2002 Nature PublishingGroup

R E V I E WS
NATURE REVIEWS | NEUROSCI ENCE VOLUME 3 | JUNE 2002 | 453
Everybody knows what stress is and nobody
knows what it is.
Hans Selye
1
The term stress originally used in the field of
engineering to describe a force that exerts physical
strain on a structure is generally defined in biological
systems as any condition that seriously perturbs the
physiological/psychological homeostasis of an organ-
ism (BOX 1). The profound physiological consequences
of stress were first shown empirically in Hans Selyes
seminal paper of 1936: A syndrome produced by
diverse nocuous agents
2
. Subsequent research
showed the broad range of adverse physiological
effects of stress in animals and humans, including
adrenal-gland enlargement, atrophy of the thymus
and lymph nodes, increased cardiovascular tone,
immune-system suppression and ulcerations
3
. In
recent decades, an important line of neuroscience
research has shown that stressful experiences can have
a negative impact on certain aspects of brain func-
tion. The field also developed an appreciation of the
fact that the acute response to stress (for example,
heightened cognition) is an adaptive mechanism,
enabling an organism to respond effectively to a real or
potential threat to its survival. However, the insidious
side of this survival strategy is that uncontrollable
stress can have severe adverse repercussions, including
deterioration in learning-and-memory capacity, an
exacerbation of the ageing-related decline in cognition,
and an enhanced susceptibility of hippocampal neu-
rons to atrophy or necrosis in response to metabolic
challenges
47
.
Recent reviews have presented a thorough discus-
sion of the neurobiological basis of the enhancement
of memory by emotional arousal
8,9
. However,
although stressful, and especially traumatic, experi-
ences are known to produce intense, long-lasting
memories of the events themselves
811
, it is also
evident that stress impairs subsequent attention
and memory
1214
, and can even induce profound
amnesia
1517
. So, to complete our understanding of
stressmemory interactions, it is important to address
how stress corrupts the memory-storage process. This
review is focused, in general, on the adverse effects of
stress on memory, with specific reference to how
stress affects hippocampus-dependent cognition and
synaptic plasticity. Our focus on the mechanisms that
underlie interactions between stress and the hippo-
campus should advance our understanding of how
stress interferes with our ability to accurately encode
information.
THE STRESSED HIPPOCAMPUS,
SYNAPTIC PLASTICITY AND
LOST MEMORIES
Jeansok J. Kim* and David M. Diamond

Stress is a biologically significant factor that, by altering brain cell properties, can disturb cognitive
processes such as learning and memory, and consequently limit the quality of human life.
Extensive rodent and human research has shown that the hippocampus is not only crucially
involved in memory formation, but is also highly sensitive to stress. So, the study of stress-
induced cognitive and neurobiological sequelae in animal models might provide valuable insight
into the mnemonic mechanisms that are vulnerable to stress. Here, we provide an overview of
the neurobiology of stressmemory interactions, and present a neuralendocrine model to
explain how stress modifies hippocampal functioning.
*Department of Psychology
and Interdepartmental
Neuroscience Program,
Yale University, New Haven,
Connecticut 06520-8205,
USA.

Departments of Psychology,
Pharmacology and
Neuroscience Program,
University of South Florida,
Tampa, Florida 33620-8200,
USA, and Veterans Hospital,
Medical Research, 13000
Bruce B. Downs Blvd,
Tampa, Florida 33612, USA.
e-mails: jeansok.kim@yale.
edu; ddiamond@chuma1.
cas.usf.edu
doi:10.1038/nrn849
2002 Nature PublishingGroup
CORTICOSTEROIDS
The principle glucocorticoids
that are synthesized by the
adrenal cortex and secreted in
response to stress (cortisol in
humans, corticosterone in rats).
454 | JUNE 2002 | VOLUME 3 www.nature.com/reviews/neuro
R E V I E WS
in the mammalian brain. One well-described neuro-
endocrine function of the hippocampus is to participate
in terminating the stress response through gluco-
corticoid-mediated negative feedback that inhibits the
hypothalamuspituitaryadrenal (HPA) axis
3,4
.
The hippocampus is enriched with the two classes of
corticosteroid receptors
26
: type I, mineralocorticoid
receptors (MRs); and type II, glucocorticoid receptors
(GRs). The adverse effects of stress on the hippocampus
seem to be mediated largely by the lower-affinity GRs,
which become heavily occupied with corticosteroids in
response to stress
2729
. In the rat hippocampus, cortico-
sterone binding to GRs has been shown to adversely
affect neuronal metabolism, cell survival, physiological
functions and neuronal morphology
37
. Consequently,
certain hippocampal functions such as learning
Stress and hippocampal memory
Although the stress response is a complex biochemical
cascade involving the release of diverse chemicals that can
affect various aspects of memory, brain structures and
physiological processes (FIG. 1), studies on isolated ele-
ments can provide valuable insights into the mnemonic
mechanisms that are especially vulnerable to stress
5
. One
brain structure that is crucially involved in both memory
and the neuroendocrine regulation of stress hormones
is the hippocampus. The hippocampus is a medial
temporal lobe structure that is necessary for the forma-
tion of stable declarative (or explicit) memory in
humans
1821
, and spatial (or relational/contextual)
memory in rodents
2125
. This sea-horse-shaped structure
is also a target of stress hormones, having one of the
highest concentrations of receptors for CORTICOSTEROIDS
Box 1 | What is stress?
The term stress is routinely used as if it were a well-defined psychophysiological measure, such as blood pressure or
galvanic skin response. On the contrary, stress is a nebulous description of diverse behavioural and physiological
responses. Some investigators have addressed the complexity of this issue
102,103
, but we are unaware of a universal definition
of stress with operationally defined terms that can be applied equally to rodents and people. A definition must address the
following two issues. First, whether stress occurs is determined not by the physical parameters of environmental
stimulation, but by how an organism perceives and reacts to the stimulus. For example, most people feel mild anxiety to
outright panic on being asked to speak in public, whereas others enjoy speaking in an open forum. So, a definition must
acknowledge that the same stimulus can be stressful to one person, but can give great pleasure to another. Second, there is
no unique physiological state that is stress specific. It is tacitly accepted that an elevation of glucocorticoid levels is an
indicator of a stress state, as rats that are exposed to a predator and people reporting high levels of stress tend to have
elevated levels of glucocorticoids
61,104
. However, glucocorticoid levels are also elevated by pleasurable activities, such as
feeding, exercise and sex
105108
. Moreover, stressor controllability has a profound influence on the impact of an aversive
experience on physiology and behaviour
56,109
, yet glucocorticoid levels are relatively insensitive to controllability
57,110,111
. So,
although glucocorticoid elevation can be useful in stress research, the level of this hormone is not a definitive measure of
the presence or magnitude of stress. Levels of other neuromodulators such as 5-hydroxytryptamine, noradrenaline,
dopamine and enkephalins can also increase under conditions that are considered to be stressful
109,112
, but we suggest
that the search for a neuromodulator that is sensitive only to a stress state will fail.
Given the problems in understanding what stress is, it is tempting to abandon the use of the term stress, and instead
focus on direct causeeffect relationships. For example, rats that were exposed to a cat showed impaired memory
62
and a
suppression of synaptic plasticity
61
. The quagmire that surrounds the stress definition is eliminated by assuming that it
was the cat, not stress, that affected memory and plasticity. However, it was still the rats perception and reaction to the
cat that produced a psychophysiological state that we interpret as stress, which then influenced hippocampal processing.
So, ignoring the issue of how to define stress does not make it go away.
With these caveats in mind, we offer a three-component definition of stress that can be applied broadly across species and
paradigms. First, stress requires heightened excitability or arousal, which can be operationally measured using
electroencephalography, behavioural (motor) activity or neurochemical (adrenaline, glucocorticoid) levels. Arousal,
however, can increase under either pleasurable or aversive conditions. So, second, the experience must also be perceived as
aversive. Aversiveness can be defined as an indication that the subject would avoid or attenuate the intensity of the stressor if
given the opportunity. Shock-avoidance conditioning in animals is stressful because it is arousing and the animals attempt
to avoid, or at least minimize, their exposure to the shock. For people who find public speaking unpleasant, the experience
satisfies the first two criteria of the stress definition, because they would find the experience arousing, and they would avoid
making the presentation if possible. Those who enjoy public speaking might find the experience arousing, but because they
dont want to avoid making the presentation, their experience would not be stressful.
The third component of the definition of stress is controllability. Two animals that are exposed to the same level of
electric shock can show arousal and attempt to make an avoidance response, but the experience can have very different
behavioural and physiological effects if one animal has control over the termination of the shock
56,109111
. Corresponding
work in people has shown that having control over an aversive experience has a profound mitigating influence on how
stressful the experience feels
113
. The element of control (and the related concept of predictability) in the stress definition
is the variable that ultimately determines the magnitude of the stress experience and the susceptibility of the individual
to develop stress-induced behavioural and physiological sequelae.
In summary, we define stress as a condition in which an individual is aroused by an aversive situation for example, a
hostile employer, unpaid bills, a predator or a pawshock. The magnitude of the stress and its physiological consequences
are influenced greatly by the individuals perception of its ability to control the presence or intensity of the stimulation.
2002 Nature PublishingGroup
NATURE REVIEWS | NEUROSCI ENCE VOLUME 3 | JUNE 2002 | 455
R E V I E WS
hippocampus-dependent recall tasks
33
. Moreover, the
administration of stress levels of cortisol to normal
human subjects selectively impairs verbal declarative
memory without affecting non-verbal (procedural)
memory
34
. In addition, people who are diagnosed with
both depression and hypercortisolaemia show hippo-
campus-dependent memory impairments; similarly,
patients with Cushings disease in which tumours in
the adrenal gland cause excess secretion of glucocorti-
coids have deficits in declarative memory
35
and
hippocampal atrophy
36
, which can be reversed after
treatment that reduces their cortisol levels
37
.
Consistent with the findings in humans, rats that are
exposed to stress or administered corticosterone (at doses
comparable to those observed in response to stress) have
deficits in spatial memory
3841
. Impairments of spatial
memory have also been observed in transgenic mice with
elevated corticosterone levels (due to the central over-
expression of CORTICOTROPIN-releasing factor, CRF)
42
.
Recently, stress has also been found to impair hippo-
campus-dependent object-recognition memory
43,44
.
Interestingly, uncontrollable stress has also been
shown to facilitate both delay
45
and trace
46
EYEBLINK
CONDITIONING in rats. Although both tasks require an intact
cerebellum
47
, the hippocampus is also necessary for trace
(but not for delay) eyeblink conditioning
48
. In contrast to
eyeblink conditioning, however, stress and corticosterone
seem to selectively affect hippocampus-dependent
contextual (but not hippocampus-independent tone)
fear conditioning in rats
49,50
. It seems, then, that stress
exerts complex effects on hippocampal memory
processing rather than uniform effects across different
and memory are susceptible to disruption by stress,
mediated in part by GR activation.
Studies carried out largely over the past two decades
have supported the idea that stress and stress hormones
impair hippocampus-dependent forms of memory in
both humans and animals
3,6,7,17,30,31
. For example,
patients with post-traumatic stress disorder (PTSD)
have hippocampal atrophy
32
and marked deficits in
CORTICOTROPIN
A polypeptide hormone, also
known as adrenocorticotropic
hormone (ACTH), which is
secreted by the adenohypophysis
of the pituitary. It stimulates the
synthesis and secretion of
corticosteroids, and the growth
of the adrenal cortex.
EYEBLINK CONDITIONING
A Pavlovian conditioning task in
which the subject learns to
respond with eyelid closure to a
conditioned stimulus (CS; tone
or light) that has been
contingently paired with an
unconditioned stimulus (US;
airpuff or shock to the eye). In
the delay paradigm, the CS
precedes and overlaps with the
US. In the trace paradigm, there
is an empty interval (or gap)
between the CS and the US.
PURSUIT ROTOR
A motor-skill learning task that
is used to assess the subjects
tracking errors and time on a
moving target.
DELAYED NONMATCHING-TO-
SAMPLE TASKS
In such recognition memory
tasks, presentation of a stimulus
is followed by a delay, after
which a choice is offered. In
matching tasks, the originally
presented stimulus must be
chosen; in nonmatching tasks, a
new stimulus must be selected.
With small stimulus sets, the
stimuli are frequently repeated,
thus becoming highly familiar.
So, typically, such tasks are more
readily solved by short-term or
working memory rather than by
long-term memory
mechanisms.
Stress
Synaptic
plasticity
Necrosis, ageing
and neurogenesis
Morphological
changes
Learning
and
memory
Neurochemical
systems
Figure 1 | Various effects on learning and memory as a
function of the magnitude of stress. At a mild level of stress,
certain neurochemical systems (for example, catecholamines,
opiates, glucocorticoids) might affect learning. As the level of
stress increases (in duration and/or intensity), several transient
and permanent changes are observed in the hippocampus,
including modifications in synaptic plasticity, morphological
changes, suppression of adult neurogenesis and neuronal
endangerment. These stress-associated changes in the brain
can potentially influence learning-and-memory processes.
Box 2 | Hippocampus-dependent and -independent memory systems
The hippocampus, which is one of the medial temporal lobe (MTL) structures, is thought to be crucial for declarative
(or explicit) memory, but not non-declarative (or implicit) memory, in humans
1821
. Declarative types of memory are
generally defined as those that involve consciously accessible records of facts and events; for example, recalling when,
where and from whom one learned to drive a car. Non-declarative types of memory are usually defined as those that
involve non-consciously expressed perceptualmotor skills; for example, the automatic act of driving a car. This concept
of memory dichotomy originated with Milner and colleagues initial discovery of preserved non-declarative memory
(mirror drawing, jigsaw-puzzle assembly, PURSUIT ROTOR and so on) in the profoundly amnesic patient H.M.
18,114
, which
was supported by subsequent neuropsychological studies
19
. Tulving
115
and others
116
further showed dual qualitatively
separable memory systems in normal subjects. Results from functional imaging studies
20
are also consistent with the
view that the hippocampus is crucially involved in declarative memory.
A similar (hippocampal/MTL-dependent and -independent) memory dichotomy seems to exist in animals. In 1978,
Mishkin
117
found that lesioning the hippocampus and the amygdala produced memory deficits in the DELAYED
NONMATCHING-TO-SAMPLE (DNMS) task (but not in the object-discrimination task) in non-human primates. Subsequent
studies found that lesions that were confined largely to the hippocampus impaired a trial-unique DNMS task (with long
delay intervals) in monkeys
19
. However, recently, there has been disagreement on whether the DNMS memory deficit is
due to damage to the hippocampus per se, or damage to its connected MTL cortical areas
118,119
.
In rodents, the hippocampus is generally considered to be involved in spatial (or relational) types of memory. OKeefe
and Dostrovsky
120
discovered place cells in the hippocampus, which fire preferentially when a rat is in a specific spatial
location in an environment. Subsequent studies showed that various hippocampal manipulations (lesions, drugs,
stimulations and gene mutations) alter performances on spatial-memory tasks, as assessed using the water maze, radial-
arm maze, circular maze and contextual conditioning in rats and mice
2125,121126
. In rabbits, the hippocampus is also
necessary for trace eyeblink conditioning (in which the conditioned stimulus and the unconditioned stimulus are
separated by a trace interval)
48
, but not for delay conditioning (in which the conditioned stimulus and the unconditioned
stimulus overlap)
47
. The selective involvement of the hippocampus in trace (but not delay) conditioning has also been
shown using fear conditioning in rats and mice
127,128
. As the hippocampus is involved in DNMS, trace conditioning, and
transitive inference and other memory tasks, which seem to have a non-spatial quality, there is a continuing debate as to
whether fundamental information processing in the hippocampus is spatial or relational in nature
129,130
.
2002 Nature PublishingGroup
456 | JUNE 2002 | VOLUME 3 www.nature.com/reviews/neuro
R E V I E WS
The restrainttailshock stress procedure that was
used in these studies was adapted from the learned-
helplessness paradigm
56
, in which animals cannot
perform any adaptive response to escape an aversive
experience. One consequence of learned helplessness is
an impairment of learning and memory. Interestingly,
rats that were given control over the termination of the
shock showed more robust hippocampal LTP than yoked
control animals
57
. So, the electrophysiological studies
parallel the behavioural findings
56
, indicating that it is the
uncontrollable psychological aspect of the shock, rather
than the pain associated with the shock, that exerted the
primary inhibitory effect on hippocampal LTP.
Likewise, other forms of psychological stress pro-
duced, for example, by forced exposure to brightly lit,
unfamiliar chambers impaired both LTP and primed-
burst potentiation (PBP, a low-threshold form of LTP) in
behaving rats
31,5860
. Recent work has shown that stress
produced by exposing rats to threatening environments
including unavoidable exposure to a predator (cat)
impaired both hippocampal PBP
61
and hippocampus-
dependent spatial memory
62
(FIG. 3). It is important to
note, however, that cat exposure blocked PBP, but did not
block LTP
61
. This finding is consistent with other studies
showing that PBP has a greater sensitivity than LTP to
modulation by ageing, drugs, hormones and neuro-
modulators, which indicates that PBP is a more sensitive
diagnostic of how behaviourally relevant variables affect
hippocampal processing
17
.
However, stress effects on hippocampal plasticity are
not limited to LTP and PBP. Recent studies have found
that stress and corticosterone enhance the induction of
homosynaptic long-term depression (LTD) in area CA1.
Specifically, restrainttailshock and exposure to a
brightly lit, novel chamber enhanced LTD, but impaired
LTP, both in vitro
63
and in vivo
58
, and the application of
corticosterone agonists to hippocampal slices enhanced
LTD and impaired LTP
64
. The fact that stress impairs LTP,
but enhances LTD, indicates that stress might alter the
physiological range of plasticity (or induce metaplasticity
a plasticity that affects ensuing synaptic plasticity),
such that subsequent synaptic plasticity favours the
development of depression over potentiation
5
.
Although the expression of synaptic plasticity can be
affected for days after a single stressful experience
53,54
,
the effects are not permanent. This conclusion is based
on work that showed that there is repeated impairment
and recovery of hippocampal PBP and spatial memory
in rats that are exposed to alternating stress-provoking
(unfamiliar) and stress-neutral (familiar) environ-
ments
17,59
. So, the modulation of hippocampal function-
ing by stress involves a dynamic process that is under
the continuous influence of an animals perception of
its environment.
Other effects of stress on the hippocampus
In addition to affecting synaptic plasticity and memory,
stress and corticosterone have been shown to alter
hippocampal dendritic morphology and to inhibit
neurogenesis in the adult brain, which can potentially
have an impact on memory-related functioning. These
hippocampus-based memories. It might be relevant to
consider the generalization that stress enhances hippo-
campus-dependent fear-related learning and memory,
and impairs the processing of spatial information that is
acquired outside the fear-conditioning context
6,17
(BOX 2).
Stress and hippocampal plasticity
For the past three decades, the primary physiological
model of memory has been long-term potentiation
(LTP) a sustained enhancement of synaptic efficacy
that is produced by a brief (less than one second)
TETANIC STIMULATION of excitatory afferent fibres (BOX 3).
If the idea that changes in synaptic efficacy underlie
information storage (Hebbs postulate
51
) is valid, then
the finding that stress impairs hippocampus-dependent
memory would predict that stress should also inter-
fere with the induction of hippocampal LTP (or LTP-
like processes). Extensive observations from in vitro
and in vivo electrophysiological studies are consistent
with this prediction, in that stress and stress hormones
do indeed impair LTP. The first demonstration of a
stress-induced suppression of LTP induction was
reported in 1987 by Thompson, Levine and col-
leagues
52
. These investigators found that hippocampal
slices prepared from adult rats that had experienced
unpredictable and inescapable restrainttailshock
stress showed marked impairments of LTP in the CA1
pyramidal cell region (FIG. 2). Subsequent studies
showed that stress-induced LTP impairment lasts for
at least 48 hours in rats
53
and 24 hours in mice
54
, and
that stress also reduces LTP in the dentate gyrus of the
hippocampus
55
.
TETANIC STIMULATION
A train of stimuli in which
afferent axons are briefly
activated at high frequency. In
LTP experiments, a 1-s train of
pulses delivered at a frequency of
100 Hz is commonly used to
potentiate transmission.
SCHAFFER COLLATERALS
Axons of the CA3 pyramidal
cells of the hippocampus that
form synapses with the apical
dendrites of CA1 neurons.
Box 3 | Hippocampal synaptic plasticity and memory
Long-lasting, activity-dependent changes in the strength of synaptic transmission, for
which long-term potentiation (LTP) is the leading candidate synaptic model, have long
been postulated to be fundamental for memory storage (Hebbs postulate
51
). LTP was
first discovered in the rabbit hippocampus by Bliss and Lomo
131
in 1973. Since then, LTP
has been shown in various brain structures and neural tissues from rats, mice and,
recently, aplysia
68,69,132
. In addition to its longevity, LTP is rapidly induced, strengthened
by repetition, shows specificity and associativity
68,69
, and occurs prominently in the
hippocampus a structure that is implicated in learning and memory
1821
. In theory,
LTP alone provides an incomplete synaptic model of learning, because the continued
potentiation of synapses will eventually lead to a saturation of all excitatory plasticity
133
.
Decreases in synaptic efficacy are also needed to reset the synapses, and are accounted for
by long-term depression (LTD) (or depotentiation, a reversal of LTP), which has also
been shown in the hippocampus after the low-frequency stimulation of afferent fibres
134
.
Both LTP and LTD at the SCHAFFER COLLATERAL/commissuralCA1 synapses seem to
be mediated by mechanisms that are triggered by Ca
2+
influx following the activation of
NMDA(N-methyl-D-aspartate) receptors
68,134
.
Several lines of evidence are consistent with the LTP (or synaptic-plasticity) hypothesis
of learning. For example, pharmacological studies have shown that NMDA receptor
antagonists, which block LTP induction in the hippocampus, also interfere with
hippocampus-dependent spatial learning and the learning of odour-discrimination
tasks
24,69
. Neurophysiological studies have shown learning-related increases in
hippocampal cell excitability
135
, indicating that LTP or LTP-like changes are operative
during learning. Recent studies from gene knockout/transgenic mice that show deficits in
hippocampal LTP and spatial memory, and alterations in the properties of hippocampal
place cells
121123,136
, provide further support. However, despite extensive correlational
evidence, some questions remain
137
, so the LTP/synaptic-plasticity hypothesis lacks
universal acceptance.
2002 Nature PublishingGroup
NATURE REVIEWS | NEUROSCI ENCE VOLUME 3 | JUNE 2002 | 457
R E V I E WS
of new granule cells in young and old rats
65
. However, it
is important to note that there is an ongoing disagree-
ment about the standard for verifying adult neurogenesis
after various experimental manipulations
73
.
The mechanisms that underlie stress effects on
granule cell proliferation are not fully understood, but
the absence of MRs and GRs on granule cell precur-
sors implies that glucocorticoids influence this process
indirectly, perhaps through an increase in glutamate-
mediated neurotransmission. Just as stress affects
hippocampus-dependent learning and LTP through
glutamate release and activation of the NMDA recep-
tor
63
, stress also seems to inhibit granule cell prolifera-
tion through NMDA receptor activation
65
. Once again,
the NMDA receptor serves as a common site of action
in both the constructive (learning and memory) and
destructive (inhibiting cell proliferation) effects of
glucocorticoids in the hippocampus.
Neurochemical mediators of stress effects
Stress activates an ensemble of neurochemical responses
including secretions of glucocorticoids (such as
corticosterone), catecholamines (such as adrenaline)
and opiates from the HPA axis and in the autonomic
nervous system
37
. So, the impairment of LTP after an
uncontrollable stress experience can be influenced by
topics have been reviewed extensively elsewhere
7,65
, and
will be addressed here only briefly.
Stress and dendritic morphology. McEwen and col-
leagues
7
have carried out an extensive analysis of the
effects of chronic stress or corticosterone administra-
tion on dendritic morphology, and have shown that
daily restraint stress, or corticosterone injections for
three weeks, produced atrophy of the apical dendrites
of CA3 pyramidal neurons. Others have shown that the
effects of chronic stress on cellular morphology can
also occur outside area CA3 for example, in the den-
tate gyrus and area CA1
66
. Stress- and corticosterone-
induced atrophy can be blocked by drugs that reduce
excitatory-amino-acid neurotransmission (phenytoin),
reduce extracellular 5-hydroxytryptamine (5-HT) lev-
els (tianeptine), or reduce general excitability through
an enhancement of GABA (-aminobutyric acid) tone
by a benzodiazepine (adinazolam)
67
, indicating that
multiple neurotransmitter and hormone systems are
involved in producing stress-induced dendritic atrophy
in the hippocampus.
One element that is common to many studies that
show stress effects on the hippocampus is the involve-
ment of NMDA (N-methyl-D-aspartate) receptors. The
crucial role of NMDA-receptor-mediated Ca
2+
ion
influx (and the triggering of particular intracellular
messenger cascades) in synaptic plasticity and memory
is well described
68,69
. However, NMDA receptor activa-
tion also seems to be a central component of the mani-
festation of the adverse short- and long-term effects of
stress on hippocampal morphology. For example, ele-
vated levels of Ca
2+
have been shown to accompany
corticosterone-enhanced cell death, which can be alle-
viated by NMDA receptor antagonists
7
. It is perhaps
ironic that hippocampal pyramidal cells deploy NMDA
receptors to store information, although the same
mechanism, if chronically overstimulated, can bring
about their demise.
Stress and adult neurogenesis. The production of gran-
ule cells in the adult dentate gyrus has been documented
in a range of species, including the rat, marmoset, rhe-
sus monkey, tree shrew and human
70
. The functional
significance of the continued production of these cells
throughout life is not fully understood, but it seems that
this process is related to the demands of learning and
memory
71
. For example, under conditions in which
hippocampus-dependent learning occurs, there is an
increase in the number of adult-generated granule cells
and in the duration for which newly formed granule
cells survive
70,72
.
The production of new granule cells seems to be
inhibited by stress or by experimentally produced eleva-
tions of corticosterone. Acute stress has been shown to
decrease the proliferation of granule cells in adult rats,
tree shrews and marmosets
65
. It has also been shown that
chronic (daily) stress can result in a persistent elevation
of cortisol levels and a decrease in the production
of granule cells in adult tree shrews. Conversely, removal
of endogenous corticosteroids stimulated the production
300
250
200
150
100
50
0 5 10 15
Minutes after tetanus
Control
Stress
P
e
r
c
e
n
t
a
g
e

o
f

p
r
e
-
t
e
t
a
n
u
s

v
a
l
u
e
20 25 30
Recording electrode
CA1
S
c
h
Stimulating
electrode
Figure 2 | Uncontrollable restrainttailshock stress impairs
LTP in the hippocampus in vitro. a | Schematic diagram of a
transverse hippocampal slice, showing the arrangement of
recording and stimulating electrodes. b | Post-tetanus
population field potentials recorded from the cell-body layer of
the CA1 field in response to test stimulation of the Schaffer
collateral/commissural fibres (Sch). Tetanizing stimulation long-
term potentiation induced in hippocampal slices prepared from
unstressed control rats but not from stressed rats. Data replotted
from REF. 52.
2002 Nature PublishingGroup
458 | JUNE 2002 | VOLUME 3 www.nature.com/reviews/neuro
R E V I E WS
corticosterone and LTP. Activation of MRs (which
would occur at low-to-intermediate levels of cortico-
sterone) has been shown to increase the magnitude of
LTP, whereas activation of GRs (which would occur at
stress levels of corticosterone) attenuated LTP but
enhanced LTD
64
. The GR-specific effects on LTP have
been confirmed in studies showing that administration
of the GR antagonist RU38486 prevented the stress-
induced impairments of LTP
76
. These findings indicate
that low-to-intermediate levels of corticosterone enhance
hippocampal synaptic plasticity by stimulating the high-
affinity MR (perhaps with a small degree of activation of
GRs)
28
. During intense stress, however, the high levels
of corticosterone produce a greater activation of the GR,
which results in a profound inhibitory effect on hippo-
campal plasticity. The relatively greater importance of
GRs (compared with MRs) in mediating the adverse
effects of corticosterone on the hippocampus is also
indicated by findings that GR agonists and MR antago-
nists can impair spatial memory
6
, and that a point
mutation in the mouse GR gene blocks the exogenous
effects of corticosterone on spatial memory
29
.
Corticosterone can also alter the intrinsic properties
of neurons in the hippocampus. Bath application of
corticosterone has been shown to prolong the after-
hyperpolarization (AHP) of CA1 pyramidal neurons by
increasing the level of internal Ca
2+
and thereby acti-
vating Ca
2+
-gated K
+
channels
77,78
. Correspondingly,
corticosterone also promotes the expression of genes
that encode channels that enhance Ca
2+
influx
79
. Recent
work has shown that stress levels of corticosterone
initially enhance MR-mediated hippocampal cellular
excitability, which is then overtaken by a GR-mediated
suppression of cellular activity
78
, which could impede
the development of LTP.
However, an increase in the corticosterone concen-
tration per se does not seem to be sufficient to affect
hippocampal LTP. This conclusion is based on findings
that LTP can still be inhibited by stress in rats that have
been depleted of corticosterone as a result of adrenalec-
tomy
80
. Moreover, in normal animals that were adminis-
tered dexamethasone (a synthetic glucocorticoid that
blocks the release of corticosterone), stress-induced
impairments of LTP still occurred
81
. More recent work
has provided three different conditions in which there
was an elevation of serum corticosterone levels in the
absence of an inhibitory effect on hippocampal process-
ing: first, rats with lesions of the amygdala did not
develop a stress-induced suppression of LTP, despite the
fact that the lesioned animals had elevated cortico-
sterone levels
82
; second, the exogenous administration of
stress levels of corticosterone, under otherwise non-
stress conditions, did not produce a spatial-memory
impairment in rats
83
; and third, male rats that were
given access to a sexually receptive female had elevations
of serum corticosterone, but no spatial-memory deficit
83
.
So, elevated corticosterone, in the absence of fear and/or
an intact amygdala is not sufficient to produce deficits in
hippocampal processing.
Other work has implicated non-glucocorticoid
neuromodulators in the manifestation of stress effects
the interaction of different types of neuromodulator.
There is strong support for the idea that one of the pri-
mary neuromodulators of LTP and memory is cortico-
sterone. The effects of corticosterone on hippocampal
plasticity are complex, in that there is a biphasic rela-
tionship between the level of corticosterone and the
magnitude of LTP, such that both low (through adrena-
lectomy) and high (produced by stress or exogenous
administration) levels of corticosterone are associated
with LTP impairments, with maximal LTP occurring in
animals with an intermediate level of corticosterone
74,75
.
Subsequent studies substantiated the mechanistic
basis of the inverted-U-shaped relationship between
Home
a
b
0
15
30
45
60
75
90
Chamber
P
e
r
c
e
n
t
a
g
e

c
h
a
n
g
e
M
e
m
o
r
y

e
r
r
o
r
s
Cat
*
WM-Home
0
1.0
0.5
1.5
2.0
2.5
3.0
3.5
RM-Cat WM-Cat
*
Figure 3 | Exposing a rat to a natural predator (a cat)
impairs cognitive and electrophysiological measures of
hippocampal functioning. A rat with no previous exposure
to a cat shows an innate freezing (fear) response in the
presence of a cat. a | After 75 minutes of exposure to the cat,
rats showed a complete suppression of synaptic plasticity
(primed-burst potentiation, or PBP) in area CA1 of the
hippocampus (bar labelled Cat). By contrast, rats placed in the
chamber for 75 minutes without the cat (Chamber) showed as
much PBP as rats that were undisturbed in their home
environment (Home). b | Rats have excellent spatial working
memory under non-stress conditions (WM-Home) when tested
in the radial-arm water maze. However, rats that were exposed
to a cat showed a selective impairment of hippocampus-
dependent working memory (WM-Cat), with no adverse effect
of stress on hippocampus-independent reference memory
(RM-Cat). Data replotted from REFS 61,62,83.
2002 Nature PublishingGroup
NATURE REVIEWS | NEUROSCI ENCE VOLUME 3 | JUNE 2002 | 459
R E V I E WS
in vivo LTP of the perforant path/dentate gyrus
96
. Abe
and colleagues
96
have found that infusions of NMDA
receptor antagonists into the basolateral nucleus of the
amygdala (BLA) impair dentate-gyrus LTP (without
affecting hippocampal baseline activity), a finding that
indicates that amygdalar NMDA receptors might be
involved in regulating hippocampal LTP. Moreover,
high-frequency stimulation of the BLA augments
dentate-gyrus LTP
96
. But more specifically, stimulation
of the amygdala (and exposing rats to stress) has a time-
dependent, biphasic (an immediate excitatory and a
longer-lasting inhibitory) effect on hippocampal LTP
97
.
Interestingly, lesions of the amygdala also prevent stress-
induced increases in catecholamine turnover
98
, which is
thought to impair the working-memory function of the
prefrontal cortex
99
.
Anatomically, the amygdala is connected both
directly (amygdalo-hippocampal bundles arise from the
magnocellular and parvicellular divisions of the baso-
lateral amygdala and terminate in area CA1 and the
subiculum) and indirectly (through the entorhinal cor-
tex) to several hippocampal regions
100
, providing various
routes by which it can influence hippocampal function-
ing. Collectively, there is a growing literature indicating
that the full expression of stress effects on the hippo-
campus requires co-activation of the amygdala and
hippocampus, in concert with the actions of neuro-
modulators (such as corticosterone, 5-HT, opiates and
CRF) directly on the hippocampus.
Putting it all together: a system-level model
In this review, we have emphasized that local effects of
neuromodulators on the hippocampus act in conjunc-
tion with influences from the amygdala to alter hippo-
campus-dependent memory and synaptic plasticity. A
similar idea of a modulating or driving input has been
proposed for the interactions between neocortical and
subcortical pathways to explain alterations in receptive-
field properties of neurons in the visual cortex
101
. On the
basis of diverse observations of plasticity in different
systems, we have developed a model that describes
a subset of the neuralendocrine interactions that are
necessary preconditions for producing stress effects on
hippocampal functioning (FIG. 4). There are five specific
premises to the model:
First, stress hormones can directly affect the hippo-
campus (denoted in FIG. 4 as the h(s) input to the
hippocampus: h(s) > 0, stress; h(s) = 0, no stress).
Supporting evidence: the hippocampus is enriched with
corticosteroid receptors
4,6,26
; there is a biphasic relation-
ship between the level of corticosterone and LTP
74,75
;
corticosterone affects the intrinsic properties of hippo-
campal neurons
77,78
, and reduces LTP and enhances
LTD in vitro
90
.
Second, stress hormones can directly or indirectly
influence the amygdala (denoted as the g(s) input to
the amygdala: g(s) > 0, stress; g(s) = 0, no stress).
Supporting evidence: the amygdala (BLA) is also
enriched with corticosteroid receptors
26
; the adrenal
hormone adrenaline, which is released during stress,
activates the vagus nerve to the nucleus solitary, which
on hippocampal functioning. First, the administration
of naltrexone (an opioid antagonist) has been reported
to block stress effects on both learning
84
and LTP
85
, indi-
cating that stress-induced opioid peptides might be
involved in mediating stress effects on synaptic plastic-
ity. Second, the 5-HT system seems to be involved in
mediating stress effects on the hippocampus. Stress ele-
vates 5-HT levels in the hippocampus
86
, and exoge-
nously applied 5-HT can inhibit CA1 LTP in vitro
87
and
in vivo in freely moving rats
88
, indicating that endoge-
nous 5-HT exerts a suppressant effect on synaptic plas-
ticity that underlies memory formation. Third, NMDA
receptor antagonists block the effects of stress on learn-
ing
89
, LTP
63
and the GR-mediated impairment of LTP
90
.
So, it seems that, in addition to glucocorticoids, other
neuromodulators are crucially involved in the effects of
stress on hippocampal plasticity.
Amygdala, stress and hippocampal functioning
An accumulating body of evidence indicates that the
amygdala is central to emotional learning
8,9
, and to
manifesting stress-related behaviours and changes in
hippocampal functioning. Lesions or pharmacological
suppression of the amygdala prevent stress-induced
gastric erosion
91
, analgesia
92
and anxiety-like behav-
iour
93
, block the modulatory effects of drugs on
hippocampus-dependent memory
94,95
, and impair
HPA axis
g(s)
h(s)
(m,h(s)); m = 0, h(s) > 0 (m,h(s)); m > 0, h(s) > 0
Stress hormones,
opioids, others
Stress
or
Hippocampus
Amygdala
or
m > 0 m = 0
Figure 4 | A hypothetical model of how neuromodulatoramygdala interactions mediate
stress effects on hippocampal plasticity. Stress activates the hypothalamuspituitaryadrenal
(HPA) axis, resulting in elevated secretion of glucocorticoids, catecholamines, enkephalins and
other neuromodulators, which affect the hippocampus and amygdala. The influence that stress
neuromodulators (variable s) have on the hippocampus is a function of their direct effects on the
hippocampus, denoted by h(s), and the effects of these neuromodulators on the amygdala,
denoted by g(s). The output from the amygdala (variable m) is a crucial component of the stress-
induced modulation of hippocampal plasticity. The interaction between mand h(s) is denoted by
f(m,h(s)). The model shows that when there is an experimentally induced reduction in the
amygdalar input to the hippocampus (as a result of damage or inactivation of the amygdala,
m= 0), plasticity in the hippocampus is intact under stress conditions (upper right). However, with
intact amygdalar input in response to stress (m> 0), plasticity in the hippocampus is disturbed
(for example, impaired long-term potentiation (LTP) and enhanced long-term depression (LTD);
lower right). The green circles on the matrices represent synapses with normal capacity to
generate plasticity (for example, normal LTP), thereby accommodating normal hippocampus-
dependent memory. The pink circles represent synapses with altered properties of plasticity (for
example, impaired LTP and enhanced LTD), which impair subsequent hippocampus-dependent
memory. The conjunction of inputs mand h(s) is required for plasticity in the hippocampus to be
disrupted. The model predicts that stress effects on the hippocampus can be prevented by
reducing the amygdalar input (upper part of the figure) to the hippocampus. Conversely, the
model predicts that these stress effects can be exacerbated by increasing the amygdalar input
to the hippocampus.
2002 Nature PublishingGroup
460 | JUNE 2002 | VOLUME 3 www.nature.com/reviews/neuro
R E V I E WS
the hippocampus. Conversely, manipulations of the
amygdala that enhance its output (which would
increase the minput to the hippocampus) during stress
would exacerbate stress effects on hippocampal func-
tioning. So, our model predicts that pharmacological
manipulations of the amygdala, such as infusing
-adrenoceptor agonists into the BLA (which would
normally enhance hippocampal memory formation
95
)
during stress, could actually exacerbate stress effects on
subsequent hippocampal plasticity and memory. By
contrast, infusing -adrenoceptor antagonists into the
amygdala (which would normally impair hippocampal
memory formation
95
) during stress would attenuate
stress effects on ensuing hippocampal functions. These
predictions should also apply to other drugs for
example, GABA receptor and opioid receptor antago-
nists and agonists, which increase or decrease (respec-
tively) noradrenaline release in the amygdala to
enhance or impair hippocampal memory formation
under normal conditions.
Concluding remarks
Stress, a naturalistic factor that contributes to memory
impairments, constitutes a significant problem in
todays increasingly populous and long-living society.
The hippocampus a structure that is involved in
memory processing and regulating the HPA axis
seems to be particularly susceptible to exposure to
uncontrollable stress. Here, we have described the
manifold effects that stress exerts on the hippocampus,
and some neurochemical systems that might be
involved in mediating stress effects. One aspect of the
modulation of hippocampal functioning by stress
is described by our model, which takes into account
the effects of hormones, such as corticosterone, on the
hippocampus, acting directly as well as through an
obligatory pathway that includes amygdala activation.
Stress effects on the hippocampus and on memory are
undoubtedly more complex than the simple neural
endocrine model that we have proposed, and are likely
to involve other neural structures and neuromodulators.
Despite its simplicity, this model helps to illustrate the
importance of studying neural structures in isolation
in vitro, in addition to the interacting components of
neural systems, to gain an understanding of how stress
affects the brain and cognition.
in turn innervates and influences the amygdala
95
; the
local application of noradrenaline or corticosterone can
modulate the formation of emotional memory
94,95
.
Third, amygdalar projections to the hippocampus
can modify hippocampal functioning (denoted as the
minputs to the hippocampus: m> 0, stress with func-
tioning amygdala; m= 0, stress with inactivated amyg-
dala). Supporting evidence: the BLA projects both
directly and indirectly (through the entorhinal cortex)
to the hippocampus (including area CA1)
100
; pharma-
cological manipulations of the amygdala modulate
hippocampus-dependent memory
94,95
; electrical stimu-
lation of the amygdala can affect the development of
hippocampal LTP
96,97
.
Fourth, the CA1 area of the hippocampus can func-
tion normally without the amygdala under non-stress
conditions. Supporting evidence: amygdala-lesioned rats
are unimpaired on a hippocampus-dependent spatial-
memory WATER-MAZE TASK
82
; hippocampal slices obtained
from intact or amygdala-lesioned rats show LTP
82
.
Fifth, the hippocampus requires both m and h(s)
inputs to undergo stress-induced alterations in synaptic
plasticity (which can alter ensuing memory); and the
minput modulates or drives the h(s) input, which deter-
mines the magnitude of stress effects in the hippo-
campus. Supporting evidence: the amygdala is necessary
for intra-hippocampus-administered drugs to modulate
(enhance or impair) memory and for mediating mem-
ory-modulatory effects of stress hormones
95
; lesions of
the amygdala (no minput to the hippocampus) prevent
stress effects on hippocampal LTP and spatial memory
82
;
fear impairs hippocampal LTP and spatial memory, but
increases in corticosterone without co-activation of the
hippocampus and amygdala (produced by exogenous
corticosterone or by copulation in male rats) do not
impair hippocampus-dependent memory
83
.
The key assumption of our model that alterations
in hippocampal functioning after stress are due to exces-
sive activity exerted by the amygdala on the hippo-
campus provides clear empirical predictions of direct
relevance to examining the role of the amygdala in
mediating stress effects on the functioning of the hippo-
campus. Specifically, experimental manipulations that
block or reduce the output from the amygdala (which
would reduce the minput to the hippocampus) during
stress should prevent or reduce the effects of stress on
WATER-MAZE TASK
A learning task in which an
animal is placed in a pool filled
with opaque water and has to
learn to escape to a hidden
platform that is placed at a
constant position. The animal
must learn to use distal cues, and
the spatial relationship between
them and the platform.
1. Selye, H. The evolution of the stress concept. Am. Psychol.
61, 692699 (1973).
2. Selye, H. A syndrome produced by diverse nocuous agents.
Nature 138, 32 (1936).
A landmark paper for the field of stress research,
which, in Seyles concept of the general adaptation
syndrome, laid an empirical foundation for the
connection between stress and disease.
3. Sapolsky, R. M. Stress, the Aging Brain, and the
Mechanisms of Neuron Death (MIT Press, Cambridge,
Massachusetts, 1992).
A comprehensive synthesis of various studies of how
stress and stress hormones influence the brain and its
functions.
4. McEwen, B. S. & Sapolsky, R. M. Stress and cognitive
function. Curr. Opin. Neurobiol. 5, 205216 (1995).
5. Kim, J. J. & Yoon, K. S. Stress: metaplastic effects in the
hippocampus. Trends Neurosci. 21, 505509 (1998).
6. de Kloet, E. R., Oitzl, M. S. & Joels, M. Stress and cognition:
are corticosteroids good or bad guys? Trends Neurosci. 22,
422426 (1999).
A synthesis of behavioural, electrophysiological and
molecular studies that assesses the complexity of
corticosteronememory interactions, providing
insight into how stress enhances and impairs
memory.
7. McEwen, B. S. Effects of adverse experiences for brain
structure and function. Biol. Psychiatry 48, 721731 (2000).
8. Cahill, L. & McGaugh, J. L. Mechanisms of emotional
arousal and lasting declarative memory. Trends Neurosci.
21, 294299 (1998).
9. LeDoux, J. E. Emotion circuits in the brain. Annu. Rev.
Neurosci. 23, 155184 (2000).
10. Jacobs, W. J. & Nadel, L. The first panic attack: a
neurobiological theory. Can. J. Exp. Psychol. 53, 92107
(1999).
11. Bremner, J. D. & Vermetten, E. Stress and development:
behavioral and biological consequences. Dev.
Psychopathol. 13, 473489 (2001).
12. Payne, J. D., Nadel, L., Allen, J. J., Thomas, K. G. & Jacobs,
W. J. The effects of experimentally induced stress on false
recognition. Memory 10, 16 (2002).
13. Gilbertson, M. W., Gurvits, T. V., Lasko, N. B., Orr, S. P. &
Pitman, R. K. Multivariate assessment of explicit memory
function in combat veterans with posttraumatic stress
disorder. J. Trauma Stress 14, 413432 (2001).
14. Heffelfinger, A. K. & Newcomer, J. W. Glucocorticoid effects
on memory function over the human life span. Dev.
Psychopathol. 13, 491513 (2001).
15. Bremner, J. D., Krystal, J. H., Charney, D. S. & Southwick,
S. M. Neural mechanisms in dissociative amnesia for
childhood abuse: relevance to the current controversy
surrounding the false memory syndrome. Am. J. Psychiatry
153, 7182 (1996).
2002 Nature PublishingGroup
NATURE REVIEWS | NEUROSCI ENCE VOLUME 3 | JUNE 2002 | 461
R E V I E WS
16. Joseph, R. The neurology of traumatic dissociative
amnesia: commentary and literature review. Child Abuse
Negl. 23, 715727 (1999).
17. Diamond, D. M., Park, C. R., Puls, M. J. & Rose, G. M. in
Neuronal Mechanisms of Memory Formation (ed. Holscher,
C.) 379403 (Cambridge Univ. Press, New York, 2001).
18. Scoville, W. B. & Milner, B. Loss of recent memory after
bilateral hippocampal lesions. J. Neurol. Neurosurg.
Psychiatry 20, 1121 (1957).
A milestone paper that indicated a crucial role for the
hippocampus in the formation of stable memories in
humans.
19. Squire, L. R. & Zola-Morgan, S. The medial temporal lobe
memory system. Science 253, 13801386 (1991).
20. Gabrieli, J. D. E. Cognitive neuroscience of human memory.
Annu. Rev. Psychol. 49, 87115 (1998).
21. Eichenbaum, H. A corticalhippocampal system for
declarative memory. Nature Rev. Neurosci. 1, 4150 (2000).
22. OKeefe, J. & Nadel, L. The Hippocampus as a Cognitive
Map (Oxford Univ. Press, New York, 1978).
23. Morris, R. G. M., Garrud, P., Rawlins, J. N. P. & OKeefe, J.
Place navigation impaired in rats with hippocampal lesions.
Nature 297, 681683 (1982).
24. Morris, R. G. M., Anderson, E., Lynch, G. S. & Baudry, M.
Selective impairment of learning and blockade of long-term
potentiation by an N-methyl-D-aspartate receptor
antagonist, APV. Nature 319, 774776 (1986).
The first demonstration that pharmacological
blockade of LTP in the hippocampus also impairs
hippocampus-dependent spatial memory.
25. Moser, E. I., Krobert, K. A., Moser, M.-B. & Morris, R. G. M.
Impaired spatial learning after saturation of long-term
potentiation. Science 281, 20382042 (1998).
26. Reul, J. M. H. M. & de Kloet, E. R. Two receptor systems for
corticosterone in rat brain: microdistribution and differential
occupation. Endocrinology 117, 25052511 (1985).
This paper showed the differential distribution of two
classes of corticosteroid receptors in various brain
structures.
27. Joels, M. & de Kloet, E. R. Mineralocorticoid receptor-
mediated changes in membrane properties of rat CA1
pyramidal neurons in vitro. Proc. Natl Acad. Sci. USA 87,
44974498 (1990).
28. Conrad, C. D., Lupien, S. J. & McEwen, B. S. Support for a
bimodal role for type II adrenal steroid receptors in spatial
memory. Neurobiol. Learn. Mem. 72, 3946 (1999).
This study provides insight into the interactions
between corticosterone receptor activation/
antagonism and memory, and indicates that optimal
memory occurs with the occupation of type I
receptors (MR) and a small degree of type II receptor
(GR) occupation.
29. Oitzl, M. S., Reichardt, H. M., Joels, M. & de Kloet, E. R.
Point mutation in the mouse glucocorticoid receptor
preventing DNA binding impairs spatial memory. Proc. Natl
Acad. Sci. USA 98, 1279012795 (2001).
30. Kirschbaum, C., Wolf, O. T., May, M., Wippich, W. &
Hellhammer, D. H. Stress- and treatment-induced elevations
of cortisol levels associated with impaired declarative
memory in healthy adults. Life Sci. 58, 14751483 (1996).
31. Diamond, D. M. & Park, C. R. Predator exposure produces
retrograde amnesia and blocks synaptic plasticity. Progress
toward understanding how the hippocampus is affected by
stress. Ann. NY Acad. Sci. 911, 453455 (2000).
32. Bremner, J. D. Alterations in brain structure and function
associated with post-traumatic stress disorder. Semin. Clin.
Neuropsychiatry 4, 249255 (1999).
33. Utto, M., Vasterling, J. J., Brailey, K. & Sutker, P. B. Memory
and attention in combat related posttraumatic-stress-disorder
(PTSD). J. Psychopathol. Behav. Assess. 15, 4352 (1993).
This paper presents evidence that Vietnam War
veterans diagnosed with PTSD show deficits in word-
recall memory tasks.
34. Newcomer, J. W., Craft, S., Hershey, T., Askins, K. &
Bardgett, M. E. Glucocorticoid-induced impairment in
declarative memory performance in adult human.
J. Neurosci. 14, 20472053 (1994).
35. Starkman, M. N., Gebarski, S. S., Berent, S. & Schteingart,
D. E. Hippocampal formation volume, memory dysfunction,
and cortisol levels in patients with Cushings syndrome. Biol.
Psychiatry 32, 756765 (1992).
36. Sapolsky, R. M. Glucocorticoids and hippocampal atrophy
in neuropsychiatric disorders. Arch. Gen. Psychiatry 57,
925935 (2000).
37. Starkman, M. N. et al. Decrease in cortisol reverses human
hippocampal atrophy following treatment of Cushings
disease. Biol. Psychiatry 46, 15951602 (1999).
38. Luine, V. N., Spencer, R. L. & McEwen, B. S. Effects of
chronic corticosterone ingestion on spatial memory
performance and hippocampal serotonergic function. Brain
Res. 616, 6570 (1993).
39. Diamond, D. M., Park, C. R., Heman, K. L. & Rose, G. M.
Exposing rats to a predator impairs spatial working memory
in the radial arm water maze. Hippocampus 9, 542552
(1999).
40. Diamond, D. M., Fleshner, M., Ingersoll, N. & Rose, G. M.
Psychological stress impairs spatial working memory:
relevance to electrophysiological studies of hippocampal
function. Behav. Neurosci. 110, 661672 (1996).
This study made the first connection between stress
effects on LTP and memory by showing that stress
selectively impaired hippocampus-dependent
(working) memory, but had no effect on
hippocampus-independent (reference) memory.
41. de Quervain, D. J.-F., Roozendaal, B. & McGaugh, J. L.
Stress and glucocorticoids impair retrieval of long-term
spatial memory. Nature 394, 787790 (1998).
42. Heinrichs, S. C. et al. Learning impairment in transgenic
mice with central overexpression of corticotropin-releasing
factor. Neuroscience 74, 303311 (1996).
43. Clark, R. E., Zola, S. M. & Squire, L. R. Impaired recognition
memory in rats after damage to the hippocampus.
J. Neurosci. 20, 88538860 (2000).
44. Baker, K. B. & Kim, J. J. Effects of stress and hippocampal
NMDA receptor antagonism on recognition memory in rats.
Learn. Mem. 9, 5865 (2002).
45. Shors, T. J., Weiss, C. & Thompson, R. F. Stress-induced
facilitation of classical conditioning. Science 257, 537539
(1992).
This study showed that stress that impairs LTP in the
hippocampus enhances hippocampus-independent,
but cerebellum-dependent, delay eyeblink
conditioning in rats.
46. Beylin, A. V. & Shors, T. J. Stress enhances excitatory
trace eyeblink conditioning and opposes acquisition of
inhibitory conditioning. Behav. Neurosci. 112, 13271338
(1998).
47. Anderson, B. J. & Steinmetz, J. E. Cerebellar and brainstem
circuits involved in classical eyeblink conditioning. Rev.
Neurosci. 5, 251273 (1994).
48. Solomon, P. R., Vander Schaaf, E. R., Thompson, R. F. &
Weisz, D. J. Hippocampus and trace conditioning of the
rabbits classically conditioned nictitating membrane
response. Behav. Neurosci. 100, 729744 (1986).
The first demonstration that lesions of the
hippocampus impair trace eyeblink conditioning.
49. Pugh, C. R., Tremblay, D., Fleshner, M. & Rudy, J. W.
A selective role for corticosterone in contextual-fear
conditioning. Behav. Neurosci. 111, 503511 (1997).
50. Rudy, J. W., Kuwagama, K. & Pugh, C. R. Isolation reduces
contextual but not auditory-cue fear conditioning: a role for
endogenous opioids. Behav. Neurosci. 113, 316323 (1999).
51. Hebb, D. O. The Organization of Behavior: a
Neuropsychological Theory (Wiley, New York, 1949).
A seminal and classic conceptualization of brain
activities and memory mechanisms.
52. Foy, M. R., Stanton, M. E., Levine, S. & Thompson, R. F.
Behavioral stress impairs long-term potentiation in rodent
hippocampus. Behav. Neural Biol. 48, 138149 (1987).
This initial study showed that stress can impair LTP in
the hippocampus, providing a possible neural basis
for stress effects on hippocampus-dependent
memory.
53. Shors T. J., Gallegos, R. A. & Breindl, A. Transient and
persistent consequences of acute stress on long-term
potentiation (LTP), synaptic efficacy, theta rhythms and
bursts in area CA1. Synapse 26, 209217 (1997).
54. Garcia, R., Musleh, W., Tocco, G., Thompson, R. F. &
Baudry, M. Time-dependent blockade of STD and LTP in
hippocampal slices following acute stress in mice. Neurosci.
Lett. 233, 4144 (1997).
55. Shors, T. J. & Dryver, E. Effect of stress and long-term
potentiation (LTP) on subsequent LTP and the theta burst
response in the dentate gyrus. Brain Res. 666, 232238
(1994).
56. Maier, S. F. & Seligman, M. E. P. Learned helplessness
theory and evidence. J. Exp. Psychol. 105, 346 (1976).
57. Shors, T. J., Seib, T. B., Levine, S. & Thompson, R. F.
Inescapable versus escapable shock modulates long-term
potentiation in the rat hippocampus. Science 244, 224226
(1989).
58. Xu, L., Anwyl, R. & Rowan, M. J. Behavioural stress
facilitates the induction of long-term depression in the
hippocampus. Nature 387, 497500 (1997).
59. Diamond, D. M., Fleshner, M. & Rose, G. M. Psychological
stress repeatedly blocks hippocampal primed burst
potentiation in behaving rats. Behav. Brain Res. 62, 19
(1994).
60. Diamond, D. M., Bennett, M. C., Stevens, K. E., Wilson, R. L.
& Rose, G. M. Exposure to a novel environment interferes
with the induction of hippocampal primed burst potentiation
in the behaving rat. Psychobiology 18, 273281 (1990).
61. Mesches, M. H., Fleshner, M., Heman, K. L., Rose, G. M. &
Diamond, D. M. Exposing rats to a predator blocks primed
burst potentiation in the hippocampus in vitro. J. Neurosci.
19, RC18 (1999).
62. Diamond, D. M., Park, C. R., Heman, K. L. & Rose, G. M.
Exposing rats to a predator impairs spatial working memory
in the radial arm water maze. Hippocampus 9, 542552
(1999).
63. Kim, J. J., Foy, M. R. & Thompson, R. F. Behavioral stress
modifies hippocampal plasticity through N-methyl-
D-aspartate receptor activation. Proc. Natl Acad. Sci. USA
93, 47504753 (1996).
64. Pavlides, C., Watanabe, Y., Margarinos, A. M. & McEwen,
B. S. Opposing roles of type I and type II adrenal steroid
receptors in hippocampal long-term potentiation.
Neuroscience 68, 387394 (1995).
65. Gould, E., McEwen, B. S., Tanapat, P., Galea, L. A. &
Fuchs, E. Neurogenesis in the dentate gyrus of the adult tree
shrew is regulated by psychosocial stress and NMDA
receptor activation. J. Neurosci. 17, 24922498 (1997).
66. Sousa, N., Lukoyanov, N. V., Madeira, M. D., Almeida, O. F.
& Paula-Barbosa, M. M. Reorganization of the morphology
of hippocampal neurites and synapses after stress-induced
damage correlates with behavioral improvement.
Neuroscience 97, 253266 (2000).
67. Brown, E. S., Rush, A. J. & McEwen, B. S. Hippocampal
remodeling and damage by corticosteroids: implications for
mood disorders. Neuropsychopharmacology 21, 474484
(1999).
68. Bliss, T. V. P. & Collingridge, G. L. A synaptic model of
memory: long-term potentiation in the hippocampus. Nature
361, 3139 (1993).
69. Martin, S. J., Grimwood, P. D. & Morris, R. G. M. Synaptic
plasticity and memory: an evaluation of the hypothesis.
Annu. Rev. Neurosci. 23, 649711 (2000).
70. Gould, E. & Gross, C. G. Neurogenesis in adult mammals:
some progress and problems. J. Neurosci. 22, 619623
(2002).
71. Nottebohm, F. Why are some neurons replaced in adult
brain? J. Neurosci. 22, 624628 (2002).
72. Gould, E., Beylin, A., Tanapat, P., Reeves, A. & Shors, T. J.
Learning enhances adult neurogenesis in the hippocampal
formation. Nature Neurosci. 2, 260265 (1999).
The first demonstration that rats undergoing
hippocampus-dependent learning (that is, trace
eyeblink conditioning) show augmented neurogenesis
in the hippocampus.
73. Rakic, P. Adult neurogenesis in mammals: an identity crisis.
J. Neurosci. 22, 614618 (2002).
74. Diamond, D. M., Bennett, M. C., Fleshner, M. & Rose, G. M.
Inverted-U relationship between the level of peripheral
corticosterone and the magnitude of hippocampal primed
burst potentiation. Hippocampus 2, 421430 (1992).
75. Bennett, M. C., Diamond, D. M., Fleshner, M. & Rose, G. M.
Serum corticosterone level predicts the magnitude of
hippocampal primed burst potentiation and depression in
urethane-anesthetized rats. Psychobiology 19, 301307
(1991).
76. Xu, L., Holscher, C., Anwyl, R. & Rowan, M. J. Glucocorticoid
receptor and protein/RNA synthesis-dependent
mechanisms underlie the control of synaptic plasticity by
stress. Proc. Natl Acad. Sci. USA 95, 32043208 (1998).
77. Kerr, D. S., Campbell, L. W., Hao, S. Y. & Landfield, P. W.
Corticosteroid modulation of hippocampal potentials:
increased effects with aging. Science 245, 15051509
(1989).
78. Joels, M. Corticosteroid actions in the hippocampus.
J. Neuroendocrinol. 13, 657669 (2001).
79. Nair, S. M. et al. Corticosteroid regulation of ion channel
conductances and mRNA levels in individual hippocampal
CA1 neurons. J. Neurosci. 18, 26852696 (1998).
80. Shors, T. J., Levine, S. & Thompson, R. F. Effect of
adrenalectomy and demedullation on the stress-induced
impairment of long-term potentiation. Neuroendocrinology
51, 7075 (1990).
81. Foy, M. R., Fory, J. G., Levine, S. & Thompson, R. F.
Manipulation of pituitaryadrenal activity affects neural
plasticity in rodent hippocampus. Psychol. Sci. 3, 201204
(1990).
82. Kim, J. J., Lee, H. J., Han, J.-S. & Packard, M. G. Amygdala
is critical for stress-induced modulation of hippocampal
long-term potentiation (LTP) and learning. J. Neurosci. 21,
52225228 (2001).
83. Diamond, D. M., Macintosh, D., Fleshner, M. & Woodson,
J. C. A fear-inducing stimulus (predator exposure) and a
sexual stimulus both increase corticosterone levels, but only
fear impairs spatial memory in male rats. Soc. Neurosci.
Abstr. (in the press).
84. Maier, S. F. Role of fear in mediating shuttle escape learning
deficit produced by inescapable shock. J. Exp. Psychol. 16,
137149 (1990).
2002 Nature PublishingGroup
462 | JUNE 2002 | VOLUME 3 www.nature.com/reviews/neuro
R E V I E WS
85. Shors, T. J., Levine, S. & Thompson, R. F. Opioid antagonist
eliminates the stress-induced impairment of long-term
potentiation (LTP). Brain Res. 506, 316318 (1990).
86. Joseph, M. H. & Kenneth, G. A. Stress-induced release of
5-HT in the hippocampus and its dependence on increased
tryptophan availability: an in vivo electrochemical study.
Brain Res. 270, 251257 (1983).
87. Corradetti, R., Ballerini, L., Pugliese, A. M. & Pepeu, G.
Serotonin blocks the long-term potentiation induced by
primed burst stimulation in the CA1 region of rat
hippocampal slices. Neuroscience 46, 511518 (1992).
88. Staubli, U. & Xu, F. B. Effects of 5-HT
3
receptor antagonism
on hippocampal theta rhythm, memory, and LTP induction in
the freely moving rat. J. Neurosci. 15, 24452452 (1995).
89. Shors, T. J. & Servatius, R. J. Stress-induced sensitization
and facilitated learning require NMDA receptor activation.
Neuroreport 6, 677680 (1995).
90. Coussens, C. M., Kerr, D. S. & Abraham, W. C.
Glucocorticoid receptor activation lowers the threshold for
NMDA-receptor-dependent homosynaptic long-term
depression in the hippocampus through activation of
voltage-dependent calcium channels. J. Neurophysiol. 78,
19 (1997).
91. Henke, P. G. Limbic system modulation of stress ulcer
development. Ann. NY Acad. Sci. 597, 201206 (1990).
92. Helmstetter, F. J. Stress-induced hypoalgesia and defensive
freezing are attenuated by application of diazepam to the
amygdala. Pharmacol. Biochem. Behav. 44, 433438
(1993).
93. Adamec, R. E., Burton, P., Shallow, T. & Budgell, J.
Unilateral block of NMDA receptors in the amygdala
prevents predator stress-induced lasting increases in
anxiety-like behavior and unconditioned startle effective
hemisphere depends on the behavior. Physiol. Behav. 65,
739751 (1999).
94. Roozendaal, B., Portillo-Marquez, G. & McGaugh, J. L.
Basolateral amygdala lesions block glucocorticoid-induced
modulation of memory for spatial learning. Behav. Neurosci.
110, 10741083 (1996).
95. McGaugh, J. L. Memory a century of consolidation.
Science 287, 248251 (2000).
A review of the role of the amygdala and its
neurochemical systems in modulating memory
consolidation in extra-amygdalar structures such as
the hippocampus.
96. Abe, K. Modulation of hippocampal long-term potentiation
by the amygdala: a synaptic mechanism linking emotion and
memory. Jpn. J. Pharmacol. 86, 1822 (2001).
A review of amygdalar modulation of LTP in the
dentate gyrus of the hippocampus.
97. Akirav, R.-L. & Richter-Levin, G. Biphasic modulation of
hippocampal plasticity by behavioral stress and basolateral
amygdalar stimulation in the rat. J. Neurosci. 19,
1053010535 (1999).
98. Goldstein, L. E., Rasmusson, A. M., Bunney, S. B. & Roth,
R. H. Role of the amygdala in the coordination of behavioral,
neuroendocrine and prefrontal cortical monoamine
responses to psychological stress in the rat. J. Neurosci. 16,
47874798 (1996).
99. Arnsten, A. F. T. Stress impairs prefrontal cortical function in
rats and monkeys: role of dopamine D1 and norepinephrine
1 receptor mechanisms. Prog. Brain Res. 126, 183192
(2000).
100. Pikkarainen, M., Ronkko, S., Savander, V., Insausti, R. &
Pitkanen, A. Projections from the lateral, basal, and
accessory basal nuclei of the amygdala to the hippocampal
formation in rat. J. Comp. Neurol. 403, 229260 (1999).
101. Sherman, S. M. & Guillery, R. W. On the actions that one
nerve cell can have on another: distinguishing drives from
modulation. Proc. Natl Acad. Sci. USA 95, 71217126
(1998).
102. Kopin, I. J. Definitions of stress and sympathetic neuronal
responses. Ann. NY Acad. Sci. 771, 1930 (1995).
103. Levine, S. in Stress-Neurobiology and Neuroendocrinology
(eds Brown, M., Koob, G. & Rivier, C.) 321 (Marcel Dekker,
New York, 1991).
104. Wolf, O. T., Schommer, N. C., Hellhammer, D. H., McEwen,
B. S. & Kirschbaum, C. The relationship between stress
induced cortisol levels and memory differs between men
and women. Psychoneuroendocrinology 26, 711720
(2001).
105. Phoenix, C. H., Dixson, A. F. & Resko, J. A. Effects of
ejaculation on levels of testosterone, cortisol, and luteinizing
hormone in peripheral plasma of rhesus monkeys. J. Comp.
Physiol. Psychol. 91, 120127 (1977).
106. Bronson, F. H. & Desjardins, C. Endocrine responses to
sexual arousal in male mice. Endocrinology 111, 12861291
(1982).
107. Kanaley, J. A., Weltman, J. Y., Pieper, K. S., Weltman, A. &
Hartman, M. L. Cortisol and growth hormone responses to
exercise at different times of day. J. Clin. Endocrinol. Metab.
86, 28812889 (2001).
108. Rosmond, R., Holm, G. & Bjorntorp, P. Food-induced
cortisol secretion in relation to anthropometric, metabolic
and haemodynamic variables in men. Int. J. Obes. Relat.
Metab. Disord. 24, 416422 (2000).
109. Maier, S. F. & Watkins, L. R. Stressor controllability, anxiety,
and serotonin. Cognit. Ther. Res. 22, 595613 (1998).
110. Prince, C. R. & Anisman, H. Situation specific effects of
stressor controllability on plasma corticosterone changes in
mice. Pharmacol. Biochem. Behav. 37, 613621 (1990).
111. Maier, S. F., Ryan, S. M., Barksdale, C. M. & Kalin, N. H.
Stressor controllability and the pituitaryadrenal system.
Behav. Neurosci. 100, 669674 (1986).
112. Arnsten, A. F. Catecholamine regulation of the prefrontal
cortex. J. Psychopharmacol. 11, 151162 (1997).
113. Fox, M. L. & Dwyer, D. J. Reducing stress by increasing
control. Clin. Leadersh. Manag. Rev. 14, 114117
(2000).
114. Milner, B., Corkin, S. & Teuber, H.-L. Further analysis of the
hippocampal amnesic syndrome: fourteen year follow-up
study of H.M. Neuropsychologia 6, 215234 (1968).
This study originated the concept of separable types
of memory with the discovery of some preserved
learning-and-memory capabilities in the profoundly
amnesic patient H.M.
115. Tulving, E. in Organization of Memory (eds Tulving, E. &
Donaldson, W.) 381403 (Academic, New York, 1972).
116. Graf, P. & Schacter, D. L. Implicit and explicit memory for
new associations in normal subjects and amnesic patients.
J. Exp. Psychol. 11, 501518 (1985).
117. Mishkin, M. Memory in monkeys severely impaired by
combined but not separate removal of the amygdala and
hippocampus. Nature 273, 297298 (1978).
The first demonstration of anterograde amnesia in
monkeys after the removal of medial temporal lobe
structures.
118. Baxter, M. G. & Murray, E. A. Opposite relationship of
hippocampal and rhinal cortex damage to delayed
nonmatching-to-sample deficits in monkeys.
Hippocampus 11, 6171 (2001).
119. Zola, S. M. & Squire, L. R. Relationship between magnitude
of damage to the hippocampus and impaired recognition
memory in monkeys. Hippocampus 11, 9298 (2001).
120. OKeefe, J. & Dostrovsky, J. The hippocampus as a spatial
map. Preliminary evidence from unit activity in the freely-
moving rat. Brain Res. 34, 171175 (1971).
The first discovery of neurons in the hippocampus
that show increased firing rates when a rat is in a
specific location in an environment (place cells).
121. Silva, A. J., Paylor, R. Wehner, J. M. & Tonegawa, S.
Impaired spatial learning in -calciumcalmodulin kinase II
mutant mice. Science 257, 206211 (1992).
122. Tsien, J. Z., Huerta, P. T. & Tonegawa, S. The essential role
of hippocampal CA1 NMDA receptor-dependent synaptic
plasticity in spatial memory. Cell 87, 13271338 (1996).
123. Silva, A. J., Smith, A. M. & Giese, K. P. Gene targeting and
the biology of learning and memory. Annu. Rev. Genet. 31,
527546 (1997).
124. Barnes, C. A. & McNaughton, B. L. An age comparison of
the rates of acquisition and forgetting of spatial information
in relation to long-term enhancement of hippocampal
synapses. Behav. Neurosci. 99, 10401048 (1985).
125. Becker, J. T., Olton, D. S., Anderson, C. A. & Breitinger,
E. R. P. Cognitive mapping in rats: the role of the
hippocampal and frontal systems in retention and reversal.
Behav. Brain Res. 3, 122 (1981).
126. Kim, J. J. & Fanselow, M. S. Modality-specific retrograde
amnesia of fear. Science 256, 675677 (1992).
127. McEchon, M. D., Bouwmeester, H., Tseng, W., Weiss, C. &
Disterhoft, J. F. Hippocampectomy disrupts auditory trace
fear conditioning and contextual fear conditioning in rats.
Hippocampus 8, 638646 (1998).
128. Huerta, P. T., Sun, L. D., Wilson, M. A. & Tonegawa, S.
Formation of temporal memory requires NMDA receptors
within CA1 pyramidal neurons. Neuron 25, 473480 (2000).
129. Eichenbaum, H., Dudchenko, P., Wood, E., Shapiro, M. &
Tanila, H. The hippocampus, memory, and place cells: is it
spatial memory or a memory space? Neuron 23, 209226
(1999).
130. OKeefe, J., Burgess, N., Donnett, J. G., Jeffery, K. J. &
Maguire, E. A. Place cells, navigational accuracy, and the
human hippocampus. Phil. Trans. R. Soc. Lond. B 353,
13331340 (1998).
131. Bliss, T. V. P. & Lomo, T. Long-lasting potentiation of synaptic
transmission in the dentate area of the anaesthetized rabbit
following stimulation of the perforant path. J. Physiol. (Lond.)
232, 331356 (1973).
The first demonstration of activity-dependent, long-
lasting enhancement of synaptic efficacy. This study
provided a synaptic model of learning and memory.
132. Murphy, G. G. & Glanzman, D. L. Mediation of classical
conditioning in Aplysia californica by long-term potentiation
of sensorimotor synapses. Science 278, 467471 (1997).
133. Sejnowski T. J. Statistical constraints on synaptic plasticity.
J. Theor. Biol. 69, 385938 (1977).
134. Dudek, S. M. & Bear, M. F. Homosynaptic long-term
depression in area CA1 of hippocampus and the effects of
NMDA receptor blockade. Proc. Natl Acad. Sci. USA 89,
43634367 (1992).
135. Thompson, R. F. The neurobiology of learning and memory.
Science 233, 941947 (1986).
136. Wilson, M. A. & Tonegawa, S. Synaptic plasticity, place cells
and spatial memory: study with second generation knockout.
Trends Neurosci. 20, 102106 (1997).
137. Shors, T. J. & Matzel, L. D. Long-term potentiation: whats
learning got to do with it? Behav. Brain Sci. 20, 597614
(1997).
Online links
DATABASES
The following terms in this article are linked online to:
LocusLink: http://www.ncbi.nlm.nih.gov/LocusLink/
glucocorticoid receptor | mineralocorticoid receptor
OMIM: http://www.ncbi.nlm.nih.gov/Omim/
Cushings disease
FURTHER INFORMATION
Encyclopedia of Life Sciences: http://www.els.net/
hippocampus | learning and memory | long-term depression and
depotentiation | long-term potentiation
MIT Encyclopedia of Cognitive Sciences:
http://cognet.mit.edu/MITECS/
hippocampus | long-term potentiation | memory | memory, animal
studies | memory, human neuropsychology | neuroendocrinology |
stress
Access to this interactive links box is free online.