1 Department of Anesthesiology, Weill Medical College of Cornell University, 525 East 68th Street, New York, NY 10021, USA. 2 Department of Anesthesiology and Critical Care Medicine, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021, USA
* Corresponding author. E-mail: kopryor{at}yahoo.com
Accepted for publication April 28, 2004.
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. Eighty-three healthy volunteers were shown a randomized sequence of 60 visual stimuli consisting of negative, positive and neutral emotive pictures, while receiving a controlled infusion of thiopental (n=31), propofol (n=31), dexmedetomidine (n=10) or placebo (n=11). After a 5 h retention interval, when drug concentration was negligible, subjects performed a recognition task with old pictures randomly mixed with new pictures. Drug effect was calculated as the proportionate reduction in recognition for images of each emotional valence.
Results. Forty-eight subjects were included in a within-subject logistic doseresponse model analysis. In the thiopental group there was a smaller drug effect seen for negative vs positive images (proportional memory reduction from baseline 0.27 (SD 0.20) vs 0.56 (0.25), P<0.001, n=20 included in analysis). A similar trend was seen in the propofol group (0.25 (0.28) vs 0.54 (0.30), n=10), but this did not attain statistical significance. No trend was seen in the dexmedetomidine group (0.33 (0.26) vs 0.24 (0.22), n=7).
Conclusions. Over a specific dose range of thiopental (target serum concentration 27 µg ml1), impairment of explicit memory for images with negative emotional valence is less than that for images with positive emotional valence. There is a strong possibility that propofol (target serum concentration 0.32.4 µg ml1) causes a similar effect. Modulation of visual memory by negative emotional content continues at sub-anaesthetic concentrations of GABAergic drugs associated with explicit memory impairment.
Keywords: anaesthetics, i.v. ; brain, disease ; brain, function ; GABA, brain ; GABA receptors ; memory, brain
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Memory processes that occur in the absence of conscious awareness are termed implicit (unconscious) memory. Implicit and explicit memory processes may have neural substrates that are at least partially independent,12 13 and recent studies strongly suggest that the principal mechanism of fear conditioning is an implicit memory process.14 At the same time, many studies suggest that some implicit memory processes can continue in the presence of clinically relevant concentrations of anaesthetic drugs, even when no explicit memory is evident.15 16 Although it is tempting to infer that any implicit memory behaviour under anaesthesia simply represents a preservation of the same neural processes that mediate implicit memory in the non-drug state, it is also possible that memory formation under anaesthesia is the result of a different or modified processone that only occurs in the presence of drug-induced changes in neuropharmacology and physiology. One method to approach the question of neural mechanisms of memory under anaesthesia is to find evidence that critical structures involved in memory formation continue to function in the presence of anaesthetic drugs. In that context, previous functional neuroimaging studies reveal that medial temporal lobe structures, possibly including the amygdalae, are relatively resistant to the effects of GABAergic drugs.1 4 17 Further, it is possible that some form of memory processing may occur in the amygdala during normal rapid eye movement sleep.18 Thus, an attractive hypothesis is that the amygdala may be centrally involved in memory formation under anaesthesia. There is some pharmacological support for this assertion: recent human studies have demonstrated that limbic structures have a unique distribution of GABA receptor subtypes, most notably a concentration of the 5-subunit.19 20 Although the full implications of these and other differences in receptor subtype distribution are not known, it is certainly possible that such differences could cause regional variability in the pharmacodynamics of GABAergic anaesthetics, thus allowing certain structures to function at drug concentrations that interrupt the function of other structures.
This study addresses that hypothesis by looking for indirect evidence of continued amygdala function in the presence of GABAergic anaesthetic drugs, through its known behavioural actions on modulating memory formation. This study investigates whether memory for negative emotional stimuli is more refractory to the effects of GABAergic drugs, in particular thiopental and propofol, than is memory for positive stimuli. Dexmedetomidine, an 2-adrenoceptor agonist that can cause memory impairment,21 was also investigated. Although dexmedetomidine does not bind directly to the GABAA receptor, it is capable of effecting changes in GABA activity though downstream effects at the ventrolateral preoptic nucleus of the anterior hypothalamus.22 Dexmedetomidine has also recently been shown to suppress fear conditioning in animal studies.23
As negative emotional stimuli are normally better remembered, this study tests whether this relationship is still evident during drug administration. This study assessed memory for visual stimuli of randomly varying emotive content during increasing drug effect. The hypothesis can be stated as follows: as drug-induced memory impairment occurs in a dose-dependent fashionsuch that explicit memory decreases as drug level increasesthe impairment of memory for negative stimuli will be less than the impairment of memory for positive stimuli.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Subjects
Eighty-three healthy right-handed volunteer subjects (51 male, 32 female) aged between 18 and 50 yr were recruited through flyers and newspaper advertisement, and were remunerated for their participation. Studies were conducted between August 2001 and March 2003. Volunteers were initially interviewed extensively by telephone, at which time a detailed medical history was taken. Exclusion criteria included a history of recreational drug abuse, use of psychoactive medications, a history of head trauma resulting in unconsciousness, neurological, cardiovascular, pulmonary or other major disease, weight greater than 30% above ideal body weight, hearing deficit, allergy to propofol or eggs, a family history of acute intermittent porphyria, and pregnancy. All subjects had completed at least a high school education, and were fluent in English, with a normal vocabulary.
Orientation session
Subjects attended an orientation session a minimum of 3 days before the study day. A further detailed medical history was taken and a brief physical examination conducted. Tests of vocabulary (vocabulary subtest of Wechsler Adult Intelligence Scale-Revised24) and handedness (Edinburgh Handedness Inventory) were administered. Detailed information on the protocol procedure was given, and subjects were familiarized with some of the equipment to be used in the study tasks. Written consent for the study was obtained at the end of the orientation session.
Study groups and procedure on study day
Subjects were allocated randomly to receive one of thiopental (n=31), propofol (n=31), dexmedetomidine (n=10) or placebo (n=11), using a random number table. Subjects were then further allocated randomly to receive a low, medium or high dose of drug. The effect-site target concentrations and assignments for each group are shown in Table 1. The concentrations were chosen on the basis of previous observations, such that the medium dose would produce a partial amnesic effect, the high dose a maximal effect, and the low dose a minimal amnesic effect.2 25 26
|
All drugs were administered by computer-controlled infusion using Stanpump software (available from S. Shafer, http://anesthesia.stanford.edu/pkpd/) and syringe-driver infusion pump (Harvard Apparatus, Inc., Holliston, MA, USA). The Stanpump algorithm used a three-compartment model to target a constant serum concentration. Thus, equilibration with the effect site (i.e. brain) took approximately 10 min.
Picture presentation (encoding task)
The picture presentation began when the infusion of drug was started, and finished in 10 min, when predicted pseudo-equilibrium with effect site (brain) had occurred. Thus, the encoding task was conducted during the period when drug concentration in the brain was increasing towards equilibrium, and ended at approximately the point where pseudo-equilibrium was achieved.
Subjects were shown a randomized sequence of 60 coloured pictures from the International Affective Picture System (IAPS, Center for the Study of Emotion and Attention, University of Florida, Gainsville, FL, USA, 1999), consisting of 20 negative, 20 positive, and 20 neutral pictures, based on normative values for emotional valence and arousal. Examples of negative images included a car accident scene and scenes of violence and disease; positive images included happy scenes of children, animals and enjoyable activities; neutral images included household items, inanimate objects and neutral faces. Normative ratings for emotional valence and arousal differed between negative images (mean valence 2.63 (SD 0.72); arousal 5.82 (0.98)), positive images (mean valence 7.52 (0.40); arousal 5.00 (0.83)) and neutral images (mean valence 5.11 (2.10); arousal 4.79 (1.30)). Images were presented using STIM software (Neuroscan, Inc., El Paso, TX, USA) on a 17-inch LCD monitor, at a distance of approximately 1.5 m. Images were presented for 9 s, with a 1 s black screen in-between, for a total test time of 600 s. Subjects were instructed to nominate as quickly as possible whether the picture was more negative or more positive using a binary button press held in the hand. Subject response did not alter the stimulus presentation, and all subjects were shown the same sequence.
Retention interval and recognition test
Following the memory task described, after equilibrium drug levels had been achieved, subjects performed a series of other tasks for separate experiments not reported here. This period lasted approximately 1 h, after which the drug infusion was stopped and the arterial line removed. Subjects were able to eat and drink once appropriate post-anaesthesia criteria had been met. Subjects rested for 4 h, at which time predicted effect-site drug concentration was negligible. The total retention interval was thus approximately 5 h.
For the recognition task, the original 60 images (old) were combined with 60 other images (new) from the IAPS that were matched for valence and intensity. Old and new images were randomly distributed through a presentation sequence of 120 total images, using the same equipment as in the encoding task. Subjects were asked to nominate the image as one of old or new using a binary button press device. Images were presented until a response was registered, at which point the next image was presented; the total time taken to complete the recognition task was thus variable.
Data analysis
Treatment of raw recognition data. As the experimental task was conducted during increasing target-site drug concentration, each successive image presentation occurred at a higher drug dose than the preceding one. Thus, the recognition data were analysed as fitting a doseresponse model, with image recognition as the response variable and trial presentation (time) qualitatively representing increasing drug dose. In the presentation sequence, images were assigned a value of 1 if they were recognized during the recognition task, and 0 if they were not. Initially, a three-parameter logistic curve was fitted to the dichotomous recognition data using SigmaPlot 8.0 Software (SPSS Inc., Chicago, IL, USA) to represent probability of recognition (Precog) as a continuous variable. Curves were fitted to all image valences combined, and then to the valence subgroups of positive, negative and neutral images.
For each subject, the presentation sequence was then divided into three drug effect zones, using the individual logistic curve generated for all image valences combined (Fig. 1). Zone 1 (no significant drug effect) contained the images presented before the point of 80% probability of recognition for stimuli of all valences combined (P80). Zone 1 represented the period when any drug effect was so small that the results were statistically indistinguishable from placebo. Zone 2 (predominantly amnesic drug effect) contained all images following P80 but before the point of 20% probability of recognition (P20), and thus represented the period when there was a definite amnesic drug effect, but some images were still being recognized. P20 was chosen as the point at which true recognition was statistically indistinguishable from false-positive responses to new stimuli. Zone 3 (predominantly sedative drug effect) contained all images presented following P20, and represented the region where no reliable image recognition took place. This very often coincided with the clinical observation that the subject had become extremely sedated during the task. The Precog curve did not always generate all three zones for a given subject. Several subjects never developed sufficient amnesia to generate a Zone 3; others had appreciable drug effect from the outset, and never generated a Zone 1. Still others transitioned rapidly from no amnesia to complete amnesia (often from one image to the next), and thus went directly from Zone 1 to Zone 3.
|
Statistical methodology and analysis. The influence of emotional valence on memory during drug effect was determined by testing for different proportional rates of memory for negative vs positive images compared with baseline (Zone 1) rates. Subjects acted as their own controls in a within-subject design, with emotional valence modelled as the independent variable and amnesic drug effect as the dependent variable. Firstly, the proportion of negative images (Zn1NEG) and positive images (Zn1POS) recalled from Zone 1 was calculated; this formed the baseline for each subject. The proportion of negative images (Zn2NEG) and positive images (Zn2POS) recalled from Zone 2 was then calculated; this represented post-intervention performance during predominantly amnesic drug effect. The relative strength of the effect was calculated as the proportionate reduction in recognition for images of each valence. Thus, for negative images, NEG=[(Zn2NEGZn1NEG)/Zn1NEG], and for positive images,
POS=[(Zn2POSZn1POS)/Zn1POS]. As an example, a subject who recalled 80% of negative images in Zone 1 and 20% of negative images in Zone 2 would have a
NEG=(0.200.80)/0.80=0.75. The values of
NEG and
POS across all subjects were then examined for significant outliers using Grubbs' test for outliers; an
value of <0.01 was considered significant, and outliers were removed from further analysis.
NEG and
POS were compared using a paired-samples t-test, with P<0.05 considered significant. Data analysis was performed using SPSS statistical software. Data are presented as mean (SD) unless otherwise indicated.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
All 83 subjects completed the protocol, and no significant adverse events were reported. Thirty-three subjects were excluded from analysis post hoc, using the criteria described above, as follows: 14 (10 in the propofol group, 4 in the thiopental group) transitioned directly from Zone 1 to Zone 3 and therefore had no Zone 2 for analysis. Four subjects had substantial drug effect from the outset, and had no Zone 1. Ten subjects (6 in the propofol group, 4 in the thiopental group) receiving low doses did not develop sufficient amnesia to transition to Zone 2. A further five subjects were excluded because they had fewer than three stimuli of each valence in each of Zones 1 and 2. Two subjects were excluded following analysis for outliers. Thus, 48 subjects were included in the final analysis (propofol n=10, thiopental n=20, dexmedetomidine n=7, placebo n=11).
Placebo and Zone 1 baseline
The fraction of images recalled by drug, zone and valence are shown in Table 2. In the placebo group, the overall fraction of images recognized was 0.86 (0.10). As expected by design, when compared across all groups by ANOVA, there was no statistically significant difference between placebo and the overall fraction of images recalled in Zone 1 for thiopental (0.93 (0.07)), propofol (0.92 (0.05)) and dexmedetomidine (0.92 (0.06), P=0.08).
|
Zone 2
The overall fraction of images recalled in Zone 2 was similar for thiopental (0.56 (0.12)), propofol (0.55 (0.11)) and dexmedetomidine (0.62 (0.15)). However, there were significant differences in the fraction of images recalled for each valence category. There was a significantly higher fraction of negative vs positive images recalled in the thiopental (0.70 (0.19) vs 0.39 (0.21), P<0.001) and propofol (0.71 (0.24) vs 0.42 (0.26), P=0.048) groups, but in the dexmedetomidine group there was no difference between the valences (0.64 (0.23) vs 0.65 (0.17), P=0.98).
Drug effect by valence
As described above, the relative strength of the drug effect was calculated as the proportionate reduction in recognition for images of each valence. Thus, for negative images, NEG=[(Zn2NEGZn1NEG)/Zn1NEG]; for positive images,
POS=[(Zn2POSZn1POS)/Zn1POS]. In the thiopental group, there was a significantly smaller drug effect seen for negative (
NEG) vs positive (
POS) images (0.27 (0.20) vs 0.56 (0.25), P<0.001). A similar trend was seen in the propofol group (0.25 (0.28) vs 0.54 (0.30), P=0.07), but this did not attain statistical significance using the methods described. In the dexmedetomidine group, there was no significant difference between stimulus classes (negative 0.33 (0.26) vs positive 0.24 (0.22), P=0.46) (Fig. 2).
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
This study used a doseresponse model to evaluate memory impairment as a function of increasing effect-site (brain) drug concentration. Because the question of interest involved the comparison of two response measurements (negative vs positive) taken nearly concurrently, actual overall drug effect was essentially the common denominator, and thus quantitative drug concentration was not that relevant. Predicted effect-site concentrations over Zone 2 were almost exactly the same as previously determined Cp50 values for amnesia for picture stimuli, being 0.8 µg ml1 for propofol and 4.4 µg ml1 for thiopental.25 Advantages to this design include the ability to accommodate a full range of pharmacokinetic/pharmacodynamic responses on a randomized basis, and the ability to use each subject's individual response as their own control. However, the design of this study has important limitations. Because the zones were a function of behavioural response, there was subject-to-subject variation in the number of measurementsand thus detailin each zone. The analysis used proportions rather than absolute numbers, and so the significance of sampling and response error was greater when there was a smaller number of measurements in a given zone, which is the reason for the exclusion of 33 subjects with fewer than three stimuli of each valence in each zone. In the subjects that were analysed, there remained variation in the number of measurements in each zone, and so the contribution of non-systematic error was not evenly distributed between subjects.
The results of this study are consistent with those of several others using visual stimuli that have shown negative emotional valence and intensity of similar visual stimuli to be significant modulators of subsequent recognition memory.2729 This study is the first to investigate the relation between the memory effects of GABAergic agents and emotional valence of stimulus material. As this study is purely behavioural, it cannot provide evidence as to any particular neuroanatomical substrate underlying this behaviour. However, there is a substantial body of pertinent knowledge that allows a reasoned hypothesis regarding the mechanism underlying the observations of this study.
In the current multiple systems model of memory, memory is seen as an interacting network of functional systems traversing multiple anatomical locations, each mediating different forms of memory behaviour. The amygdala complex is an important modulator of memory encoding, storage and consolidation processes, which have been localized to other brain regions.8 Human functional neuroimaging studies clearly demonstrate that the amygdala is prominently involved in modulating memory processes during states of negative emotional arousal.27 29 32 Amygdala activity appears to modulate processes occurring in the hippocampus,33 possibly through long-term potentiation,34 and also the caudate nucleus, possibly through modulation of stimulus-response processes33a finding consistent with the pattern of predominantly glutamatergic neuronal projections from amygdala nuclei found in animal studies.35 The modulatory role of the amygdala is likely to be enabled via the release and action of stress hormones such as epinephrine and cortisol.36 In humans, as in animals, this modulation is enhanced or attenuated by the actions of centrally acting ß-adrenergic agonists37 and antagonists,38 39 respectively.
There has been no systematic investigation of the effects of GABAergic drugs on the amygdala and other limbic structures in humans. Animal literature strongly suggests that the basolateral amygdala complex may be critically involved in mediating the amnesic effects of GABAergic drugs.911 Indeed, removal of the basolateral amygdala complex has been shown to completely block the amnesic effects of propofol9 and diazepam.11 On the other hand, human functional neuroimaging studies have found cortical structures associated with memory processes, notably the dorsolateral prefrontal and parietal cortices, to be specifically depressed by propofol1 and midazolam.4 17 In these studies, medial temporal lobe structures, including the amygdala, are relatively unaffected by these drugs when administered in concentrations producing significant memory impairment. It should be noted that the amygdala is a very small structure, and is difficult to image in functional neuroimaging studies.
Considering evidence from both animal and human studies, two broad possibilities can be formulated. The first is that the basolateral amygdala complex functions in some way as an essential cofactor or gateway for the amnesic effects occurring in cortical structures. The second possibility is that amygdala-mediated memory processes are distinct from those mediated by widely distributed cortical networks. Although this behavioural study provides no data on neural correlates or mechanisms, one intriguing possible explanation for the findings is that the cortically mediated amnesic effect of GABAergic drugs occurs at lower drug concentrations than does any amygdala-mediated amnesic effect. Were this to occur, the modulatory effects of the amygdala on memory would remain robust as an increasing effect of GABAergic drugs caused a failure of cortically mediated memory processes.
Although the number of subjects studied was small, it is interesting that this study found no trend to emotive modulation of memory impairment in subjects who received dexmedetomidine. Dexmedetomidine exerts much of its sedative action through 2A-adrenoceptor binding at the locus ceruleus, which is an important modulator of vigilance40 41 and a principal source of noradrenergic innervation to the forebrain. As the modulatory action of the amygdala is heavily influenced by adrenergic activity, a reduction in noradrenergic output from the locus ceruleus would reasonably be expected to attenuate amygdala-mediated enhancement of memory with negative emotional content. This is corroborated by a recent study demonstrating that a reduction in fear conditioning in mice administered dexmedetomidine was associated with reduced retrieval-associated expression of c-fos and p-CREB in the amygdala.23 Although dexmedetomidine is able to exert changes in GABA activity, these appear to be restricted to activity in the ventrolateral preoptic nucleus in the anterior hypothalamus, via effects downstream from its action at the locus ceruleus.22 There is no evidence that dexmedetomidine exerts indirect GABAergic activity at the amygdala.
The finding of differential memory effects for common GABAergic anaesthetic drugs has important clinical implications. The likelihood that these effects involve the amygdala is especially interesting, given that the peri-operative/peri-anaesthesia period is associated with significant elevations in ß-adrenergic and other endogenous stress hormones. The effects of those hormonal surges on memory behaviour in the peri-anaesthetic setting are not known, but they would be expected to exert significant modulatory actions. It is possible that these effects may be involved in several poorly understood peri-anaesthetic phenomena, such as post-anaesthesia nightmares, dysphoria, depression, anxiety, peri-anaesthesia disinhibition, emotional lability and aggression. An intriguing related area for future study is the effect of ß-adrenergic drugs on some of these peri-anaesthetic phenomena. It remains to be seen whether ß-blockers will modulate memory formation during anaesthesia, whether they might have a place in the prevention of post-anaesthesia psychological/visceral sequelae or whether they can attenuate the development of post-traumatic stress disorder following anaesthesia recall. The establishment of neuroanatomical correlates for the findings presented in the current study will aid these investigations.
In conclusion, this study has demonstrated that, over a specific dose range of thiopental and possibly propofol, impairment of explicit memory occurs at different rates depending on the emotional valence of stimuli, which in this study were visual. Stimuli presented during drug effect with negative emotional valence are more strongly remembered. By inference, this suggests that the processes underlying the modulation of memory by negative emotion may be relatively resistant to the effects of GABAergic drugs. It is possible that these findings represent a differential involvement of cortical and amygdalar memory systems and function. Further investigation is needed to determine such structuralfunctional correlates, using techniques such as functional neuroimaging and event-related potentials. Future behavioural studies should extend the use of the doseresponse model presented here to one in which the effect-site drug concentration is held constant, effectively suspending subjects in each zone while a set number of stimuli are presented.
![]() |
Acknowledgments |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Veselis RA, Reinsel RA, Feshchenko VA. Drug-induced amnesia is a separate phenomenon from sedation: electrophysiologic evidence. Anesthesiology 2001; 95: 896907[ISI][Medline]
3 Ghoneim MM, Hinrichs JV. Drugs, memory and sedation: Specificity of effects. Anesthesiology 1997; 87: 7346[CrossRef][ISI][Medline]
4 Veselis RA, Reinsel RA, Beattie BJ, et al. Midazolam changes cerebral blood flow in discrete brain regions: an H2(15)O positron emission tomography study. Anesthesiology 1997; 87: 110617[CrossRef][ISI][Medline]
5 Baddeley A, Bueno O, Cahill L, et al. The brain decade in debate: I. Neurobiology of learning and memory. Braz J Med Biol Res 2000; 33: 9931002[ISI][Medline]
6 Bright P, Kopelman MD. Learning and memory: recent findings. Curr Opin Neurol 2001; 14: 44955[CrossRef][ISI][Medline]
7 James W. Principles of Psychology. New York: H Holt and Company, 1890
8 Packard MG, Cahill L. Affective modulation of multiple memory systems. Curr Opin Neurobiol 2001; 11: 7526[CrossRef][ISI][Medline]
9 Alkire MT, Vazdarjanova A, Dickinson-Anson H, White NS, Cahill L. Lesions of the basolateral amygdala complex block propofol-induced amnesia for inhibitory avoidance learning in rats. Anesthesiology 2001; 95: 70815[ISI][Medline]
10 Brioni JD, Nagahara AH, McGaugh JL. Involvement of the amygdala GABAergic system in the modulation of memory storage. Brain Res 1989; 487: 10512[CrossRef][ISI][Medline]
11 Tomaz C, Dickinson-Anson H, McGaugh JL. Basolateral amygdala lesions block diazepam-induced anterograde amnesia in an inhibitory avoidance task. Proc Natl Acad Sci USA 1992; 89: 361519[Abstract]
12 Rugg MD, Mark RE, Walla P, Schloerscheidt AM, Birch CS, Allan K. Dissociation of the neural correlates of implicit and explicit memory. Nature 1998; 392: 5958[CrossRef][ISI][Medline]
13 Paller KA, Hutson CA, Miller BB, Boehm SG. Neural manifestations of memory with and without awareness. Neuron 2003; 38: 50716[CrossRef][ISI][Medline]
14 Knight DC, Nguyen HT, Bandettini PA. Expression of conditional fear with and without awareness. Proc Natl Acad Sci USA 2003; 100: 15 2803
15 Jelicic M, Bonke B, Appelboom DK. Indirect memory for words presented during anaesthesia. Lancet 1990; 336: 249
16 Block RI, Ghoneim MM, Sum Ping ST, Ali MA. Human learning during general anaesthesia and surgery. Br J Anaesth 1991; 66: 1708[Abstract]
17 Reinsel RA, Veselis RA, Dnistrian AM, Feshchenko VA, Beattie BJ, Duff MR. Midazolam decreases cerebral blood flow in the left prefrontal cortex in a dose-dependent fashion. Int J Neuropsychopharmacol 2000; 3: 11727[CrossRef][ISI][Medline]
18 Mavanji V, Siwek DF, Patterson EH, Spoley EE, Datta S. Effects of passive-avoidance training on sleep-wake state-specific activity in the basolateral and central nuclei of the amygdala. Behav Neurosci 2003; 117: 7519[CrossRef][ISI][Medline]
19 Maeda J, Suhara T, Kawabe K, et al. Visualization of alpha5 subunit of GABAA/benzodiazepine receptor by 11C Ro15-4513 using positron emission tomography. Synapse 2003; 47: 2008[CrossRef][ISI][Medline]
20 Lingford-Hughes A, Hume SP, Feeney A, et al. Imaging the GABA-benzodiazepine receptor subtype containing the alpha5-subunit in vivo with [11C]Ro15 4513 positron emission tomography. J Cereb Blood Flow Metab 2002; 22: 87889[CrossRef][ISI][Medline]
21 Hall JE, Uhrich TD, Barney JA, Arain SR, Ebert TJ. Sedative, amnesic, and analgesic properties of small-dose dexmedetomidine infusions. Anesth Analg 2000; 90: 699705
22 Nelson LE, Lu J, Guo T, Saper CB, Franks NP, Maze M. The alpha2-adrenoceptor agonist dexmedetomidine converges on an endogenous sleep-promoting pathway to exert its sedative effects. Anesthesiology 2003; 98: 42836[CrossRef][ISI][Medline]
23 Frances Davies M, Tsui J, Flannery JA, Li X, DeLorey TM, Hoffman BB. Activation of alpha(2) adrenergic receptors suppresses fear conditioning: expression of c-fos and phosphorylated CREB in mouse amygdala. Neuropsychopharmacology 2004; 29: 22939[CrossRef][ISI][Medline]
24 Oldfield RC. The assessment and analysis of handedness: The Edinburgh Inventory. Neuropsychologica 1971; 9: 97113[CrossRef]
25 Veselis RA, Reinsel RA, Feshchenko VA, Wronski M. The comparative amnesic effects of midazolam, propofol, thiopental, and fentanyl at equisedative concentrations. Anesthesiology 1997; 87: 74964[ISI][Medline]
26 Veselis RA, Glass P, Dnistrian A, Reinsel R. Performance of computer-assisted continuous infusion at low concentrations of intravenous sedatives. Anesth Analg 1997; 84: 104957[Abstract]
27 Erk S, Kiefer M, Grothe J, Wunderlich AP, Spitzer M, Walter H. Emotional context modulates subsequent memory effect. Neuroimage 2003; 18: 43947[CrossRef][ISI][Medline]
28 Taylor SF, Liberzon I, Fig LM, Decker LR, Minoshima S, Koeppe RA. The effect of emotional content on visual recognition memory: a PET activation study. Neuroimage 1998; 8: 18897[CrossRef][ISI][Medline]
29 Canli T, Zhao Z, Brewer J, Gabrieli JD, Cahill L. Event-related activation in the human amygdala associates with later memory for individual emotional experience. J Neurosci 2000; 20: RC99[Medline]
30 Cahill L, Haier RJ, Fallon J, et al. Amygdala activity at encoding correlated with long-term, free recall of emotional information. Proc Natl Acad Sci USA 1996; 93: 801621
31 Morris JS, Dolan RJ. Involvement of human amygdala and orbitofrontal cortex in hunger-enhanced memory for food stimuli. J Neurosci 2001; 21: 530410
32 Cahill L, Haier RJ, White NS, et al. Sex-related difference in amygdala activity during emotionally influenced memory storage. Neurobiol Learn Mem 2001; 75: 19[CrossRef][ISI][Medline]
33 Packard MG, Cahill L, McGaugh JL. Amygdala modulation of hippocampal-dependent and caudate nucleus-dependent memory processes. Proc Natl Acad Sci USA 1994; 91: 847781[Abstract]
34 Kim JJ, Lee HJ, Han JS, Packard MG. Amygdala is critical for stress-induced modulation of hippocampal long-term potentiation and learning. J Neurosci 2001; 21: 52228
35 Pitkanen A, Pikkarainen M, Nurminen N, Ylinen A. Reciprocal connections between the amygdala and the hippocampal formation, perirhinal cortex, and postrhinal cortex in rat. A review. Ann NY Acad Sci 2000; 911: 36991
36 McGaugh JL, Roozendaal B. Role of adrenal stress hormones in forming lasting memories in the brain. Curr Opin Neurobiol 2002; 12: 20510[CrossRef][ISI][Medline]
37 Cahill L, Alkire MT. Epinephrine enhancement of human memory consolidation: interaction with arousal at encoding. Neurobiol Learn Mem 2003; 79: 1948[CrossRef][ISI][Medline]
38 Cahill L, Prins B, Weber M, McGaugh JL. Beta-adrenergic activation and memory for emotional events. Nature 1994; 371: 7024[CrossRef][ISI][Medline]
39 Cahill L, van Stegeren A. Sex-related impairment of memory for emotional events with beta-adrenergic blockade. Neurobiol Learn Mem 2003; 79: 818[CrossRef][ISI][Medline]
40 Aston-Jones G, Rajkowski J, Kubiak P, Alexinsky T. Locus coeruleus neurons in monkey are selectively activated by attended cues in a vigilance task. J Neurosci 1994; 14: 446780[Abstract]
41 Delagrange P, Canu MH, Rougeul A, Buser P, Bouyer JJ. Effects of locus coeruleus lesions on vigilance and attentive behaviour in cat. Behav Brain Res 1993; 53: 15565[ISI][Medline]