BrainResearchButletin,Vol. 35, Nos. 516,pp.

457-465, 1994

Copyright0 1994ElsevierScienceLtd
Printedin the USA. Allrights reserved

Pergamon

0361-9230/94$6.00
+ .OO

036%9230(94)00135-9

Locus Coeruleus-evoked Responses

in Behaving Rats: A Clue to the Role of
Noradr~nalin~ in Memory
SUSAN J. SARA,
Institut des Neurosciences,

ANDREY

VANKOV

AND ANNE HERVfZ

Universitb P & M Curie, 9 quai St. Bernard,

7500.5 Paris, Frunce

SARA, S. J., A. VANKOV AND A. HERVfi. Locus coeruleus-evoked responses in behaving rats: A clue IO the role of noradrenuline in memory. BRAIN RES BULL 35(5/Q 457-465, 1994.-Neuromodulatory
properties of noradrenaline (NA) suggest
that the coreruieo-cortical NA projection should play an important role in attention and memory processes. Our research is aimed
at providing some behavioral evidence. Single units of the locus coeruleus (LC) are recorded during controlled behavioral situations, in order to relate LC activation to specific behavioral contexts. LC cells respond in burst to imposed novel sensory stimuli
or to novel objects encountered during free exploration. When there is no predictive value of the stimulus or no behavioral response
required, there is rapid habituation of the LC response. When a stimulus is then associated with reinforcement, there is a renewed
response, which is transient. During extinction, LC neuronal responses reappear. Thus, LC cells respond to novelty or change in
incoming imformation, but do not have a sustained response to stimuli, even when they have a high level of biological significance.
The gating and tuning action of NA released in target sensory systems would promote selective attention to relevant stimuli at the
critical moment of change. The adaptive behavioral outcome would result from the integration of retrieved memory with the
sensory info~ation selected from the envi~nment.

Memory

Novelty response

Locus coeruleus

Habituation

Single-unit activity

Behaving rat

Hole board

Noradrenaline

discrete signals and tune receptive fields of individual cells in

thalamus (16), somatosenso~ cortex (28,29), visual cortex (12),
and auditory cortex (5). b) Moreover, the presence of NA can
change the spontaneous mode of activity of thalamic neurons
from burst to spike (13). The spike mode is necessary for the
accurate transfer of incoming information to the cerebral cortex
and is always associated with cortical arousal (26,27). c) Finally,
NA promotes and may even permit long-term potentiation in the
hip~ampus,
a process believed by many to underly memory
(14,24,25).
All of these authors concur that the actions of NA on cells
and synapses in various target structures should be of fundamental importance for cognitive processes such as attention, perception, leaming, and memory [(6,8) for reviews]. Our view is that
the noradrenergic projection from the locus coeruleus plays an
important role in memory processes by promoting or permitting
changes in network configurations at critical times when environmental events may require changes in behavioral output.
Although the effects of neuromodulatory influences on behavioral output have been extensively studied in invertebrate systems [(9) for review], the functional significance of neuromodulatory effects on target cells and networks in higher species has
mostly been a subject of intense speculation, with many authors
promoting the idea that such systems should play a fundamental
role in the regulation of attention and information processing
(6,8,13,23,28,29). Such a proposition, based on the electrophys-

MEMORY is a biological adaptation that permits an organism to

use past experience to adjust its behavior to changing environmental imperatives and it is assumed to be a result of leaminginduced, long-lasting changes in functional connectivity among
neurons. The significance of the memory for adaptive behavior
lies in its retrieval, which depends, not upon static connections
among neurons, but upon the outputs of dynamic networks.
These outputs, in turn, are not fixed, but are very much dependent
upon the state of the system. In this way, a single network can
perform several functions (7,8) and underly more than one memory. The state of the system is mediated by the action of neuromodulatory influences, originating, for the most part, in brain
stem and basal forebrain nuclei, principally, locus coeruleus (LC)
and nucleus basalis. Thus biological theories of memory must
take into account these brain stem influences on forebrain activity, reactivity, and plasticity. Nemom~ulators
influence fimctional connectivity within a network by modifying both the intrinsic properties of the cellular components and their synaptic
interactions [(2) for general review]. As early as 1970, it was
proposed that noradrenaline (NA) should play an essential role
in learning and memory by permitting or even accentuating
activity in (synapses) that are ~~s~~ing
novel or significant
stimuli (11). Indeed, many subsequent studies have shown that
NA has several postsynaptic actions that would lend support to
Ketys intuitive hypothesis. a) Although inhibiting spontaneous
activity, NA can enhance, or even permit, cellular responses to
To whom requests for reprints should be addressed.
457

458

SARA,

_J

idaroxan
T

AND HERVfi

necessary to efficiently perform the task (Fig. 1). These experiments suggest that stimulation of noradrenergic function facilitates shifts in attention and behavioral adaptation to changes in
response-reinforcement
contingencies.

*
T

7-

Are LC Cells Differentially

of the Learning Process?

b
i6;
;

VANKOV

Engaged During Different Phases

If noradrenergic neuromodulatory influences are necessary for

cognitive and behavioral adaptation to changes in environmental
contingencies,
then the LC neurons should be particularly responsive to such changes. To evaluate this, we studied the firing
pattern of single units in the LC in rats during the whole course
of learning-from
response to the novel discriminanda through

5-

1
1

I
4

Trials

CS + (LF TONE)

FIG. I. Spatial memory deficit after a change in extramaze cues: facilitation by a single injection of idazoxan. Rats were trained for 10 days in
the eight-arm radial maze until they reached asymptotic performance.
The maze was moved to another room that induced a marked decrement
in performance (trial 1). Rats treated with idazoxan before the first trial
showed the same decremant as control rats, but on the next trial, 24 h
later, the performance level was equal to that of the last training trial,
while control rats continued to show the decrement. Squares: saline;
circles: idazoxan (2 mg/kg, IP). Arrow indicates the single injection of
idazoxan, 30 min before the first trial after the room change. *p < 0.01.

iological evidence cited above, although intuitively attractive,

must be tested in behaving animals, where attempts to evaluate
this hypothesis have not provided unequivocal support. Rats with
noradrenergic
depletion,
although exhibiting
some behavioral
deficits, are perfectly well able to learn and remember even complex
tasks [( 15) for review]. Moreover, pharmacological or electrical
stimulation of LC has little effect on straightforward learning and
retention [(17,18) for reviews]. Nevertheless, we have defined
specific behavioral situations in which such stimulation of the
noradrenergic
system improves cognitive performance.
Using
electrical stimulation of the LC nucleus or the alpha, antagonist
idazoxan to activate cells in the LC, we found a marked facilitation of memory retrieval in a spontaneous forgetting paradigm,
with no effect on acquisition or performance of the task (4,19).
An important
factor in retrieval process is directing attention and
selecting relevant stimuli from the available array, particularly
after an extended retention interval, when forgetting has occurred
(21).
To further examine the effect of noradrenergic stimulation on
such processes we designed experiments that explicitly tested the
ability of the rat to shift attention from one array of stimuli to
another. Rats were trained in the same linear maze as was used
for the forgetting experiments and after acquistion, the responsereinforcement contingencies were changed within the same apparatus. Now, instead of the fixed path, the rat was required to
follow cues, the position of which was changed on each trial.
Rats were treated with idazoxan
at the beginning
of each daily
trial. There was no effect of the drug on the acquisition of either
aspect of the task alone, but a marked facilitation of acquisition
of the new response pattern (4).
In a similar vein, we trained rats in a radial arm maze where
the animal has to use visual cues outside the maze to navigate
around and make a spatial map of the situation. If the external
cues are changed after the rat is well trained, there is a marked
decrement in performance, which endures for several subsequent
trials. Rats treated with idazoxan on the day of the change recover
their performance
much more quickly,
presumably
are better able to shift their attention
to the newly

because
relevant

they
cues

CS - (HF TONE)

acquisition

IL_

L
over

training

CS + (HF TONE)

CS - (LF TONE)

reversal
tone

dipper

tone

FIG. 2. Activity of a single unit in the locus coeruleus during differential

appetitive conditioning.
Top: acquisition trials 40-60;
left: CS+ (rewarded) trials: a tone of 4 kHz (LF), 500 ms duration, was followed by
a 200 ms delay and then by the presentation of a dipper containing sweet
water. Right: CS- (nonrewarded) trials: a 8 kHz (HF) tone was presented
and nonreinforced.
Note the larger response of the LC unit to the CS+
than to the CS-. Middle: overtraining
trials IO& 120, with stimulusreinforcement
contingencies
above. There are no longer any LC responses to the CSs during this overtrained phase, even though the be
havioral response to the dipper is unchanged and the rat continues to
consume the reinforcement.
Bottom: reversal trials: the stimulus-reinforcement contingencies are reversed so that now the HF tone is followed
by the dipper presentation and the LF tone is not reinforced. There is an
immediate reengagement
of the LC unit, as reflected by increased responding to both tones, with no significant increase in baseline responding. Bin size: 100 ms, tone duration 500 ms [adapted from (21)].

LOCUS

COERULEUS

ACTIVITY

459

AND MEMORY

BEHAVIOR

CELL
cs+

cs+

CS-

cs:j
::
::
u,
::
:
:::
:i

::
::
::
::
:
:::
:i

Ilk_
:i::::
:j:/
AL
1 :
:
j
:
j
:
:
:
:
:
:
:
:
:
:

lipper

m$

:
j
:
j
:
:
:
:
:
:
:
:
:
i

dipper

FIG. 3. Left: activity of a single unit in the locus coeruleus during differential appetitive conditioning. Right: development of
the behavioral response to the CSf and CS-. The behavioral measure was a head movement towards the water dipper. Note
the absence of LC response at the early trials (the training sessions were preceded by ten habituation trials), followed by a
differential neuronal response appearing in the second block (trials 21-40). The neuronal differential response precedes the
behavioral differential response by many trials; behavioral response to the CS+ appears in the third block (trials 41-60) and by
the end of the session, the behavior is controlled by the appearance of the dipper (trials 61-80). The neuronal response in the
LC is greatly reduced when the differential behavioral response to CS+ trials is well established. Bin size: 100 ms; tone duration
500 ms [adapted from (21)].

learning,
reversal
learning, and extinction. Singleunit activity in the LC was recorded in freely moving rats in a
discrimination
learning paradigm, where tones are associated
with foot shock or, in other experiments, with sweet water. The
aversive situation was a stimulus-stimulus
association paradigm,
while the appetitive reinforcement
paradigm required a head
movement toward the dipper to consume the reinforcement after
the CS+. LC cells responded when the tone was novel; the response habituated very rapidly when the tone was not followed
by shock; as soon as shock was introduced, the cells responded
again vigorously to the tone, for the early conditioning trials, but
dropped out during overtraining. During the extinction procedure, when the rat was expecting the shock but it did not appear,
there was a vigorous response of the LC cells, which usually
lasted for 10 or 20 trials (Fig. 2). In the appetitive version of the
task, where one tone was followed by sweet water and a second
tone was never reinforced, the LC cells responded differentially
to the positive stimulus well before the rat expressed discrimination between the two tones at a behavioral level by poking its
head in the hole to drink only after the CS+. The LC response
disappeared when differential behavioral responses were established. When the significance of the stimuli was reversed, the LC
cells immediately began to fire to both stimuli (Fig. 2). All of
these fluctuations in phasic responses were directly related to
changes in the significance of the stimulus but were independent
of changes in baseline firing of the cell, as illustrated in Fig. 2
(22). In contrast to the present results, Aston-Jones
and Bloom
(1) reported that there was no habituation of the response of LC
cells to an auditory tone pip.
differential

These experiments show that LC cells are engaged during

learning whenever there is a change in significance of the stimulus. When a modification of the behavioral response is required,
as in the case of the appetitive paradigm, the LC neuronal response to the CSs disappears when the appropriate differential
operant behavior is established, i.e., head movement to the CS+
and not to the CS-. The LC differential response to the CS+
preceded by many trials the differential head movement response
to consume the reinforcement (Fig. 3).
These results are significant in that they are the first report of
LC cells being involved in learning situations that do not use
stressful reinforcement. In the only other study of the activity of
LC cells during learning, Jacobs found that there was a conditioned response to CSs that predicted shock, but no response to
CSs that announced delivery of food, leading to the conclusion
that the LC is involved more in stress than in information processing per se. These experiments were carried out in the cat,
and differences from our results may be due to species differences, as Jacobs, himself, suggests (10).
Are LC Cells involved in Reponse to Novelty and Change
Outside of Formal Learning Situations?
To address this question we used a hole board apparatus in
which novel objects could be placed. It consists of a 60 x 60 x
40 cm box with nine holes of 4 cm diameter, symmetrically
placed in the floor. The apparatus was equipped with stategically
placed photoelectric cells that monitored exactly where, when,
and for how long the rat puts its nose in a hole. Locomotor ac-

460

SARA,

E
.-

34

.5
5
x
ag

24

VANKOV

AND HERVfi

14

SAL

SAL

IDA

IDA

no objects

objects

A
2.5 -

FIG. 4. Hole
Photoelectric
quence, and
some of the
ments, only

board apparatus used in the object exploration experiments.

cells monitor locomoter activity and the frequency, seduration of each visit to a hole. Objects can be placed in
holes, as illustrated. During the LC unit recording experitwo holes were exposed.

tivity and rearing was monitored at the same time, and the information was stored on a computer for offline analysis. The
apparatus is described in detail in reference 3. In most experiments, the rats are familiarized with the apparatus, and baseline
data is collected before pharmacological
treatment or recording
and before objects are placed in the holes.
Rats spend much more time examining the holes containing
objects than the empty holes, especially if they have been in the
hole board before and have never seen the objects. Rats treated
with idazoxan (2 mg/kg) before the session spend even more time
looking at the novel objects than control rats (Fig. 5A) (3,18).
This is a dose that markedly increases the firing rate of LC cells
(18). On the other hand, when treated with very low doses of
clonidine-doses
low enough to significantly decrease spontaneous firing of LC cells but to have no detectable sedative effect-the
preference for the holes with objects is strongly
attenuated (Fig. 5B) (20). There is no effect of either drug treatment on locomotor activity, rearing, or hole visits in general; it
is restricted to the expression of preference for the holes containing the novel objects.
We recorded single-unit activity of LC cells in rats in this
behavioral situation. Rats were anethesitized with pentobarbital
(60 mg/kg) and were implanted with movable microelectrodes in

FIG. 6. LC unit activity during exploration

10

t&kg

6
FIG. 5. (A) Effect of idazoxan on locomotor activity and on exploration
of holes with and without objects. Animals were submitted to a preexperimental session 2 days before the drug session to obtain a baseline of
actvity. The results are expressed in terms of the differences between the
two sessions, total duration of visits to holes with objects (left) and empty
holes (right). Control rats show a preference for the holes containing the
novel object; rats treated with 2 mg/kg idazoxan. a dose that increases
firing of LC neurons and enhances release of NA, spend an even greater
amount of time in contact with the holes containing objects (from Sara,
1991). (B) Clonidine, at the threshold dose for inhibiting firing of LC
units, blocks the preference for novelty, while having no effect on general
exploratory activity. In this experiment, there was no predrug habituation
session and the results are expressed in terms of the mean duration of
visits to holes containing objects (striped bars) and empty holes (white
bars).

the LC under stereotaxic and electrophysiological

control. Single-unit activity was recorded in the awake, freely moving animal. The recording techniques used were the same as for our
previous conditioning experiments described above. (A full de-

in the hole board. Session one, with both holes empty. (A) Chronology of hole visits during the IO-min
session. (B) Behavior and unit activity for 10 s around each hole visit. Behavior (hole 1,hole 2); each deflection represents a period during which the
rat interupts the photocell beam under the hole, i.e., has its nosed inside the hole. The dots represent the mean instantaneous
frequency of firing of
the unit (calculated as one/interspike interval). Thus, each dot represents one spike. Session two, with a novel salient object in hole 1. (A) Chronology
of hole visits during the IO-min session. (B) Behavior and unit activity for 10 s around each hole visit. Note that the first visit to hole one consists of
four very short nose pokes. The LC unit fires in burst of action potentiels that are time-locked with the encounter with the novel object in the hole.
On the visit to hole two, which is empty, the rat emits a similar behavior pattern, but there is no LC response. On the two subsequent visits to the
hole containing the object, the burst response is much smaller, indicating habituation.

LOCUS COERULEUS ACTIVITY

AND MEMORY

461

LC UNIT RESPONSE TO A NOVEL OBJECT

SESSION 1
BOTH HOLES EMPTY

hole 1
head DIPS
II~iI
Il.
la& zm.

hole 1

hole2

hole 2
w
1IIl-llIf
3i.
48fl. an.
688.

i,

I
,

r---r;

A-

lK!

SESSION 2
OBJECT IN HOLE ONE

8.

Me.

I_
288.

hole2
sax
3m.

4m.

!im.

SARA,

462

Trials 1-5

VANKOV

AND HERVti

Trials 6- 10

j,,.

,,(,,,!,,,,H_

0.9

13

8000 Hz

8000 Hz

230 Hz

230 Hz

FIG. 7. Response of a single LC unit to tones of four different frequencies, in an awake, freely moving rat after the hole board exercise. A series of
10 stimuii were presented for each frequency; interstimulus interval was 30 s, tone duration was I s. Left: sum of responses, block of first five stimuli.
Right: sum of responses, block of last five stimuli. Bin size: IO ms from top to bottom: responses to tones of 8 kHz, 230 Hz, 12 Hz, 4 kHz, respectively.
The response is limited to a burst at the onset of the tone. Note the decrease of responses on the second block of trials (right side).

scription of the recording method can be found in reference 22).

After a cell was isolated and determined by its firing characteristics and wave form to be a putative LC cell, activity was recorded while the rats explored the hole board, usually for IO-mitt

sessions. Output from the window discriminator was fed to one

channel of an interface (Cambridge Electronics Device). The behavioral output from the holeboard photoelectric detectors, ineluding the precise time and duration of each nose poke, went to

LOCUS

COERULEUS

ACTIVITY

463

AND MEMORY

Trials l-5

Triak

6-10

4000 Hz

12000 Hz

12000 Hz

FIG.8.Response of a single unit of the LC to tones of three different frequencies. Protocol the same as for Fig. 4. Left: sum of responses, block of
first five stimuli. Right: sum of responses, block of last five stimuli. Bin size: 10 ms. Top: 8 kHz tone; middle: 4 kHz tone; bottom: 12 k.Hz tone; no
response to the 230 Hz tone. Note that the pattern of response of this cell is different for each frequency: a brief single phasic response at 8 kHz, a
biphasic response to 4 kHz, and a more diffuse response to 12 kHZ. There is rapid habituation at all frequencies.
another channel. In this way, LC activity and hole board exploration activity could be analyzed together offline (Spike2 software, CED).
In the first series of experiments,
the animals were very
familiar with the apparatus. In the first IO-min recording session,
there were no objects in either of the two exposed holes. After a
lo-min postsession rest period spent quietly in a familiar cage,
the rat was replaced in the apparatus where there was now an
object in one of the holes. Tbe LC unit responded with a burst

on the first encounter with the novel object. There was no LC

response when the rat put its nose in the hole without the object.
This burst response to the encounter with the unexpected novel
object habituated rapidly, with no response at all after the first or
second encounter (Fig. 6).
Rapid habituation of responses of LC cells to nonreinforced
novel stimuli has been previously noted in our auditory learning
experiments, described above, but it has not been reported in the
literature as a particular feature of LC response mode. Thus fur-

464

SARA,

ther study of the persistence of the LC response to novel tones

in the absence of reinforcement was made with the same animals
after the exploration session. Four different tone frequencies were
used. The tones were presented to the still, awake rat in a series
of 10 stimuli, with a 30-s intertone interval; the order of presentation of the different frequencies was different for each rat, and
this had no influence on the probability or type of response. LC
cells exhibited a robust response to the first few presentations of
the tone, although not all rats responded at all frequencies. Most
responses occurred to tones of 4 kHz and 8 kHz, with some
responses at 12 kHz; very few rats responded to tones of 230 Hz.
The intensity of the tones (about 70 dB) did not produce a
startle response and usually did not elicit a clear orienting
response either. The most frequently
observed cellular response was a burst consisting of IO- 12 spikes, followed by a
short inhibitory period, as illustrated in Fig. 7. The latency of
the response was about 40 ms.
The same cell that responded consistently to a tone of a given
frequency sometimes presented a biphasic response to a tone of
another frequency, as illustrated in the middle panel of Fig. 8,
where the initial response is a burst of seven action potentiels
with a latency of 50 ms, followed by a second response at 580
ms. In this rat, a similar type of response is observed to the 12
kHz tone. This is a response of the same cell, in the same rat
during a still, awake state. There was rapid habituation independently of the type of response exhibited; in most cases, the cell
ceased to respond after the first few presentations of the stimulus.
This is illustrated in both Figs. 7 and 8. Rapid habituation of the
auditory response to tones of similar intensity, duration, and frequency has been observed in all awake rats we have tested in this
protocole (n = 7), as well as in most rats in our previous experiments, where the habituation procedure was part of the conditioning protocol (21). We are conducting further parametric studies of auditory habituation and conditioning
of LC cells in
~e~esitized
rats, where prelimin~
results replicate our observations in the awake rat (20). One important parameter determining the rate of habituation to nonreinforced stimuli might be
the duration of the stimulus; our experiments use tones of OS1 s duration while others, who do not report habituation, use pips
of 50 ms duration (1).
CONCLUSION

AND PERSPECTIVES

The LC is engaged in cognitive situations similar to those in

which stimulation of the system is particularly effective, i.e.,
when there is novelty or a change in the significance of a stimulus, requiring a new behavioral strategy. Recordings of single
unit activity in the LC during learning, during spontaneous exploration of a new environment and unf~ili~
objects, and during presentation auditory stimuli confirm that LC neurons respond to novel stimuli, habituate rapidly, respond again when the
stimulus changes its predictive value, well before differential responding at a behavioral level. An important element of these

VANKOV

AND HERVB

observations is that the LC conditioned response is not sustained

once the behavioral response is established; the phasic response
to the CSs disappears and this is independent of any changes in
the tonic firing rate. As soon as the stimulus-reinforcement
contingencies are changed, i.e., during reversal or extinction, the
cells respond anew.
Such a transient mode of responding of LC neurons must be
taken into account when considering how the noradrenergic system might contribute to information processing, learning, and
memory retrieval. If NA were necessary for sustained attention
during high level performance, one would expect that the responses to relevant stimuli should persist throughout the performance trials to assure continuous release of NA. Results consistent with such a view have been obtained from two very different
paradigms. In monkeys, the LC has been observed to repond
consistently through hundreds of trials to odd-ball stimuli (2). In
the rat engaged in a reaction time task involving a warning signal
and a target signal, both auditory, the LC cells responded consistently over many trials to the warning signal, suggesting an
involvement in preparatory set (30). On the other hand, the rapid
LC responses to environmen~l
changes, together with the
equally rapid habituation of these responses, seen in our experiments, encourage the view that the noradrenergic system is implicated in shifting attention to environmental imperatives. The
fact that the neuronal differential response in the LC clearly precedes the differential behavioral response in our learning experiments lends support to this view. The release of NA during these
LC phasic responses would change the functional state of the
system and, consequently, the output of already established networks. Retrieval of relevent memories might occur in this way.
At the same time, the gating and tuning action of NA in target
sensory systems would promote selective attention to relevant
sensory stimuli. The adaptive behavioral outcome would result
from the integration of the retrieved memory with the sensory
info~ation
selected from the environment at the critical moment
of change.
The results of both the stimulation and recording experiments
reinforce current hypotheses concerning the role of the coeruleoforebrain noradrenergic system in selective attention and sensory
and neuronal plasticity, for the most part formulated from electrophysiological
studies of the postsynaptic actions of NA. The
challenge now is to combine these approaches in the behaving
animal to study the relationship between LC activation, postsynaptic neuromodulatory
effects such as sensory tuning in the cortex, on the one hand, and performance and retention on a relevant
behavioral task, on the other.

This research was supported by grants from the Cognisciences Programme of the CNRS and by the Minister for Research and Technology
of France (MRT#91-C0956)
to S.J.S. A.H. is a predoctoral fellow of
MRT; A.V. was supported by a postdoctoral fellowship from the Fondation pour la Recherche Medicale.

~FERENCES
Aston-Jones, G.; Bloom, F. E. Norepinephrine-containing
locus
coeruleus neurons in behaving rats exhibit pronounced
responses
to nonnoxious
environmental
stimuli. J. Neurosci.
1:887-900;
1981.
Aston-Jones,
G.; Chiang, C.; Alexinsky, T. Discharge of noradrenergic IWUS coeruleus neurons in behaving rats and monkeys suggests
a role in vigilance. Prog. Brain Res. 88:502-517;
1991.
Devauges, V.; Sara, S. J. Activation of the noradrenergie
system

facilitates an attentional shift in the rat. Behav. Brain Res. 39:1928; 1990.
4. Devauges, V.; Sara, S. J. Memory retrieval enhancement
by locus
coeruleus stimulation: Evidence for mediation by beta receptors. Behav. Brain Res. 43:93-97;
1991,
5. Edeline, J. M.; Sara. S. J. Sharpening frequency receptive fields of
auditory cortex neurons by idazoxan-induced
release of norepinephtine. Sot. Neurosci. Abstr. 748.4; 1993.

LOCUS COERULEUS ACTIVITY

AND MEMORY

465

6. Foote, S.; Morrison, J. Extrathalamic

modulation of cortical function. Annu. Rev. Neurosci. 10~67-95; 1987.
I. Getting, P. Emerging principles governing the operation of neural
networks. Annu. Rev. Neurosci. 12: 185-204;
1989.
in arousal, emotion and learn8. Harley, C. A role for norepinepherine
ing?: Limbic modulation by norepinepherine
and the Kety hypothesis. Prog. Neuropsychopharmacol.
Biol. Psychiatry
11:419-458;
1987.
9. Harris-Wanick,
R. M.; Marder, E. Modulation of neural networks
for behavior. Annu. Rev. Neurosci. 14:39-57;
1991.
10. Jacobs, B. Single unit activity of locus coeruleus neurons in behaving animals. Prog. Neurobiol. 27:183-194;
1986.
Il. Kety, S. The biogenic amines in the central nervous system: Their
possible roles in arousal, emotion and learning. In: Schmitt, F. O.,
ed. The neurosciences
second study program. New York: Rockefeller University Press; 1970:324-335.
12. Madar, Y.; Segal, M. The functional role of the noradrenergic system
in the visual cortex. Activation of the noradrenergic
pathway. Exp.
Brain Res. 41:814; 1980.
13. McCormmick,
D. A. Cholinergic and noradrenergic
modulation of
tbalamocortical
processing. Trends Neurosci. 12(6):215-221;
1989.
14. Neumann, R. S.; Harley, C. Long-lasting potentiation of the dentate
gyms population spike by norepinepherine.
Brain Res. 273: 162165; 1983.
15. Robbins, T. W.; Everit, B. J. Comparative
functions of the central
noradrenergic,
dopaminergic
and cholinergic
systems. Neuropharmacology 26:893-901;
1987.
16. Rogawski, M. A.; Aghajanian, G. K. Modulation of lateral geniculate neurone excitability by noradrenaline
microiontophoresis
or locus cocruleus stimulation. Nature 287:731-734;
1980.
17. Sara, S. J. The locus coeruleus and cognitive function: Attempts to
relate noradrenergic
enhancement of signal/noise in the brain. Physiol. Psycho]. 13(3):151-162;
1985.
18. Sara, S. J. Noradrenaline
and memory: Neuromodulatory
effects on
retrieval. In: Weinman, J.; Hunter, J., eds. Memory: Neumchemical
and clinical aspects. London:
Harwood
Academic
Publishers;
1991:105-128.
19. Sara, S. J.; Devauges, V. Priming stimulation of locus coeruleus
This

article

is being

published

without

the authors

review

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

facilitates memory retrieval in the rat. Brain Res. 438(1/2):299-303;

1988.
Sara, S. J.; Dyon-Laurent,
C.; HervC, A. Novelty seeking behavior
in the rat is dependent upon the integrity of noradrenergic
system.
(submitted).
Sara, S. J.; Hervt, A. Plasticity of auditory responses of locus coeruleus neurons of anesthetized rats. Sot. Neurosci. Abstr. 173.1 I;
1993.
Sara, S. J.; Segal, M. Plasticity of sensory reponses of locus coeruleus neurons in behaving rats: Implication for cognition. Prog. Brain
Res. 88:571-585;
1991.
Segal, M.; Bloom, F. The action of norepinephrine
in the rat hippocampus. III Hippocampal
cellular responses to locus coeruleus
stimulation in the awake rat. Brain Res. 107:499-512;
1976.
Stanton, P. K.; Sarvey, J. M. Norepinephrine
regulates
longterm potentiation of both the population spike and dendritic EPSP
in hippocampal
dentate gyrus. Brain Res. Bull. 18:115-l 19;
1986.
Stanton, P. K.; Sarvey, J. M. Depletion on norepinephrine,
but not
serotonin, reduces long-term potentiation in the dentate gyrus of rat
hippocampal slices. J. Neurosci. 5:2169-2176;
1985.
Steriade, M. The excitaory-inhibitory
response sequence in thalamic
and neocortical cells: Stat related changes and regulaory systems.
In: Eldelman, X.; Gall, X.; Cowan, X., eds. Dynamic aspects of
neocortical function. New York: Wiley Interscience;
1984: 107157.
Steriade, M.; Llinas, R. The functional states of the thalamus and
the associated
neuronal interplay. Physiol. Rev. 68(3):649-742;
1988.
Waterhouse,
B. D.; Woodward,
D. Interaction of norepinephrine
with cerebrocortical
activity evoked by stimulation of somatosensory afferent pathways in rat. Exp. Neural. 67: 1 l-34; 1980.
Waterhouse, B. D.; Sessler, F. M.; Cheng, J.-T.; et al. New evidence
for a gating action of norepinephrine

30.

neu-

rons associated
118; 1989.
of the proofs,

with preparatory

which

were

set in rats. Neurosci.

not available

at press

Lett. 106: 112-

time.