retells it immediately, and then retells it again after a filled 30-minute delay. Immediate recall probes whether the material was understood and held in working memory; delayed recall probes whether any of it was carried into long-term storage. The diagnostic signature of anterograde amnesia is a normal or near-normal immediate score followed by a collapse at delay.

Suppose the story contains 25 idea units. A healthy control recalls 15 immediately and 12 after the delay. A patient with dense anterograde amnesia recalls 11 immediately — within the normal range, because working memory and comprehension are intact — but 0 after the delay. We can quantify each person's retention as the proportion of what they initially recalled that survived the delay:

retention = (delayed recall) ÷ (immediate recall)

For the control, retention = 12 ÷ 15 = 0.80, so 80% of what was initially grasped was still available after 30 minutes — ordinary, modest forgetting. For the patient, retention = 0 ÷ 11 = 0.00: nothing survived the delay, despite a normal starting point. The savings over the delay — the percentage lost — is 20% for the control and 100% for the patient. The contrast is the point. The patient's normal immediate score rules out a problem with attention, language, or comprehension; the zero delayed score, following that normal start, isolates the failure precisely at the transfer of new information into durable memory. It is this dissociation within a single patient — intact immediate hold, absent delayed retention — rather than any single low number, that defines the anterograde deficit and separates it from a general cognitive impairment, in which the immediate score would fall too.

Discussion

Anterograde amnesia has done more to shape the cognitive psychology of memory than any other clinical condition, because it delivers, in one person, the dissociations that theory most needs. The intact digit span beside the absent delayed recall separates working memory from long-term memory. The spared skill learning beside the lost fact learning separates nondeclarative from declarative memory. The graded retrograde loss beside the total anterograde loss separates storage from formation and motivates the theory of consolidation. Each of these boundaries, now standard in the textbook architecture of memory, was drawn with evidence that a single circumscribed lesion can subtract one capacity while leaving its neighbours intact (Squire, 2009).

The condition also carries a lesson about the relationship between memory and the self. H.M. and patients like him retain their personality, their sense of humour, their pre-injury identity and knowledge, yet they live without the accumulating record of new experience that, for most people, constitutes the ongoing story of a life. That they remain recognisably themselves shows that the self is anchored substantially in consolidated, pre-injury memory and in preserved semantic knowledge, not only in the continuous laying-down of new episodes. At the same time, the loss is real and profound: the capacity to update one's autobiography, to learn from yesterday, to recognise a face met last week, depends on exactly the system anterograde amnesia removes (Corkin, 2002).

For cognitive theory, the open questions concern storage rather than formation. That the hippocampus is required to form new declarative memories is not in dispute; what remains contested is how long, and in what way, a memory continues to depend on it — the issue that separates the standard consolidation model from multiple-trace and contextual-binding accounts (Yonelinas et al., 2019). Resolving it matters for prognosis and rehabilitation, because it bears on which memories a hippocampal lesion should be expected to erase and which it should spare.

Current Directions

Contemporary work on anterograde amnesia is driven less by new patients than by new tools that let the memory trace itself be manipulated. The most influential development is the engram research programme, which uses activity-dependent tagging in animals to label the specific neurons that encode a given memory and then to reactivate or silence them. This work has reframed the classic question of amnesia — is a lost memory gone, or merely unreachable? — as an empirical one. In mouse models of early Alzheimer-type pathology, memories that could not be retrieved by natural cues were nonetheless recovered by directly stimulating the tagged engram cells, implying that in at least some conditions the failure is one of retrieval access rather than of storage (Roy et al., 2016). Whether the same is true of human amnesias remains unknown, but the finding sharpens the storage-versus-formation distinction that consolidation theory rests on (Josselyn & Tonegawa, 2020).

A second front is the finer anatomy of the human hippocampus. High-resolution imaging now resolves individual hippocampal subfields in living patients, so that a focal loss of CA1 or CA3 can be tied to a specific memory profile rather than attributed to the medial temporal lobe as a whole (Bartsch et al., 2011; Miller et al., 2017). A third is theoretical: contextual binding theory has revived the debate over whether a consolidation gradient reflects a slow transfer of storage or the accumulating interference and contextual drift that any binding system would show, a reframing that makes new, testable predictions about when remote memories should and should not survive hippocampal damage (Yonelinas et al., 2019). None of these lines has overturned the core facts H.M. established, but together they are turning the study of anterograde amnesia from a description of what is lost into a mechanistic account of why.

Common Misconceptions

“Amnesia means losing one's identity and past.”

This is the film version, and it is almost the opposite of the commonest real amnesia. Dense anterograde amnesia leaves identity, personality, and most of the pre-injury past intact; what is lost is the ability to form new memories going forward. The dramatic loss of self is a rare and usually functional (psychogenic) presentation, not the medial temporal lobe syndrome (Corkin, 2002).

“Anterograde amnesics cannot learn anything at all.”

They cannot learn new facts and events consciously, but they learn skills, habits, and conditioned responses at a normal rate — while denying they have ever practised. This spared nondeclarative learning is one of the most important findings the condition produced (Milner et al., 1968).

“A blow to the head that erases the last few years is anterograde amnesia.”

Losing memories formed before an injury is retrograde amnesia. Anterograde amnesia concerns memories that would have formed after it. The two are dissociable and frequently co-occur in different proportions (Squire et al., 2015).

“All memory loss comes from hippocampal damage.”

The hippocampus is central, but diencephalic structures (Korsakoff syndrome) produce a clinically similar amnesia, and drugs such as benzodiazepines produce a fully reversible anterograde amnesia with no structural lesion at all (Ghoneim, 2004).

Commonly Confused With

Anterograde amnesia vs. retrograde amnesia

The prefix names the direction in time relative to the injury, and that is the rule to apply. Ask when the lost memories would have formed: if the failure is to record events happening after the onset, it is anterograde; if intact events from before the onset can no longer be retrieved, it is retrograde. Patient H.M. is the anchor case — a severe anterograde deficit, unable to lay down new episodic memories, with only limited, graded retrograde loss. The two dissociate and usually co-occur in different proportions, so classify each direction separately rather than assuming one implies the other.

Glossary

Anterograde amnesia.

The inability to form new long-term declarative memories after the onset of brain injury or disease, with immediate memory and pre-injury memory relatively spared.

Consolidation.

The process by which a newly encoded memory is stabilised over time, progressing from a labile, hippocampus-dependent state to a durable form; it operates at synaptic and systems scales.

Declarative memory.

Conscious memory for facts and events that can be brought to mind and “declared”; it depends on the medial temporal lobe and is the form anterograde amnesia impairs.

Diencephalic amnesia.

Anterograde amnesia arising from damage to midline diencephalic structures — the mammillary bodies and thalamus — most often in Korsakoff syndrome.

Digit span.

The number of digits a person can repeat back in order immediately after hearing them; a measure of immediate/working memory that is spared in anterograde amnesia.

Engram.

The physical substrate of a stored memory — the specific pattern of neurons and synaptic changes that encode it; now directly labelled and manipulated in animal studies.

Episodic memory.

Declarative memory for personally experienced events, located in a specific time and place; its formation is the capacity most completely lost in anterograde amnesia.

Korsakoff syndrome.

A chronic amnesic disorder caused by thiamine (vitamin B1) deficiency, usually in the context of alcohol misuse, producing diencephalic anterograde and retrograde amnesia.

Medial temporal lobe.

The set of structures on the inner surface of the temporal lobe — the hippocampus and the entorhinal, perirhinal, and parahippocampal cortices — that together form the declarative memory system.

Nondeclarative memory.

The collection of learning systems — skills, habits, priming, conditioning — that operate without conscious recollection and are spared in anterograde amnesia; also called implicit memory.

Retrograde amnesia.

The loss of memories acquired before the onset of injury; dissociable from anterograde amnesia and often temporally graded.

Ribot's law.

The generalisation that in retrograde amnesia recent memories are more vulnerable than remote ones, producing a temporal gradient of loss.

Semantic memory.

Declarative memory for general facts and knowledge, independent of the context in which it was learned; sometimes partly acquirable despite hippocampal damage.

Systems consolidation.

The slow, hippocampus-coordinated reorganisation over weeks to years by which a memory becomes retrievable from neocortex without the hippocampus.

Temporal gradient.

The pattern in which memories nearer in time to an injury are more likely to be lost than older ones; the key behavioural evidence for consolidation.

Transient global amnesia.

A sudden, self-limiting episode of dense anterograde amnesia lasting several hours, linked to reversible changes in the hippocampal CA1 subfield, that resolves without a permanent lesion.

Key Researchers

Suzanne Corkin

(1937–2016). Neuroscientist at the Massachusetts Institute of Technology who studied patient H.M. for more than four decades, mapping the extent of his lesion and the precise scope of his spared and lost abilities; her synthesis remains the definitive account of the case (Corkin, 2002). Wikidata

Eleanor A. Maguire

(1970–2025). Cognitive neuroscientist at University College London whose neuroimaging work on the hippocampus in autobiographical memory and spatial navigation framed much of the modern understanding of what the medial temporal lobe contributes to remembering. ORCID

Brenda Milner

(living). Founding figure of neuropsychology at the Montreal Neurological Institute; her examination of patient H.M. established the medial temporal lobe's role in memory, and her discovery of his intact motor-skill learning opened the study of nondeclarative memory (Scoville & Milner, 1957; Milner et al., 1968). Wikidata

Larry R. Squire

(living). Neuroscientist at the University of California, San Diego, who developed the influential taxonomy of declarative and nondeclarative memory and the standard model of systems consolidation that explains the temporal gradient (Squire, 2009; Squire et al., 2015). Wikidata

Endel Tulving

(1927–2023). Cognitive psychologist at the University of Toronto who drew the distinction between episodic and semantic memory, giving anterograde amnesia the vocabulary to describe precisely which form of declarative memory it impairs (Tulving, 2002). Wikidata

Faraneh Vargha-Khadem

(living). Cognitive neuroscientist at University College London whose studies of developmental amnesia showed that early hippocampal injury can devastate episodic memory while leaving substantial semantic learning intact, reshaping theories of how the two relate (Vargha-Khadem et al., 1997). ORCID

Frequently Asked Questions

What is the difference between anterograde and retrograde amnesia?

Anterograde amnesia is the inability to form new memories after an injury; retrograde amnesia is the loss of memories formed before it. They are defined relative to the same event and are dissociable — a person can have severe anterograde amnesia with little retrograde loss, or the reverse.

Can someone with anterograde amnesia lead a normal conversation?

Yes, at the moment. Immediate and working memory are intact, so a patient can follow and contribute to a conversation in real time. The deficit shows only afterwards: minutes later they typically have no memory that the conversation took place.

Is anterograde amnesia permanent?

It depends on the cause. Amnesia from structural damage — surgery, stroke, encephalitis — is usually enduring. Amnesia caused by drugs such as benzodiazepines or anaesthetics is temporary and lifts as the drug is cleared.

Can people with anterograde amnesia learn anything new?

They can learn skills, habits, and conditioned responses at a normal rate, even while unable to remember practising — this spared nondeclarative learning is one of the field's key findings. What they cannot reliably form is new conscious memory for facts and events.

Which part of the brain is damaged in anterograde amnesia?

Most often the medial temporal lobe, especially the hippocampus and adjacent cortex. Damage to midline diencephalic structures (the mammillary bodies and thalamus), as in Korsakoff syndrome, produces a clinically similar amnesia.

Why do older memories survive when recent ones are lost?

Because of consolidation: over time a memory becomes less dependent on the hippocampus as it is stabilised in the neocortex. A hippocampal lesion therefore tends to erase recent, still-dependent memories while sparing remote, fully consolidated ones — the temporal gradient.

Was patient H.M. able to remember anything about his own past?

Yes. H.M. retained his identity, personality, and most memories from well before his 1953 surgery; his retrograde loss was limited and graded. His profound deficit was anterograde — he could form almost no new lasting memories for the rest of his life.

Is drug-induced memory loss really the same condition?

It is the same functional deficit — a failure to encode new declarative memories with immediate memory intact — produced pharmacologically rather than by a lesion. Its reversibility makes it a clean demonstration that memory formation is a discrete, interruptible process.

References