The amygdala modulates memory for changes in reward magnitude: involvement of the amygdaloid GABAergic system

Behavioral neuroscience of psychological pain

Pain is a common word used to refer to a wide range of physical and mental states sharing hedonic aversive value. Three types of pain are distinguished in this article: Physical pain, an aversive state related to actual or potential injury and disease; social pain, an aversive emotion associated to social exclusion; and psychological pain, a negative emotion induced by incentive loss. This review centers on psychological pain as studied in nonhuman animals. After covering issues of terminology, the article briefly discusses the daily-life significance of psychological pain and then centers on a discussion of the results originating from two procedures involving incentive loss: successive negative contrast-the unexpected devaluation of a reward-and appetitive extinction-the unexpected omission of a reward. The evidence reviewed points to substantial commonalities, but also some differences and interactions between physical and psychological pains. This evidence is discussed in relation to behavioral, pharmacological, neurobiological, and genetic factors that contribute to the multidimensional experience of psychological pain.

Retrieval-induced NMDA receptor-dependent Arc expression in two models of cocaine-cue memory

The association of environmental cues with drugs of abuse results in persistent drug-cue memories. These memories contribute significantly to relapse among addicts. While conditioned place preference (CPP) is a well-established paradigm frequently used to examine the modulation of drug-cue memories, very few studies have used the non-preference-based model conditioned activity (CA) for this purpose. Here, we used both experimental approaches to investigate the neural substrates of cocaine-cue memories. First, we directly compared, in a consistent setting, the involvement of cortical and subcortical brain regions in cocaine-cue memory retrieval by quantifying activity-regulated cytoskeletal-associated (Arc) protein expression in both the CPP and CA models. Second, because NMDA receptor activation is required for Arc expression, we investigated the NMDA receptor dependency of memory persistence using the CA model. In both the CPP and CA models, drug-paired animals showed significant increases in Arc immunoreactivity in regions of the frontal cortex and amygdala compared to unpaired controls. Additionally, administration of a NMDA receptor antagonist (MK-801 or memantine) immediately after cocaine-CA memory reactivation impaired the subsequent conditioned locomotion associated with the cocaine-paired environment. The enhanced Arc expression evident in a subset of corticolimbic regions after retrieval of a cocaine-context memory, observed in both the CPP and CA paradigms, likely signifies that these regions: (i) are activated during retrieval of these memories irrespective of preference-based decisions, and (ii) undergo neuroplasticity in order to update information about cues previously associated with cocaine. This study also establishes the involvement of NMDA receptors in maintaining memories established using the CA model, a characteristic previously demonstrated using CPP. Overall, these results demonstrate the utility of the CA model for studies of cocaine-context memory and suggest the involvement of an NMDA receptor-dependent Arc induction pathway in drug-cue memory interference.

Sex-dependent impacts of low-level lead exposure and prenatal stress on impulsive choice behavior and associated biochemical and neurochemical manifestations

A prior study demonstrated increased overall response rates on a fixed interval (FI) schedule of reward in female offspring that had been subjected to maternal lead (Pb) exposure, prenatal stress (PS) and offspring stress challenge relative to control, prenatal stress alone, lead alone and lead+prenatal stress alone (Virgolini et al., 2008). Response rates on FI schedules have been shown to directly relate to measures of self-control (impulsivity) in children and in infants (Darcheville et al., 1992, 1993). The current study sought to determine whether enhanced effects of Pb±PS would therefore be seen in a more direct measure of impulsive choice behavior, i.e., a delay discounting paradigm. Offspring of dams exposed to 0 or 50ppm Pb acetate from 2 to 3 months prior to breeding through lactation, with or without immobilization restraint stress (PS) on gestational days 16 and 17, were trained on a delay discounting paradigm that offered a choice between a large reward (three 45mg food pellets) after a long delay or a small reward (one 45mg food pellet) after a short delay, with the long delay value increased from 0s to 30s across sessions. Alterations in extinction of this performance, and its subsequent re-acquisition after reinforcement delivery was reinstated were also examined. Brains of littermates of behaviorally-trained offspring were utilized to examine corresponding changes in monoamines and in levels of brain derived neurotrophic factor (BDNF), the serotonin transporter (SERT) and the N-methyl-d-aspartate receptor (NMDAR) 2A in brain regions associated with impulsive choice behavior. Results showed that Pb±PS-induced changes in delay discounting occurred almost exclusively in males. In addition to increasing percent long delay responding at the indifference point (i.e., reduced impulsive choice behavior), Pb±PS slowed acquisition of delayed discounting performance, and increased numbers of both failures to and latencies to initiate trials. Overall, the profile of these alterations were more consistent with impaired learning/behavioral flexibility and/or with enhanced sensitivity to the downshift in reward opportunities imposed by the transition from delay discounting training conditions to delay discounting choice response contingencies. Consistent with these behavioral changes, Pb±PS treated males also showed reductions in brain serotonin function in all mesocorticolimbic regions, broad monoamine changes in nucleus accumbens, and reductions in both BDNF and NMDAR 2A levels and increases in SERT in frontal cortex, i.e., in regions and neurotransmitter systems known to mediate learning/behavioral flexibility, and which were of greater impact in males. The current findings do not fully support a generality of the enhancement of Pb effects by PS, as previously seen with FI performance in females (Virgolini et al., 2008), and suggest a dissociation of the behaviors controlled by FI and delay discounting paradigms, at least in response to Pb±PS in rats. Collectively, however, the findings remain consistent with sex-dependent differences in the impacts of both Pb and PS and with the need to understand both the role of contingencies of reinforcement and underlying neurobiological effects in these sex differences.

The involvement of the GABAergic system in the formation and expression of the extinction memory in the crab Neohelice granulata

There is growing interest in the neurobiological mechanisms involved in the extinction of aversive memory. This cognitive process usually occurs after repeated or prolonged presentation of a conditioned stimulus that was previously associated with an unconditioned stimulus. If extinction is considered to be a new memory, the role of the γ-aminobutyric acid system (GABAergic system) during extinction memory consolidation should be similar to that described for the original trace. It is also accepted that negative modulation of the GABAergic system before testing can impair extinction memory expression. However, it seems possible to speculate that inhibitory mechanisms may be required in order to acquire a memory that is inhibitory in nature. Using a combination of behavioral protocols, such as weak and robust extinction training procedures, and pharmacological treatments, such as the systemic administration of GABAA agonist (muscimol) and antagonist (bicuculline), we investigated the role of the GABAergic system in the different phases of the extinction memory in the crab Neohelice granulata. We show that the stimulation of the GABAergic system impairs and its inactivation facilitates the extinction memory consolidation. Moreover, fine variations in the GABAergic tone affect its expression at testing. Finally, an active GABAergic system is necessary for the acquisition of the extinction memory. This detailed description may contribute to the understanding of the role of the GABAergic system in diverse aspects of the extinction memory.

Involvement of basolateral amygdala GABAA receptors in the effect of dexamethasone on memory in rats

In this study we investigated whether GABA(A) receptors of the basolateral amygdala (BLA) interact with the effect of dexamethasone on the retrieval stage of memory. Adult male Wistar rats were bilaterally cannulated in the BLA by stereotaxic surgery. The animals were trained in step-through apparatus by induction of electric shock (1.5 mA, 3 s) and were tested for memory retrieval after 1 d. The time of latency for entering the dark compartment of the instrument and the time spent by rats in this chamber were recorded for evaluation of the animals' retrieval in passive avoidance memory. Administration of dexamethasone (0.3 and 0.9 mg/kg, subcutaneously (s.c.)), immediately after training, enhanced memory retrieval. This effect was reduced by intra-BLA microinjection of muscimol (0.125, 0.250 and 0.500 µg/rat), when administered before 0.9 mg/kg of dexamethasone. Microinjection of bicuculline (0.75 µg/rat, intra-BLA) with an ineffective dose of dexamethasone (0.1 mg/kg, s.c.) increased memory retrieval. However, the same doses of muscimol and bicuculline without dexamethasone did not affect memory processes. Our data support reports that dexamethasone enhances memory retrieval. It seems that GABA(A) receptors of the BLA mediate the effect of dexamethasone on memory retrieval in rats.

Reward processing by the opioid system in the brain

The opioid system consists of three receptors, mu, delta, and kappa, which are activated by endogenous opioid peptides processed from three protein precursors, proopiomelanocortin, proenkephalin, and prodynorphin. Opioid receptors are recruited in response to natural rewarding stimuli and drugs of abuse, and both endogenous opioids and their receptors are modified as addiction develops. Mechanisms whereby aberrant activation and modifications of the opioid system contribute to drug craving and relapse remain to be clarified. This review summarizes our present knowledge on brain sites where the endogenous opioid system controls hedonic responses and is modified in response to drugs of abuse in the rodent brain. We review 1) the latest data on the anatomy of the opioid system, 2) the consequences of local intracerebral pharmacological manipulation of the opioid system on reinforced behaviors, 3) the consequences of gene knockout on reinforced behaviors and drug dependence, and 4) the consequences of chronic exposure to drugs of abuse on expression levels of opioid system genes. Future studies will establish key molecular actors of the system and neural sites where opioid peptides and receptors contribute to the onset of addictive disorders. Combined with data from human and nonhuman primate (not reviewed here), research in this extremely active field has implications both for our understanding of the biology of addiction and for therapeutic interventions to treat the disorder.

Basolateral amygdala lesions and sensitivity to reinforcer magnitude in concurrent chains schedules

Previous studies show that the basolateral amygdala (BLA) is required for behavior to adjust when the value of a reinforcer decreases after satiation or pairing with gastric distress. This study evaluated the effect of pre- or post-training excitotoxic lesions of the BLA on changes in preference with another type of contingency change, reinforcer magnitude reversal. Rats were trained to press left and right levers during a variable-interval choice phase for 50 microl or 150 microl sucrose delivered to consistent locations after a 16-s delay. Tones were presented during the first and last 2s of the delay to reinforcement. The tone frequency predicted the magnitude of sucrose reinforcement in baseline conditions. All groups acquired stable preference for the lever on the large (150 microl) reinforcer side. However, nose poking during the delay to large reinforcement was highly accurate (i.e., to the reinforced side) for all groups except the rats with BLA lesions induced before training, suggesting impaired control of behavior by the tone. After the acquisition of stable preference, the locations of the reinforcer magnitudes were unpredictably reversed for a single session. Pre-training lesions blunted changes in preference when the reinforcer magnitudes were reversed. Lesions induced after stable preference was acquired, but prior to reversal, did not disrupt changes in preference. The data suggest that the BLA contributes to the adaptation of choice behavior following changes in reinforcer magnitude. Impaired learning about the tone-reinforcer magnitude relationships may have disrupted discrimination of the reinforcer magnitude reversal.

Age-related changes in conditioned flavor preference in rats

Age-related changes have been documented in regions of the brain shown to process reward information. However, few studies have examined the effects of aging on associative memory for reward. The present study tested 7- and 24-month-old rats on a conditioned flavor preference task. Half of the rats in each age group received an unsweetened grape-flavored solution (CS-) on odd-numbered days and a sweetened cherry-flavored solution (CS+) on even-numbered days. The remaining rats in each age group received a sweetened grape-flavored solution (CS+) on odd-numbered days and an unsweetened cherry-flavored solution (CS-) on even-numbered days. During the acquisition phase of testing, the designated solution (CS+ or CS-) was presented to each rat for 15 min daily across six consecutive days. On the preference phase, each rat received unsweetened cherry and unsweetened grape-flavored solutions simultaneously for 15 min daily across four consecutive days. The 7-month-old rats showed a significant preference for the flavor that was previously sweetened during the acquisition phase (CS+) compared to the previously unsweetened solution (CS-) when the two unsweetened solutions were presented simultaneously during the preference phase of testing. In contrast, the 24-month-old rats did not show a preference and consumed roughly equal amounts of the previously sweetened (CS+) and unsweetened (CS-) solutions. Thus, the data suggest that the ability to form flavor-reward associations declines with increasing age, resulting in impaired conditioned flavor preference.

Functional interactions between the nucleus tractus solitarius (NTS) and nucleus accumbens shell in modulating memory for arousing experiences

The shell division of the nucleus accumbens receives noradrenergic input from neurons in the nucleus of the solitary tract (NTS) that transmit information regarding fluctuations in peripheral hormonal and autonomic activity. Accumbens shell neurons also receive converging inputs from limbic areas such as the hippocampus and amygdala that process newly acquired information. However, few studies have explored whether peripheral information regarding changes in emotional arousal contributes to memory processing in the accumbens. The beneficial effects on memory produced by emotional arousal and the corresponding activation of NTS neurons may be mediated through influences on neuronal activity in the accumbens shell during memory encoding. To explore this putative relationship, Experiment 1 examined interactions between the NTS and the accumbens shell in modulating memory for responses acquired after footshock training in a water-motivated inhibitory avoidance task. Memory for the noxious shock was significantly improved by posttraining excitation of noradrenergic NTS neurons. The enhanced retention produced by activating NTS neurons was attenuated by suppressing neuronal activity in the accumbens shell with bupivacaine (0.25%/0.5 microl). Experiment 2 examined the direct involvement of accumbens shell noradrenergic activation in the modulation of memory for psychologically arousing events such as a reduction in perceived reward value. Noradrenergic activation of the accumbens shell with phenylephrine (1.0 microg/0.5 microl) produced an enhancement in memory for the frustrating experience relative to control injections as evidenced by runway performance on an extended seven-day retention test. These findings demonstrate a functional relationship between NTS neurons and the accumbens shell in modulating memory following physiological arousal and identifies a role of norepinephrine in modulating synaptic activity in the accumbens shell to facilitate this process.

GABA(A) receptors of hippocampal CA1 regions are involved in the acquisition and expression of morphine-induced place preference

In the present study, the effects of bilateral intra-hippocampal CA1 (intra-CA1) injections of GABA(A) receptor agonist and/or antagonist on the acquisition and expression of morphine-induced place preference in male Wistar rats have been investigated. The conditioning treatments with subcutaneous (s.c.) injections of different doses of morphine (0.5-7.5 mg/kg) induced a conditioned place preference (CPP) for the drug-associated place in a dose-dependent manner. Intra-CA1 administration of the GABA(A) receptor agonist, muscimol (0.25, 0.5 and 1 microg/rat) significantly inhibited the morphine (5 mg/kg, s.c.)-induced CPP. Intra-CA1 injections of different doses of the GABA(A) receptor antagonist, bicuculline (0.25, 0.5 and 1 microg/rat), in combination with an ineffective dose of morphine (0.5 mg/kg, s.c.) elicited a significant CPP. However, muscimol or bicuculline by themselves did not elicit any effect on place conditioning. Furthermore, the muscimol-induced inhibition of morphine response was reversed by bicuculline (1 microg/rat, intra-CA1) administration. On the other hand, the bilateral intra-CA1 injections of muscimol (0.25, 0.5 and 1 microg/rat) or bicuculline (0.5, 1 and 2 microg/rat) significantly decreased the expression of morphine-induced CPP. Intra-CA1 administration of different doses of muscimol or bicuculline had no effect on locomotor activity in the testing phase. Our data indicated that the GABA(A) receptors of the hippocampal CA1 regions may play an important role in the acquisition and expression of morphine-induced place preference.

Neural systems implicated in delayed and probabilistic reinforcement

This review considers the theoretical problems facing agents that must learn and choose on the basis of reward or reinforcement that is uncertain or delayed, in implicit or procedural (stimulus-response) representational systems and in explicit or declarative (action-outcome-value) representational systems. Individual differences in sensitivity to delays and uncertainty may contribute to impulsivity and risk taking. Learning and choice with delayed and uncertain reinforcement are related but in some cases dissociable processes. The contributions to delay and uncertainty discounting of neuromodulators including serotonin, dopamine, and noradrenaline, and of specific neural structures including the nucleus accumbens core, nucleus accumbens shell, orbitofrontal cortex, basolateral amygdala, anterior cingulate cortex, medial prefrontal (prelimbic/infralimbic) cortex, insula, subthalamic nucleus, and hippocampus are examined.

GABA receptor subtype antagonists in the nucleus accumbens shell and ventral tegmental area differentially alter feeding responses induced by deprivation, glucoprivation and lipoprivation in rats

GABA(A) and GABA(B) receptor agonists stimulate feeding following microinjection into the nucleus accumbens shell and ventral tegmental area, effects blocked selectively and respectively by GABA(A) and GABA(B) receptor antagonists. GABA antagonists also differentially alter opioid-induced feeding responses elicited from these sites. Although GABA agonists and antagonists have been shown to modulate feeding elicited by deprivation or glucoprivation, there has been no systematic examination of feeding elicited by homeostatic challenges following GABA antagonists in these sites. Therefore, the present study examined the dose-dependent ability of GABA(A) (bicuculline, 75-150 ng) and GABA(B) (saclofen, 1.5-3 microg) antagonists administered into the nucleus accumbens shell or ventral tegmental area upon feeding responses elicited by food deprivation (24 h), 2-deoxy-D-glucose-induced glucoprivation (500 mg/kg) or mercaptoacetate-induced lipoprivation (70 mg/kg). A site-specific effect of GABA receptor antagonism was observed for deprivation-induced feeding in that both bicuculline and saclofen administered into the nucleus accumbens shell, but not the ventral tegmental area, produced short-term (1-4 h), but not long-term (24-48 h) effects upon deprivation-induced intake without meaningfully altering body weight recovery. In contrast to the relative inability of GABA receptor antagonism in both sites to alter 2-deoxy-D-glucose-induced intake, mercaptoacetate-induced intake was eliminated by saclofen and significantly reduced by bicuculline in the nucleus accumbens shell and eliminated by both bicuculline and saclofen in the ventral tegmental area. These data reinforce the findings that GABA(A) and GABA(B) receptors in the nucleus accumbens shell and ventral tegmental area are not only important in the modulation of pharmacologically induced feeding responses, but also participate in differentially mediating the short-term feeding response to food deprivation in the nucleus accumbens shell as well strongly modulating lipoprivic, but not glucoprivic feeding responses in both sites.

Enhancement of conditioned fear extinction by infusion of the GABAAagonist muscimol into the rat prefrontal cortex and amygdala

In auditory fear conditioning, repeated presentation of the tone in the absence of the shock leads to extinction of the acquired fear response. Both the infra limbic prefrontal cortex (IL) and the basolateral amygdala (BLA) are involved in extinction. In this study, we examine the involvement of these two regions in extinction by manipulating the gamma-aminobutyric acid (GABA)ergic system, in the Sprague-Dawley rat. We microinfused a low dose of the GABA(A) agonist muscimol into the IL or BLA. Muscimol infused to IL before extinction training, but not after either a short (five-trials) or long (15-trials) extinction training, resulted in long-term facilitation of extinction. Infusion of muscimol to the BLA following a short (five-trial) extinction session facilitated extinction at least 48-h post-drug infusion. The differences in the temporal parameters of the effects of muscimol in the IL or BLA, suggest differential involvement of these structures in long-term extinction of fear memory. We propose a facilitating role for GABA(A) neurotransmission in the IL in triggering the onset of fear extinction and its maintenance, whereas in the BLA, GABA(A) neurotransmission facilitates extinction consolidation. The involvement of GABA(A) receptors in fear extinction in the prefrontal cortex and amygdala is of particular interest, because of the role of these areas in emotional processes, and the role of the GABA(A) receptors in anxiety states.

Neural substrates of cocaine-cue associations that trigger relapse

Learned associations that occur during the process of repeated drug use in addiction can later manifest as trigger factors in relapse to renewed drug-seeking and drug-taking behavior. The process of conditioned-cued relapse of drug-seeking behavior has been successfully modeled in animals using the reinstatement procedure, in which chronic drug self-administration can be extinguished or withheld, and then reinstated using conditioned stimuli previously paired with the drug. Our laboratory has extensively studied the neural circuitry underlying conditioned-cued drug-seeking during the expression of reinstatement. In order to study the learning process of drug-cue pairings, we further developed a procedure whereby discrete cocaine-cue pairings can be conducted in a single pavlovian training session in animals previously trained to self-administer cocaine. Presentation of these cues during later reinstatement trials produces robust responding over extinction levels at levels similar to those seen when animals experience the cues on a daily basis. In a series of experiments, we have shown that reversible pharmacological inactivation of the basolateral complex of the amygdala just prior to acquisition of cocaine-cue associations blocks the ability of cocaine-paired stimuli to elicit conditioned-cued reinstatement. This learning process is mediated in part by muscarinic acetylcholine and dopaminergic inputs to the basolateral complex of the amygdala, as intra-amygdala infusion of selective receptor antagonists at the time of acquisition significantly affects reinstatement. We have also recently found that disruption of neural activity within the basolateral complex of the amygdala at the time of consolidation (just after cocaine-cue pairings) will disrupt reinstatement. Taken together, these results reveal the importance of the amygdala in the acquisition, consolidation, and expression of drug-stimulus learning that drives relapse to drug-seeking behavior.

The role of the GABA(A) agonist muscimol on memory performance: reward contingencies determine the nature of the deficit

A matching-to-position (MTP) paradigm was altered to influence the type of associations a rat would use to solve the task. Our main behavioral manipulation was the application of the differential outcomes procedure (DOP). The DOP involves correlating each to-be-remembered event with a distinct reward condition. This procedure results in the development of unique reward expectancies that enhance and guide choice behavior. Such distinct reward expectancies are not formed when either a common or random assignment of reward is used (a non-differential outcomes procedure [NOP]). Intracerebroventricular infusions of the amnestic agent muscimol (GABA(A) agonist) or aCSF were delivered to male rats trained on a delayed MTP task that implemented either the DOP or the NOP. Muscimol impaired performance in a dose dependent fashion in both groups--but the nature of the deficit differed as a function of reinforcement contingencies. Rats trained with the DOP displayed a non-mnemonic delay-independent impairment: performance at all delay intervals was disrupted. In contrast, NOP-trained rats displayed a delay-dependent impairment demonstrating that muscimol can also have memory-disrupting effects. The difference in pattern of impairment appears to be a function of the associations formed during training and the type of cognitive strategies involved in maintaining behavior on a conditional delayed discrimination task when reinforcement contingencies are varied. Thus, these results demonstrate that increasing GABA(A) receptor activation impairs a range of associative and memory functions.

Glucose regulation of memory for reward reduction in young and aged rats

Although baseline blood glucose levels in aged Fischer-344 rats are comparable to those of young rats, the rise in blood glucose in response to training-related stress is substantially attenuated. The diminished response may contribute to increased depletion of extracellular brain glucose levels during training in aged rats; the depletion is blocked and memory is enhanced by systemic injections of glucose. The present experiment examined the role of glucose in regulating memory for reward reduction training. Blood glucose levels exhibited a significant rise after reward reduction trials in young adult but not 2-year-old rats. Although young and aged rats exhibited comparable learning during the day of reward reduction training, aged rats exhibited more rapid forgetting of the learning response. Post-training glucose injections (200 mg/kg, i.p.) facilitated memory formation and slowed the rate of forgetting in young and old rats, consistent with the view that deficiencies in circulating glucose responses to training may contribute to the rapid forgetting evident in aged Fischer-344 rats.

GABA(A) receptors in the basolateral amygdala are involved in mediating morphine reward

In the present study, the effects of intra-basolateral amygdala (BLA) injection of GABA(A) receptor agonist and antagonist on morphine-induced conditioned place preference (CPP) in male Wistar rats have been investigated. Subcutaneous (s.c.) administration of different doses of morphine sulfate (1-9 mg/kg) produced a dose-dependent CPP. Using a 3-day schedule of conditioning, it was found that the GABA(A) receptor agonist, muscimol (0.125, 0.25 and 0.5 microg/rat) or the GABA(A) receptor antagonist, bicuculline (0.125, 0.25 and 0.5 microg/rat), did not produce a significant place preference or place aversion. Intra-BLA administration of muscimol (0.25 and 0.5 microg/rat) decreased the acquisition of CPP induced by morphine (6 mg/kg). On the other hand, intra-BLA injection of bicuculline (0.25 and 0.5 microg/rat) in combination with an ineffective dose of morphine (1 mg/kg) elicited a significant CPP. The response of different doses of muscimol was attenuated by bicuculline (0.125 and 0.25 microg/rat). Furthermore, intra-BLA administration of bicuculline but not muscimol before testing significantly decreased the expression of morphine (6 mg/kg)-induced place preference. The administration of the higher doses of bicuculline (0.25 and 0.5 microg/rat) during acquisition and the higher dose of muscimol (2 microg/rat) on the test day decreased the locomotor activity of the animals on the testing phase. It can be concluded that GABA(A) receptors in the amygdala are involved in morphine reward.

Post-training intra-basolateral amygdala infusions of norepinephrine enhance consolidation of memory for contextual fear conditioning

Post-training infusions of drugs, including noradrenergic agonists and antagonists, into the basolateral amygdala (BLA) influence the consolidation of memory for training in several tasks, including inhibitory avoidance. There is, however, conflicting evidence concerning whether post-training intra-BLA drug infusions modulate the consolidation of contextual fear conditioning (CFC). In the present study, norepinephrine (NE) was infused bilaterally into the BLA of male Sprague Dawley rats immediately after training on two CFC tasks: a Y-maze and a straight alley. Post-training intra-BLA infusions enhanced memory of CFC training in the Y-maze, as assessed by percentage of time spent freezing and shock arm entrance latencies. Post-training intra-BLA infusions of NE enhanced 48 hr retention of CFC training in the straight alley, as assessed by shock compartment entrance latencies and the number of shocks required to learn to avoid entering the shock compartment. These findings indicate that the consolidation of memory for CFC, like that for inhibitory avoidance training, is influenced by post-training neuromodulatory influences within the BLA. Thus, the findings provide additional evidence consistent with the hypothesis that the BLA has a general role in modulating memory consolidation.

Role of the basolateral amygdala in memory consolidation

Typically, emotionally charged events are better remembered than neutral ones. This paper reviews data indicating that the amygdala is responsible for this facilitation of memory by emotional arousal. Pharmacological and behavioral studies have shown that the release of adrenal stress hormones facilitates memory consolidation. The available evidence suggests that this effect depends on a central action of stress hormones involving the release of the neuromodulators noradrenaline (NA) and acetylcholine in the basolateral complex of the amygdala (BLA). Indeed, BLA lesions block the memory modulating effects of stress hormones. Moreover, microdialysis studies have revealed that BLA concentrations of NA and acetylcholine are transiently (2h) elevated following emotionally arousing learning episodes. Last, post-learning intra-BLA injections of beta-adrenergic or muscarinic receptor antagonists reduce retention. These results have led to the hypothesis that NA and acetylcholine increase the activity of BLA neurons in the hours after the learning episode. In turn, the BLA would facilitate synaptic plasticity in other brain structures, believed to constitute the storage sites for different types of memory. Consistent with this, post-learning treatments that reduce or enhance the excitability of BLA neurons respectively decrease or improve long-term retention on various emotionally charged learning tasks. However, a number of issues remain unresolved. Chief among them is how the BLA facilitates synaptic plasticity elsewhere in the brain. The present review concludes with a consideration of this issue based on recent advances in our understanding of the BLA. Among other possibilities, it is suggested that rhythmic BLA activity at the theta frequency during arousal as well as the uniform conduction times of BLA axons to distributed rhinal sites may promote plasticity in co-active structures of the temporal lobe.

Selective inactivation of the dorsomedial prefrontal cortex and the basolateral amygdala attenuates conditioned-cued reinstatement of extinguished cocaine-seeking behavior in rats

RATIONALE

Environmental stimuli previously paired with cocaine can induce craving in humans and reinstate extinguished cocaine-seeking behavior in laboratory animals. Previous evidence has implicated the amygdala and the prefrontal cortex (PFC) as possible substrates for conditioned-cued relapse.

OBJECTIVES

In order to test directly the role of the PFC in a model of relapse, the present study examined the effects of reversible inactivation of three medial PFC areas, the anterior cingulate (ACing), the prelimbic cortex (PL), and the infralimbic cortex (IL), on the expression of conditioned-cued reinstatement of extinguished cocaine-seeking behavior. We also tested the involvement of the basolateral amygdala (BLA) and the parietal cortex immediately dorsal to the BLA, sensory cortex area 1 - barrel field (S1BF).

METHODS

During daily 3-h sessions, rats pressed a lever for IV cocaine infusions that were paired with a light-tone (LT) presentation. Following extinction of lever pressing in the absence of the LT, reinstatement of extinguished lever pressing was measured during response-contingent presentations of the LT in the absence of cocaine. For localized reversible inactivation, tetrodotoxin (TTX) (5 ng/0.5 micro l/side) or vehicle was bilaterally infused just prior to reinstatement testing.

RESULTS

TTX inactivation of the BLA, ACing, or PL impaired the ability of LT presentations to reinstate extinguished lever pressing for cocaine-paired stimuli. In contrast, inactivation of the IL or the S1BF had no effect on conditioned-cued reinstatement. Furthermore, there was no effect of TTX in any of the tested brain regions on general locomotor activity. CONCLUSIONS. These results support a role for the dorsomedial PFC and the BLA in the circuitry that mediates drug-seeking behavior elicited by cocaine-associated stimuli. Placed within the context of recent studies using drug-primed and stress-induced reinstatement models, we suggest that the dorsomedial PFC may serve as a common link in the neural circuitry underlying reinstatement of drug-seeking behavior.

Effects of expectations for different reward magnitudes on neuronal activity in primate striatum

In behavioral science, it is well known that humans and nonhuman animals are highly sensitive to differences in reward magnitude when choosing an outcome from a set of alternatives. We know that a realm of behavioral reactions is altered when animals begin to expect different levels of reward outcome. Our present aim was to investigate how the expectation for different magnitudes of reward influences behavior-related neurophysiology in the anterior striatum. In a spatial delayed response task, different instruction pictures are presented to the monkey. Each image represents a different magnitude of juice. By reaching to the spatial location where an instruction picture was presented, animals could receive the particular liquid amount designated by the stimulus. Reliable preferences in reward choice trials and differences in anticipatory licks, performance errors, and reaction times indicated that animals differentially expected the various reward amounts predicted by the instruction cues. A total of 374 of 2,000 neurons in the anterior parts of the caudate nucleus, putamen, and ventral striatum showed five forms of task-related activation during the preparation or execution of movement and activations preceding or following the liquid drop delivery. Approximately one-half of these striatal neurons showed differing response levels dependent on the magnitude of liquid to be received. Results of a linear regression analysis showed that reward magnitude and single cell discharge rate were related in a subset of neurons by a monotonic positive or negative relationship. Overall, these data support the idea that the striatum utilizes expectancies that contain precise information concerning the predicted, forthcoming level of reward in directing general behavioral reactions.

Cholinergic modulation of memory in the basolateral amygdala involves activation of both m1 and m2 receptors

Muscarinic cholinergic activation is a critical component of basolateral amygdala (BLA)-mediated modulation of memory consolidation. The receptor(s) mediating this activation during consolidation have not been elucidated. This study investigated the roles of muscarinic subtype 1 (m1) and subtype 2 (m2) receptors in memory enhancement, by post-training intra-BLA infusions of the non-selective muscarinic agonist oxotremorine. Rats received intra-BLA infusions of either oxotremorine alone (10 microg in 0.2 microl per side), oxotremorine together with the selective m1 antagonist telenzipine (1.7, 5.0, 17 or 50 nmol/side), oxotremorine with the selective m2 antagonist methoctramine (1.7, 5.0, 17 or 50 nmol/side), oxotremorine with a combination of the above doses of telenzipine and methoctramine, or only vehicle, immediately after inhibitory avoidance training. Performance on a 48-hour retention test was significantly enhanced in oxotremorine-treated rats relative to vehicle-infused controls. Intra-BLA co-infusion of oxotremorine with either telenzipine (5, 17 or 50 nmol/side) or methoctramine (17 or 50 nmol/side) blocked the oxotremorine-induced enhancement. Combinations of these antagonists did not act additively to block memory enhancement by oxotremorine. These findings indicate that modulation of memory consolidation induced by cholinergic influences within the BLA requires activation of both m1 and m2 receptor synapses. Plausible mechanisms for m1- and m2-mediated influences on BLA circuitry are discussed.

The role of the amygdala in conditioned flavor preference

Long-Evans rats with control or amygdala lesions were tested in a conditioned flavor preference task. Half of the rats in each lesion group received an unsweetened grape-flavored solution on odd-numbered days and a sweetened cherry-flavored solution on even-numbered days. The remaining rats received a sweetened grape-flavored solution on odd-numbered days and an unsweetened cherry-flavored solution on even-numbered days. The appropriate solution was presented once a day for 15 min to each rat in the homecage. After six days of testing, each rat received unsweetened cherry and grape flavored solutions simultaneously for 15 min daily across four days. When the two unsweetened flavored solutions were presented simultaneously control rats showed a significant preference for the flavor that was sweetened during training compared to the unsweetened solution. However, amygdala-lesioned rats did not show a preference. The data suggest that the amygdala may be critically involved in mediating reward-based conditioned flavor preference.

GABAergic regulation of the central nucleus of the amygdala: implications for sleep control

It is becoming established that the amygdala has a strong influence on arousal state, with most evidence indicating a role in the regulation of rapid eye movement sleep (REM). Electrically activating the central nucleus of the amygdala (CNA) can increase subsequent REM and enhance REM-related phenomena. However, drugs that may be inhibitory to CNA have been typically reported to reduce REM. This suggests that enhancing activity in CNA could promote REM, and that inhibiting activity in CNA could suppress REM. We reversibly inactivated CNA using the GABA(A) agonist, muscimol, or blocked GABAergic inhibition with the GABA(A) antagonist, bicuculline, and examined the effects on sleep and wakefulness. Rats (90-day-old male Sprague-Dawley) were implanted with electrodes for recording EEG and EMG. Cannulae were aimed into CNA for microinjecting muscimol (0.001, 0.3 and 1.0 microM/0.2 microl saline) or bicuculline (56 and 333 pM/0.2 microl saline). Each animal received bilateral microinjections of muscimol, bicuculine or saline alone followed by 6-h sleep recordings. Microinjections of low concentrations of muscimol into CNA produced relatively selective decreases in total REM and number of REM episodes that lasted up to 6 h. In contrast, microinjections of bicuculline into CNA produced significant increases in REM. There were no significant reductions in NREM or wakefulness. These findings demonstrate that inactivating CNA can produce a relatively selective suppression of REM. The possible role that spontaneous activity in CNA may play in REM initiation and/or maintenance is discussed.

Appetitive conditioning-induced plasticity is expressed during paradoxical sleep in the medial geniculate, but not in the lateral amygdala

This study examined whether neurons in the medial division of the medial geniculate (MGm) and the dorsal part of the lateral amygdala (LAd) express learning-induced plasticity in paradoxical sleep (PS) after appetitive conditioning, as they do in PS after fear conditioning. Rats received tone-food pairings in 3 sessions. After each session, the tone was presented at a nonawakening intensity during PS. Multiunit activity was simultaneously recorded in MGm and LAd. During waking, increases in tone-evoked discharges developed in MGm and LAd; however, as training continued, they lessened in LAd, but not in MGm. During PS, conditioned tone responses were expressed in MGm, but not in LAd. Thus, these results demonstrate dissociation of MGm and LAd plasticity. Moreover, compared with fear conditioning results, they suggest that expression of amygdalar plasticity in PS depends on the emotional salience of the stimulus.

The amygdala but not the hippocampus is involved in pattern separation based on reward value

A total of 32 male Long-Evans rats were tested on a modified version of Flaherty, Turovsky, and Krauss's (1994) anticipatory contrast paradigm to assess pattern separation for reward value. Prior to testing, each rat received either a control, a hippocampal, or an amygdala lesion. In the home cage, each rat was allowed to drink a water solution containing 2% sucrose for 3 min followed by a water solution containing 32% sucrose for 3 min. Across 10 days of testing, the rats in each lesion group showed significantly increased anticipatory discriminability as a function of days. To assess the operation of a pattern separation mechanism, each rat was then tested using the same procedure except the 2% solution was followed by a 16% solution for 10 days and then by an 8% solution for 10 days. Control and hippocampal-lesioned rats continued to show high discriminability when the 2% solution was followed by a 16% solution; however, the amygdala-lesioned rats showed low anticipatory discriminability. On trials where the 2% sucrose solution was followed by an 8% sucrose solution, all groups showed low discriminability scores, suggesting that when two reward values are very similar even control animals are not able to separate the reward values in memory. However, the results of a preference task revealed that all groups can perceptually discriminate between a 2% and an 8% sucrose solution. The data suggest that the amygdala but not the hippocampus is involved in the separation of patterns based on reward value.

Long-term depression in the basolateral amygdala of the mouse involves the activation of interneurons

Long-term depression (LTD) in the basolateral amygdala, following low frequency stimulation (1 Hz/900 pulses) of the lateral amygdala, was studied in an in vitro slice preparation of 2-3 weeks and 2-4 months old mice. Whole-cell patch-clamp recordings of neurons, visualized by means of infrared videomicroscopy, and extracellular field potential recordings were performed. Loading single neurons with the calcium chelator BAPTA (30 mM) did not reduce the excitatory postsynaptic currents following low frequency stimulation. However, buffering presynaptic calcium with BAPTA-AM, and application of the specific Ca2+/calmodulin-stimulated protein kinase II antagonist KN-62 (1-[N,O-bis(5-isoquinoline sulfonyl)-N-methyl-L-tyrosyl]-4-phenylpiperizine), blocked low frequency stimulation-induced LTD. The induction of LTD was reduced by the competitive N-methyl-D-aspartate receptor antagonist D-(-)-2-amino-5-phosphonopentanoic acid (50 microM), and blocked by the metabotropic glutamate receptor antagonist (-)-amino-4-carboxy-methyl-phenylacetic acid (1 mM), and by 5-hydroxytryptamine (5-HT; 30 microM) via the activation of 5-HT(1A) receptors. Also blocking GABA(A) receptor-mediated synaptic transmission with bicuculline (10 microM) or picrotoxin (20 microM) reduced the induction of LTD. Visually and electrophysiologically identified interneurons in slices from 2 weeks old mice, expressed in contrast to adult mice (2-4 months), pronounced LTD. Principal neurons showed only weak LTD after low frequency stimulation.A synopsis of these findings suggests a pivotal role of GABAergic interneurons and serotonergic afferents in the induction of LTD in the basolateral nucleus of the amygdala.

Synapses on GABAergic neurons in the basolateral nucleus of the rat amygdala: double-labeling immunoelectron microscopy

Although the basolateral nucleus (BL) of the amygdala is known to contain an abundance of gamma-aminobutyric acid (GABA)ergic neurons that regulate the amygdaloid projection neurons and influence storage and consolidation of memory, it remains to be determined what type of neuronal input controls GABAergic neurons in the BL. We examined the synapses that GABAergic neurons form with GABAergic and noradrenergic neurons and terminals with unknown transmitters by double-labeling immunoelectron microscopy using anti-GABA and dopamine-beta-hydroxylase (DBH) antisera. The medium and small dendrites of the GABAergic neurons were shown to receive symmetric, inhibitory-type synapses from GABAergic axon terminals and asymmetric, excitatory-type synapses from noradrenergic axon terminals. Each segment of the GABAergic neurons from perikarya to dendritic spines received both symmetric and asymmetric synapses from unlabeled axon terminals of various forms and sizes. The incidence rates of the two types of synapses were almost identical. Our results suggest that GABAergic neurons in the BL of the rat amygdala might be affected by the excitatory influence of the noradrenergic system and the inhibitory influence of the GABAergic system. Furthermore, these neurons are also strongly influenced by both excitatory and inhibitory-type synapses from neuronal systems other than the GABAergic and noradrenergic systems.

Induction of NGFI-B mRNA following contextual fear conditioning and its blockade by diazepam

Expression of the immediate-early gene, NGFI-B (nerve growth factor inducible gene B), was examined in the amygdala, hippocampus, and neocortex following contextual fear conditioning. Rats were either handled, placed within the testing context without receiving the footshock, received a footshock immediately upon placement within the context, or received a footshock after a 3-min delay (delayed-shock). Only the delayed-shock group displayed a fear response (freezing) in the post-shock period and in a retention test 24 h after fear conditioning. Expression of NGFI-B mRNA was increased in the dorsolateral part of the lateral nucleus of the amygdala (LaDL) and the neocortex 30 min following conditioning in the delayed-shock group compared to the other three groups. In addition, following a retention test conducted 24 h after fear conditioning, NGFI-B mRNA expression was increased in the neocortex of the delayed-shock group compared to the handled group. In a subsequent experiment, the effects of pretreatment with the anxiolytic drug, diazepam, on fear conditioning and the concomitant increases in NGFI-B mRNA were investigated. Rats administered a 2.5 mg/kg, i.p. dose of diazepam before fear conditioning did not acquire contextual fear as demonstrated by a lack of freezing in a retention test. Although diazepam blocked fear conditioning while the 40% propylene glycol, 10% ethanol vehicle solution did not, both diazepam and the vehicle reduced the conditioning-induced increase in NGFI-B expression in the LaDL. In contrast, the fear-conditioning-induced NGFI-B increase in the neocortex was blocked by diazepam, but not by the vehicle. The data suggest that the transcriptional factor NGFI-B in the LaDL and neocortex may play a functional role in learning and memory of contextual fear, but blocking the increase in NGFI-B expression in the LaDL is not essential for diazepam to interfere with fear conditioning.

Differential amygdala responses to winning and losing: a functional magnetic resonance imaging study in humans

The amygdala has been shown to respond to many distinct types of affective stimuli, including reward and punishment feedback in animals. In humans, winning and losing situations can be considered as reward and punishment experiences, respectively. In this study, we used functional magnetic resonance imaging (fMRI) to measure regional brain activity when human subjects were given feedback on their performance during a simple response time task in a fictitious competitive tournament. Lexical stimuli were used to convey positive 'win' or negative 'lose' feedback. The frequency of positive and negative trials was parametrically varied by the experimenters independently from the subjects' actual performance and unbeknownst to them. The results showed that the parametric increase of winning was associated with left amygdala activation whereas the parametric increase of losing was associated with right amygdala activation. These findings provide functional evidence that the human amygdala differentially responds to changes in magnitude of positive or negative reinforcement conveyed by lexical stimuli.

Differential expression of EGR-1 mRNA in the amygdala following diazepam in contextual fear conditioning

The amygdaloid complex is thought to be a major site of action of anxiolytic benzodiazepine agonists. To investigate whether activity in the amygdaloid complex is altered with anxiolytic effects of diazepam, mRNA expression of the immediate-early gene EGR-1 was examined in the amygdala following blockade of fear conditioning by diazepam. It was previously shown that mRNA expression of EGR-1 (also called, NGFI-A, Zif 268, Krox 24) increases in the lateral nucleus of the amygdala (LA) shortly following contextual fear conditioning. It was therefore hypothesized that diazepam would block both contextual fear and the concomitant increase in EGR-1 mRNA expression in the LA induced by fear conditioning. Rats administered systemic diazepam before fear conditioning displayed both anxiolytic effects during the post-shock period and amnesic effects during a retention test 24 h later. Diazepam blocked the fear-conditioning-induced increase in EGR-1 expression in the LA. In addition, diazepam significantly increased EGR-1 mRNA expression in the central nucleus of the amygdala (CeA) in a dose-dependent manner. The results reveal differential regulation of EGR-1 by diazepam in the central and lateral nuclei of the amygdala suggesting that these two amygdala nuclei act in a reciprocal manner during the anxiolytic and amnesic action of the benzodiazepine agonist.

Alcohol drinking produces brain region-selective changes in expression of inducible transcription factors

Mapping the effects of alcohol consumption on neural activity could provide valuable information on mechanisms of alcohol's effects on behavior. The present study sought to identify effects of alcohol consumption on expression of inducible transcription factors (ITFs) in mouse brain. C57BL/6J mice were trained to consume 10% ethanol/10% sucrose solution during a 30-min limited access period. Control animals were given access to 10% sucrose solution or water. Following the final day of the procedure, animals were sacrificed and immunohistochemical analyses were performed for three ITFs (c-Fos, FosB, and Zif268). Alcohol-consuming animals had increased ITF expression in several brain areas. Specifically, c-Fos was significantly induced in the nucleus accumbens core (AcbC), the medial posteroventral portion of the central nucleus of the amygdala (CeMPV), and the Edinger-Westphal nucleus (EW). Expression of c-Fos was significantly lower in the dentate gyrus of alcohol-consuming animals vs. sucrose-consuming animals. However, it was not significantly different from the water controls. Induction of c-Fos in AcbC, CeMPV and EW was significantly related to blood alcohol concentrations (BAC). Furthermore, FosB expression in the CeMPV and the EW was also significantly higher in the alcohol-consuming animals vs. water controls. FosB expression in the EW was significantly related to BAC. The significance of these results is two-fold. First, our experiments demonstrate that ITF mapping is an effective strategy in identifying alcohol-induced changes following voluntary consumption. Second, they suggest a relationship between ITF expression in AcbC, CeMPV and EW and the level of alcohol intoxication.

Serotonergic modulation of neurotransmission in the rat basolateral amygdala

Whole cell patch-clamp recordings were obtained from projection neurons and interneurons of the rat basolateral amygdala (BLA) to understand local network interactions in morphologically identified neurons and their modulation by serotonin. Projection neurons and interneurons were characterized morphologically and electrophysiologically according to their intrinsic membrane properties and synaptic characteristics. Synaptic activity in projection neurons was dominated by spontaneous inhibitory postsynaptic currents (IPSCs) that were multiphasic, reached 181 +/- 38 pA in amplitude, lasted 296 +/- 27 mS, and were blocked by the GABAA receptor antagonist, bicuculline methiodide (30 microM). In interneurons, spontaneous synaptic activity was characterized by a burst-firing discharge patterns (200 +/- 40 Hz) that correlated with the occurrence of 6-cyano-7-nitroquinoxaline-2,3-dione-sensitive, high-amplitude (260 +/- 42 pA), long-duration (139 +/- 19 mS) inward excitatory postsynaptic currents (EPSCs). The interevent interval of 831 +/- 344 mS for compound inhibitory postsynaptic potentials (IPSPs), and 916 +/- 270 mS for EPSC bursts, suggested that spontaneous IPSP/Cs in projection neurons are driven by burst of action potentials in interneurons. Hence, BLA interneurons may regulate the excitability of projection neurons and thus determine the degree of synchrony within ensembles of BLA neurons. In interneurons 5-hydroxytryptamine oxalate (5-HT) evoked a direct, dose-dependent, membrane depolarization mediated by a 45 +/- 6.9 pA inward current, which had a reversal potential of -90 mV. The effect of 5-HT was mimicked by the 5-HT2 receptor agonist, alpha-methyl-5-hydroxytryptamine (alpha-methyl-5-HT), but not by the 5-HT1A receptor agonist, (+/-) 8-hydroxydipropylaminotetralin hydrobromide (8-OH-DPAT), or the 5-HT1B agonist, CGS 12066A. In projection neurons, 5-HT evoked an indirect membrane hyperpolarization ( approximately 2 mV) that was associated with a 75 +/- 42 pA outward current and had a reversal potential of -70 mV. The response was independent of 5-HT concentration, blocked by TTX, mimicked by alpha-methyl-5-HT but not by 8-OH-DPAT. In interneurons, 5-HT reduced the amplitude of the evoked EPSC and in the presence of TTX (0.6 microM) reduced the frequency of miniature EPSCs but not their quantal content. In projection neurons, 5-HT also caused a dose-dependent reduction in the amplitude of stimulus evoked EPSCs and IPSCs. These results suggest that acute serotonin release would directly activate GABAergic interneurons of the BLA, via an activation of 5-HT2 receptors, and increase the frequency of inhibitory synaptic events in projection neurons. Chronic serotonin release, or high levels of serotonin, would reduce the excitatory drive onto interneurons and may act as a feedback mechanism to prevent excess inhibition within the nucleus.

Posttraining intraamygdala infusions of oxotremorine and propranolol modulate storage of memory for reductions in reward magnitude

These experiments examined the effects of posttraining intraamygdala administration of the muscarinic agonist, oxotremorine, and the beta-noradrenergic antagonist, propranolol, on memory for reduction in reward magnitude. Male Sprague-Dawley rats (175-200 g) implanted with bilateral intraamygdala cannulae were food deprived (maintained at 80% of body weight) and trained to run a straight alley (six trials/day) for either ten 45-mg food pellets (high reward) or one 45-mg food pellet (low reward) for 10 days. In Experiment One, the animals in the high-reward group were than shifted to a one-pellet reward and immediately given intraamygdala infusions (0.5 microliter/side) of either oxotremorine (10 ng) or phosphate buffer. Shifted training continued for 4 more days and no further injections were given. Shifted animals given the buffer solution displayed an increase in runway latencies but returned to preshift latencies by the fifth day of shifted training. In contrast, animals given oxotremorine exhibited increased latencies through the fifth day. In Experiment Two, rats were trained as in Experiment. One but immediately following the shift received intraamygdala infusions of oxotremorine (10 ng), propranolol (0.3 microgram), both, or phosphate buffer. Shifted vehicle-injected rats returned to preshift performance by the fifth day of shifted training. Shifted propranolol rats returned to preshift latencies by the third day of shifted training. In contrast, the shifted oxotremorine and the shifted oxotremorine/propranolol rats displayed longer latencies than unshifted controls through 5 days of shifted training. The findings indicate that the muscarinic cholinergic and beta-noradrenergic systems within the amygdala interact in regulating memory and support the view that noradrenergic influences are mediated through cholinergic activation.

Involvement of the amygdala in memory storage: interaction with other brain systems

There is extensive evidence that the amygdala is involved in affectively influenced memory. The central hypothesis guiding the research reviewed in this paper is that emotional arousal activates the amygdala and that such activation results in the modulation of memory storage occurring in other brain regions. Several lines of evidence support this view. First, the effects of stress-related hormones (epinephrine and glucocorticoids) are mediated by influences involving the amygdala. In rats, lesions of the amygdala and the stria terminalis block the effects of posttraining administration of epinephrine and glucocorticoids on memory. Furthermore, memory is enhanced by posttraining intraamygdala infusions of drugs that activate beta-adrenergic and glucocorticoid receptors. Additionally, infusion of beta-adrenergic blockers into the amygdala blocks the memory-modulating effects of epinephrine and glucocorticoids, as well as those of drugs affecting opiate and GABAergic systems. Second, an intact amygdala is not required for expression of retention. Inactivation of the amygdala prior to retention testing (by posttraining lesions or drug infusions) does not block retention performance. Third, findings of studies using human subjects are consistent with those of animal experiments. beta-Blockers and amygdala lesions attenuate the effects of emotional arousal on memory. Additionally, 3-week recall of emotional material is highly correlated with positronemission tomography activation (cerebral glucose metabolism) of the right amygdala during encoding. These findings provide strong evidence supporting the hypothesis that the amygdala is involved in modulating long-term memory storage.