Glucocorticoid receptor activation in the rat nucleus of the solitary tract facilitates memory consolidation: involvement of the basolateral amygdala

These experiments examined the involvement of glucocorticoid receptors (GRs or type II) located in the A2-noradrenergic cell group of the rat nucleus of the solitary tract (NTS) in modulating memory storage. Bilateral intra-NTS infusions (0.5 microL) of the specific GR agonist RU 28362 (11beta, 17beta-dihydroxy-6, 21-dimethyl-17alpha-pregna-4,6-trien-20yn-3-one), in doses ranging from 0.01 to 10.0 ng, immediately after inhibitory avoidance training produced a dose-dependent enhancement of 48 h retention performance. Infusions of 0.1 or 1.0 ng of the agonist enhanced retention, whereas lower or higher doses were ineffective. Post-training infusions of the GR antagonist RU 38486 [17beta-hydroxy-11beta-(4-dimethylaminophenyl)-17alpha-(1-pr opynyl)-o estra-4,9-dien-3-one, 0.01-10.0 ng] into the NTS did not significantly affect retention performance, but shifted the dose-response effects of post-training systemic injections of the synthetic glucocorticoid dexamethasone to the right. These results indicate that activation of GRs in the NTS can influence memory formation for inhibitory avoidance training, and suggest that the effects of circulating glucocorticoids on memory are mediated, in part, by an activation of GRs in the NTS. Additionally, pretraining infusions of the beta1-adrenergic antagonist atenolol (0.5 microg in 0.2 microL) into the basolateral nucleus of the amygdala (BLA), a brain structure which receives noradrenergic projections from the NTS and is implicated in memory storage modulation, blocked the memory-enhancing effects of the GR agonist (1.0 ng) infused into the NTS. These findings provide evidence that memory storage is modulated by glucocorticoid binding to GRs in noradrenergic cell bodies in the NTS and suggest that these modulatory effects are conveyed by ascending projections to the BLA.     

Glucocorticoid enhancement of memory consolidation in the rat is blocked by muscarinic receptor antagonism in the basolateral amygdala

Glucocorticoid-induced memory enhancement is known to depend on beta-adrenoceptor activation in the basolateral amygdala (BLA). Additionally, inactivation of muscarinic cholinergic receptors in the rat amygdala blocks memory enhancement induced by concurrent beta-adrenergic activation. Together, these findings suggest that glucocorticoid-induced modulation of memory consolidation requires cholinergic as well as adrenergic activation in the BLA. Two experiments investigated this issue. The first experiment examined whether blockade of muscarinic cholinergic receptors in the BLA with atropine alters the memory-enhancing effects of the systemically administered glucocorticoid dexamethasone. Dexamethasone (0.3, 1.0 or 3.0 mg/kg, s.c.) administered to rats immediately after inhibitory avoidance training produced dose-dependent enhancement of 48-h retention. Concurrent bilateral infusions of the muscarinic cholinergic antagonist atropine (0.5 microg in 0.2 microL per side) into the BLA blocked the memory enhancement. The second experiment investigated whether the BLA is a locus of interaction between glucocorticoid and muscarinic activation. The specific glucocorticoid receptor (GR or type II) agonist RU 28362 (1.0, 3.0 or 10 ng) was infused into the BLA either alone or together with atropine immediately after training. The GR agonist produced dose-dependent memory enhancement and atropine blocked the memory enhancement. These findings indicate that muscarinic cholinergic activation within the BLA is critical for enabling glucocorticoid enhancement of memory consolidation and that enhancement of memory induced by GR activation in the BLA requires cholinergic activation within the BLA.     

Glucocorticoids interact with the basolateral amygdala beta-adrenoceptor--cAMP/cAMP/PKA system in influencing memory consolidation

Infusion of a beta-adrenoceptor antagonist into the basolateral nucleus of the amygdala (BLA) blocks memory enhancement induced by systemic or intra-BLA administration of a glucocorticoid receptor (GR) agonist. As there is evidence that glucocorticoids interact with the noradrenergic signalling pathway in activating adenosine 3prime prime or minute,5prime prime or minute-cyclic monophosphate (cAMP), the present experiments examined whether glucocorticoids influence the beta-adrenoceptor--cAMP system in the BLA in modulating memory consolidation. Male, Sprague--Dawley rats received bilateral infusions of atenolol (a beta-adrenoceptor antagonist), prazosin (an alpha1-adrenoceptor antagonist) or Rp-cAMPS (a protein kinase A inhibitor) into the BLA 10 min before inhibitory avoidance training and immediate post-training intra-BLA infusions of the GR agonist, RU 28362. Atenolol and Rp-cAMPS, but not prazosin, blocked 48-h retention enhancement induced by RU 28362. A second series of experiments investigated whether a GR antagonist alters the effect of noradrenergic activation in the BLA on memory consolidation. Bilateral intra-BLA infusions of the GR antagonist, RU 38486, administered 10 min before inhibitory avoidance training completely blocked retention enhancement induced by alpha1-adrenoceptor activation and attenuated the dose--response effects of post-training intra-BLA infusions of clenbuterol (a beta-adrenoceptor agonist). However, the GR antagonist did not alter retention enhancement induced by post-training intra-BLA infusions of 8-Br-cAMP (a synthetic cAMP analogue). These findings suggest that glucocorticoids influence the efficacy of noradrenergic stimulation in the BLA on memory consolidation via an interaction with the beta-adrenoceptor--cAMP cascade, at a locus between the membrane-bound beta-adrenoceptor and the intracellular cAMP formation site.     

1999 Curt P. Richter award. Glucocorticoids and the regulation of memory consolidation

This paper summarizes recent findings on the amygdala's role in mediating acute effects of glucocorticoids on memory consolidation in rats. Posttraining activation of glucocorticoid-sensitive pathways involving glucocorticoid receptors (GRs or type II) enhances memory consolidation in a dose-dependent inverted-U fashion. Selective lesions of the basolateral nucleus of the amygdala (BLA) or infusions of beta-adrenoceptor antagonists into the BLA block the memory-modulatory effects of systemic injections of glucocorticoids. Additionally, posttraining infusions of a specific GR agonist administered directly into the BLA enhance memory consolidation, whereas those of a GR antagonist impair. These findings indicate that glucocorticoid effects on memory consolidation are mediated, in part, by an activation of GRs in the BLA and that the effects require beta-adrenergic activity in the BLA. Other findings indicate that the BLA interacts with the hippocampus in mediating glucocorticoid-induced modulatory influences on memory consolidation. Lesions of the BLA or inactivation of beta-adrenoceptors within the BLA also block the memory-modulatory effects of intrahippocampal administration of a GR agonist or antagonist. These findings are in agreement with the general hypothesis that the BLA integrates hormonal and neuromodulatory influences on memory consolidation. However, the BLA is not a permanent locus of storage for this information, but modulates consolidation processes for explicit/associative memories in other brain regions, including the hippocampus.     

The memory-modulatory effects of glucocorticoids depend on an intact stria terminalis

This study examined the effects of stria terminalis (ST) lesions on glucocorticoid-induced modulation of memory formation for inhibitory avoidance training and spatial learning in a water maze. Systemic (s.c.) posttraining injections of the glucocorticoid receptor agonist dexamethasone (0.3 or 1.0 mg/kg) enhanced memory for inhibitory avoidance training in rats with sham ST lesions. Removal of the adrenal glands (adrenalectomy; ADX) significantly impaired spatial memory in a water maze, and immediate posttraining injections of dexamethasone (0.3 mg/kg) attenuated the memory impairment. Bilateral lesions of the ST did not significantly affect retention of these two tasks. However, ST lesions did block the effects of short-term ADX and dexamethasone administration on memory for both tasks. These results are similar to those of previous experiments examining the effects of lesions of the basolateral nucleus of the amygdala on the glucocorticoid-induced modulation of memory for both tasks. These findings suggest that the integrity of the ST, which connects the amygdala with other brain structures, is essential for the modulating effects of glucocorticoids on memory storage.     

Excitotoxic basolateral amygdala lesions potentiate the memory impairment effect of muscimol injected into the medial septal area

In rats, the septo-hippocampal system is important for memory encoding. Previous reports indicate that muscimol, a specific GABAergic agonist induces learning and memory deficits when infused into the medial septal area. The basolateral nucleus of the amygdala (BLA) modulates memory encoding in other brain areas, including the hippocampus. To explore the interactions between the septo-hippocampal system and amygdala in memory, we studied the effects of intra-medial septal infusions of muscimol in rats with BLA lesions. Animals received sham surgery or excitotoxic BLA lesions and were given infusions of either vehicle or muscimol (5 nmol) into the medial septal area 5 min prior to training sessions in inhibitory avoidance and water maze tasks. In the inhibitory avoidance task, muscimol-induced memory impairment was potentiated by BLA amygdala lesions. Additionally, in the water maze task, BLA-lesioned rats given muscimol infusions into the medial septal also showed memory impairment. These findings indicate that the MSA interacts with the BLA in the processing of memory storage.     

Involvement of a basolateral amygdala complex-nucleus accumbens pathway in glucocorticoid-induced modulation of memory consolidation

Systemic or intracerebral administration of glucocorticoids modulates memory consolidation in several tasks. Previously, we have shown that these memory-modulatory effects depend on an intact basolateral complex of the amygdala (BLC) and efferents from the BLC that run through the stria terminalis. It is currently unknown, however, what BLC efferent structures mediate these effects. The present experiments were designed to determine whether the nucleus accumbens (NA), which receives BLC efferents through the stria terminalis and is involved in several BLC-dependent behaviours, is involved in glucocorticoid-induced modulation of memory consolidation. In experiment 1, rats with bilateral sham or N-methyl-D-aspartate (NMDA)-induced lesions of the NA were trained on a one-trial, footshock-motivated inhibitory avoidance task, and given immediate post-training injections of either the synthetic glucocorticoid dexamethasone (0.3 or 1.0 mg/kg, s.c.) or vehicle. Testing 48 h later revealed that dexamethasone significantly enhanced retention in sham-lesioned rats but that the enhancing effect was blocked in NA-lesioned rats. An asymmetrical, or crossed-lesion design was employed in experiment 2. Rats with a unilateral NMDA-induced lesion of the BLC and a unilateral lesion of either the ipsilateral or contralateral NA were trained as in experiment 1. Testing 48 h later revealed that dexamethasone enhanced retention in ipsilaterally lesioned rats, but that this effect was blocked in contralaterally lesioned rats. These findings indicate that an intact BLC-NA pathway is critical for the enhancing effects of glucocorticoids on memory consolidation, and are consistent with the view that the BLC regulates memory consolidation in other brain regions.     

Systems mediating acute glucocorticoid effects on memory consolidation and retrieval

It is well established that glucocorticoid hormones, secreted by the adrenal cortex after a stressful event, influence cognitive performance. This article reviews recent findings from this laboratory on the acute effects of glucocorticoids in rats on specific memory phases, i.e., memory consolidation and memory retrieval. Posttraining activation of glucocorticoid-sensitive pathways involving glucocorticoid receptors (GRs) enhances memory consolidation in a dose-dependent manner. Glucocorticoid influences on memory consolidation depend on noradrenergic activation of the basolateral complex of the amygdala (BLA) and interactions of the BLA with other brain regions. By contrast, memory retrieval processes are usually impaired with high circulating levels of glucocorticoids or following infusions of GR agonists into the hippocampus. Although the BLA does not appear to be a site of glucocorticoid action in influencing memory retrieval, an intact BLA is required for enabling glucocorticoid effects on memory retrieval. The BLA appears to be a key structure in a memory-modulatory system that regulates, in concert with other brain regions, stress and glucocorticoid effects on both memory consolidation and memory retrieval.     

Differential involvement of hippocampal and amygdalar NMDA receptors in contextual and aversive aspects of inhibitory avoidance memory in rats

Adult male rats bilaterally implanted with guide canullae aimed either at the dorsal hippocampus (dHIP) or the basolateral nucleus of the amygdala (BLA) were trained in a step-down inhibitory avoidance task (IA) and tested for retention 24 h after training. Immediately after training, animals were given a bilateral infusion of the N-methyl-D-aspartate (NMDA) glutamate receptor antagonist D,L-2-amino-5-phosphonopentanoic acid (AP5) (5.0 microg) into the dHIP or the BLA. Both intrahippocampal and intraamygdala infusions of AP5 blocked IA retention. Preexposure to the training box, but not to a different environment 24 h prior to training prevented the impairing effect of intrahippocampal infusion of AP5 on retention. Preexposure did not affect the retention impairment induced by intraamygdala infusion of AP5. These data suggest that hippocampal NMDA receptors might be involved in the contextual and spatial aspects, while amygdalar NMDA receptors might be involved in the aversive aspects of memory for IA.     

Role of norepinephrine in mediating stress hormone regulation of long-term memory storage: a critical involvement of the amygdala.

There is extensive evidence indicating that the noradrenergic system of the amygdala, particularly the basolateral nucleus of the amygdala (BLA), is involved in memory consolidation. Infusions of norepinephrine or beta-adrenoceptor agonists into the BLA enhance memory for inhibitory avoidance as well as water maze training. Other findings show that alpha 1-adrenoceptor activation also enhances memory for inhibitory avoidance training through an interaction with beta-adrenergic mechanisms. The central hypothesis guiding the research reviewed in this chapter is that stress hormones released during emotionally arousing experiences activate noradrenergic mechanisms in the BLA, resulting in enhanced memory for those events. Findings from experiments using rats have shown that the memory-modulatory effects of the adrenocortical stress hormones epinephrine and glucocorticoids are mediated by influences involving activation of beta-adrenoceptors in the BLA. In addition, both behavioral and microdialysis studies have shown that the noradrenergic system of the BLA also mediates the influences of other neuromodulatory systems such as opioid peptidergic and GABAergic systems on memory storage. Other findings indicate that this stress hormone-induced activation of noradrenergic mechanisms in the BLA regulates explicit/declarative memory storage in other brain regions.     

Lesions of the stria terminalis attenuate the amnestic effect of amygdaloid stimulation on avoidance responses

The present study investigated the involvement of two amygdala pathways, the stria terminalis (ST) and the ventral amygdalofugal pathway (VAF), in the effect of post-training electrical stimulation of the amygdala on retention. Rats with implanted amygdaloid electrodes and ST lesions, VAF transections or sham pathway operations, were trained on an inhibitory avoidance task and an active avoidance task. Electrical stimulation of the amygdala was given immediately after training and retention was tested 24 h later. In rats with sham ST lesions, post-training amygdaloid stimulation impaired retention in both tasks. Lesions of the ST did not significantly affect retention in the unstimulated rats. However, the ST lesions attenuated the amnestic effect of amygdaloid stimulation. In rats with sham VAF transections, stimulation of the amygdala impaired retention in the inhibitory avoidance task but enhanced retention in the active avoidance task. Transecting the VAF impaired retention performance of the unstimulated rats in the inhibitory avoidance task. However, the VAF transections did not alter the effect of amygdaloid stimulation: in both tasks, the retention performance of stimulated rats with VAF transections did not differ from that of stimulated rats with sham transections. These findings suggest that the ST may be involved in mediating the influences of the stimulated amygdala in modulating memory storage processing in the brain.     

Impairment of object recognition memory by rapamycin inhibition of mTOR in the amygdala or hippocampus around the time of learning or reactivation

The role of the basolateral complex of the amygdala (BLA) in recognition memory remains poorly understood. The mammalian target of rapamycin (mTOR) in the BLA and other brain areas has been implicated in synaptic plasticity and memory. We have recently shown that mTOR signaling in both the BLA and the dorsal hippocampus (DH) is required for formation and reconsolidation of inhibitory avoidance, a fear-motivated memory task. Here we examined the effects of infusions of the mTOR inhibitor rapamycin into the BLA before or after either training or reactivation on retention of novel object recognition (NOR) memory in rats, and compared the effects with those obtained using intra-DH infusions. Male Wistar rats received bilateral infusions of vehicle or rapamycin into the BLA or DH before or after NOR training or reactivation. Rapamycin impaired NOR retention tested 24 h after training when given either before or immediately after training into the BLA or DH. Rapamycin also impaired retention measured 24 h after reactivation when infused before reactivation into the BLA or DH, or immediately after reactivation into the BLA, but not when given 6 h after reactivation into either the BLA or DH. The results suggest that mTOR signaling in the BLA and DH is involved in NOR memory formation and stabilization.     

Involvement of the amygdala GABAergic system in the modulation of memory storage

These experiments examined the involvement of the intrinsic GABAergic system of the amygdaloid complex in the modulation of memory storage. Rats were chronically implanted with bilateral cannulae in the amygdala, trained in an inhibitory avoidance task, and given post-training bilateral intra-amygdala injections of either the GABA receptor antagonist bicuculline methiodide (BMI) (0.1-1.0 nmol) or the GABAA receptor agonist muscimol (0.001-0.1 nmol). As indicated by performance on a 48 h retention test, BMI enhanced retention of the inhibitory avoidance conditioning, while muscimol impaired retention. The memory-enhancement obtained with BMI (0.1 nmol) was produced by a dose lower than that necessary to induce convulsions. Post-training injections of BMI did not affect retention when injected into the caudate-putamen dorsal to the amygdala. These results suggest that the amygdaloid GABAergic system is involved in the modulation of memory storage.     

Amygdala and dorsal hippocampus lesions block the effects of GABAergic drugs on memory storage

These experiments examined the effects of posttraining systemic administration of the GABAergic agonist muscimol and the GABAergic antagonist bicuculline on retention in mice with bilateral lesions of the amygdala, dorsal hippocampus or caudate nucleus. Unoperated male CD1 mice and mice with either sham lesions or electrolytically induced lesions of these 3 brain regions were trained in a one-trial inhibitory avoidance task and, immediately after training, received i.p. injections of either muscimol, (1.0, 2.0 or 3.0 mg/kg), bicuculline, (0.25, 0.5 or 1.0 mg/kg), or control solutions. Retention was tested 24 h after training. Lesions of the 3 brain regions produced comparable impairment of retention. In the unoperated controls and sham controls muscimol and bicuculline produced dose-dependent impairment and enhancement, respectively, of retention. The drug effects on retention were blocked by lesions of the amygdala and hippocampus, but were not blocked by lesions of the caudate nucleus. These findings are consistent with other recent evidence suggesting that the amygdala and hippocampus are involved in mediating posttraining neuromodulatory influences on memory storage.     

Relaxin receptor activation in the basolateral amygdala impairs memory consolidation

The peptide-hormone relaxin has well-established actions in male and female reproductive tracts, and has functional effects in circumventricular regions of brain involved in neurohormonal secretion. In the current study, we initially mapped the distribution of mRNA encoding the relaxin receptor--leucine-rich repeat-containing G-protein-coupled receptor 7 (LGR7)- and [33P]-human relaxin-binding sites in extra-hypothalamic sites of male Sprague-Dawley rats. The basolateral amygdala (BLA) expressed high levels of LGR7 mRNA and relaxin-binding sites and, although relaxin peptide was not detected in the BLA, several brain regions that send projections to the BLA were found to contain relaxin-expressing neurons. As it is well established that the BLA is involved in regulating the consolidation of memory for emotionally arousing experiences, we investigated whether activation of LGR7 in the BLA modulated memory consolidation for aversively motivated inhibitory avoidance training. Bilateral infusions of human relaxin (10-200 ng in 0.2 microL) into the BLA immediately after inhibitory avoidance training impaired 48-h retention performance in a dose-dependent manner. Delayed infusions of relaxin into the BLA 3 h after training were ineffective, indicating that the retention impairment was due to influences on memory consolidation. Post-training infusions of relaxin into the adjacent central amygdala, which is devoid of LGR7, did not impair retention. These findings suggest a novel function for endogenous relaxin-LGR7 signalling in rat brain involving regulation of memory consolidation.     

Post-training intrabasolateral amygdala infusions of dopamine modulate consolidation of inhibitory avoidance memory: involvement of noradrenergic and cholinergic systems

There is extensive evidence that several neurotransmitter systems within the basolateral amygdala (BLA) influence memory consolidation. The present study investigated the influence of dopamine (DA) in the BLA on the consolidation of memory for inhibitory avoidance (IA) training. Male Sprague-Dawley rats (approximately 300 g) were trained on a step-through IA task and, 48 h later, tested for retention as indexed by their latencies to enter the shock compartment on the test day. Drugs were infused into the BLA or central amygdala nucleus (CEA) immediately or 3 h after training via bilateral cannulae. DA infused into the BLA immediately after training enhanced retention, whereas DA infused into the BLA 3 h after training or into the CEA did not affect retention. Infusions of the dopaminergic antagonist cis-Flupenthixol together with DA blocked the DA-induced memory enhancement. Immediate post-training intra-BLA infusions of the D1 receptor antagonist SCH 23390 or the D2 receptor antagonist sulpiride impaired retention. beta-adrenergic or muscarinic cholinergic receptor antagonists coinfused into the BLA with DA blocked the memory enhancing effects of DA. These findings indicate that dopaminergic activation within the BLA modulates memory consolidation and that the modulation involves activation of both D1 and D2 receptors and concurrent activation of beta-adrenergic and cholinergic influences within the BLA.     

Muscimol injected into the right or left amygdaloid complex differentially affects retention performance following aversively motivated training

The effects of intra-amygdala infusion of the GABA_A agonist, muscimol, prior to retention testing was examined. In Expt. 1, rats were trained in a one-trial inhibitory-avoidance task and given bilateral intra-amygdala infusions of vehicle or muscimol or simultaneous unilateral infusions of each, 5 min before the retention test 24 h after training. Expt. 2 used the same procedures as those in Expt. 1 but two retention measures were taken: initial step-through latency and the number of trials to reach criterion during continuous multiple-trial inhibitory-avoidance (CMIA) training. Groups given bilateral infusions of muscimol or unilateral infusion of muscimol into the right amygdala had significantly shorter latencies to enter the dark compartment than groups given bilateral infusions of vehicle or unilateral infusion of muscimol into the left amygdala. Bilateral muscimol infusions impaired acquisition of CMIA relative to bilateral vehicle infusions or unilateral muscimol infusion into the left amygdala. These results suggest differential involvement of the right and left amygdalae in memory.     

(+/-)-Alpha-methyl-4-carboxyphenylglycine, a metabotropic glutamate receptor blocker, impairs retention of an inhibitory avoidance task in rats when infused into the basolateral nucleus of the amygdala

The amygdala is important for memory processes of emotionally motivated learning and the amygdala glutamatergic system may play a key role in this process. In this study we assessed the effect of the infusion of (+/-)-alpha-methyl-4-carboxyphenylglycine (MCPG), a metabotropic glutamate receptor (mGluR) antagonist, into the basolateral complex of the amygdala (BLA) on the learning and retention of an emotionally motivated task. Rats received either vehicle or three different doses of MCPG (0.2, or 1.0, or 5.0 microg/0.2 microl/side, respectively) bilaterally into the BLA, 5 min before they were trained in a continuous multiple-trial inhibitory avoidance (CMIA) task. Response latencies during the training were recorded. Retention was assessed 8 days later. MCPG in the doses given did not significantly affect the acquisition of the CMIA task. However, MCPG at a dose of 5.0 microg/0.2 microl/side impaired the long-term retention test performance. Additionally, a nociception test indicated that dose of MCPG infused into the BLA did not affect the footshock sensitivity. Our results indicate that MCPG, when infused into the BLA of rats prior to the training, impaired long-term memory of aversive training without affecting acquisition.     

Post-training amygdaloid lesions impair retention of an inhibitory avoidance response

The study examined the effect of pre- and post-training bilateral amygdaloid lesions on retention of a one-trial inhibitory avoidance response. Groups of rats, including unimplanted controls and implanted controls, were trained and tested for retention at 4, 7 or 12 days following training. The lesions were made at one of several intervals before or after training: 2 days before, immediately after, or 2, 5 or 10 days after. At all retention intervals the retention of implanted controls was poorer than that of unimplanted controls and, in comparison with both control groups, the retention of animals lesioned before training was impaired. Retention was also impaired by the post-training lesions. The degree of impairment varied with the interval between the training and the lesion: lesions made within 2 days following training impaired retention, while lesions made 10 days following training had no impairing effect. These findings suggest that posttraining lesions of the amygdala affect retention by impairing time-dependent processes involved in memory storage. With a sufficiently long training-lesion interval (10 days) an intact amygdala is not essential for retention.     

Impaired retention of visual discriminated escape training produced by subseizure amygdala stimulation.

These experiments examined the effects on memory of posttrial, subseizure, electrical stimulation of the amygdala. Rats were trained in a visual discriminated avoidance Y-maze. Each animal received 6 tirals on the training day. Retention, tested the following day, was measured both by the number of correct choices on the first 6 retraining trials and by the number of trials to a criterion of 5 of 6 correct choices. If administered 2 min, 1 h, or 4 h, but not 10 h, after training, bilateral amygdala stimulation significantly impaired retention as measured 24 h after training. In a second experiment, rats received unilateral amygdala stimulation in order to examine better the anatomical localization of effective stimulation sites. The unilateral stimulation was administered either 2 min, 10 min, 1 h, or 4 h after training. The behavioral procedures were the same as those used in the first experiment. For animals stimulated 2 min after training, the optimal stimulation region was one which extended rostrally from the ventrolateral portion of the basomedial nucleus to the dorsomedial region of the amygdala near the stria terminalis and nucleus centralis. For animals stimulated after a 10 min training-treatment interval, this amygdala region was not an effective stimulation site. However, in these animals, stimulation of the basolateral nucleus impaired later retention. Unilateral, posttraining amygdala stimulation administered 1 or 4 h after training did not appear to produce retention deficits. The findings of these experiments thus indicate that posttrial unilateral or bilateral amygdala stimulation impairs retention of discriminated avoidance training. Furthermore, the specific amygdala site at which posttrial stimulation impairs later retention varies with the training-treatment interval.