Behavioural Brain Research 191 (2008) 67–71
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Aspects of learned fear related to the hippocampus are sleep-dependent David N. Ruskin ∗ , Gerald J. LaHoste Applied Biopsychology Program, Department of Psychology, University of New Orleans, New Orleans, LA, USA
a r t i c l e
i n f o
Article history: Received 20 December 2007 Received in revised form 22 February 2008 Accepted 10 March 2008 Available online 16 March 2008 Keywords: Conditioned fear Associative learning Pavlovian conditioning Sleep deprivation Sleep restriction Amygdala Hippocampus
a b s t r a c t Reduced sleep interferes with contextual but not cued learned fear, and it was suggested that this selectivity reﬂects underlying neural substrates. The apparent lack of contextual fear in sleep-deprived animals, however, could be secondary to hyperactivity. Also, changing the parameters of cued conditioning can change the neural pathways involved, such that some types of cued fear might be sensitive to sleep loss. To address these issues, we measured fear expressed with conditioned defecation as well as behavior, and used a trace cued learning paradigm. Using the platform-over-water method, male Sprague–Dawley rats were continuously sleep-deprived for 3 days, or for 20 h/day for 3 days. Animals then underwent fear conditioning, and were tested for learning the next day. Sleep-deprived or -restricted animals showed a lack of contextual fear at testing, as conditioned freezing and defecation were minimal. Sleep deprivation also blocked cued fear after trace conditioning. Therefore, reduced sleep impairs contextual learning, and impairs cued learning only when the hippocampus is involved. The data support a model in which sleep loss interferes with hippocampal function while sparing amygdala function. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Deprivation, restriction, or fragmentation of sleep is a common behaviorally based problem in subjects including (but not limited to) students, military personnel, medical personnel, transportation workers, and parents of young children. Disturbed sleep is also present in a number of medical disorders, such as restless-legs syndrome, asthma, sleep apnea, and some psychiatric conditions. Lack of sleep has adverse consequences for society (e.g. vehicle accidents or medical care mistakes) as well as for the affected individual (e.g. injuries or poor health) . Several studies have noted that sustained wakefulness and ethanol intoxication comparably degrade cognition, attention, and motor responses [12,14,23,37]. Learning and memory are some of the cognitive functions impaired by lack of sleep. Different forms of learning and memory, however, are differentially affected by reduced sleep, a pattern demonstrated in experimental animals by fear conditioning. In this paradigm, Pavlovian associations form between a conditioned stimulus and a temporally contingent (typically co-terminating) unconditioned stimulus (footshock), as well as between the learning context and the unconditioned stimulus. Aversive learning is
∗ Corresponding author. Present address: Department of Psychology, Trinity College, 300 Summit Street, Hartford, CT 06106, USA. Tel.: +1 860 297 2337; fax: +1 860 297 2538. E-mail addresses: [email protected]
(D.N. Ruskin), [email protected]
(G.J. LaHoste). 0166-4328/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.bbr.2008.03.011
later expressed, typically by freezing behavior or cessation of operant behavior, during re-exposure to the cue or the context. Brain lesion studies have shown partially different neural substrates for the cued and contextual associations. In particular, damage to the hippocampus degrades contextual learning, while damage to the amygdala degrades predominantly cued learning [28,35,40,42]. To date, studies using either pre- or post-conditioning sleep deprivation (SD) have shown impaired contextual, but not cued, fear conditioning [3,17,27,39] (but see ), suggesting that lack of sleep interferes with normal hippocampal function without affecting the amygdala. Whereas there is abundant evidence in humans and animals that sleep loss interferes with learning and memory, a number of concerns remain regarding the above interpretation of a differential effect of SD on amygdala-dependent and hippocampus-dependent learning in animals. For instance, SD produces hyperactivity in rodents ; therefore, a lack of freezing during a test for conditioned fear might represent an inability to freeze rather than a lack of fear. Whereas the completely intact freezing during presentation of the conditioned stimulus in the cue test in SD animals [3,17,27,39] argues against this interpretation, a hyperactive animal might be able to freeze during the typically brief (30–60 s) conditioned stimulus re-presentation but not during the typically longer contextual fear test. The antagonistic drives for freezing and hyperactivity would thus give the impression of intact cued but impaired contextual learning in SD animals. To address this confound, in the present studies we used conditioned defecation as an index of learned fear in parallel with freezing. Lesion studies
D.N. Ruskin, G.J. LaHoste / Behavioural Brain Research 191 (2008) 67–71
have shown that conditioned defecation to an aversive context is hippocampus-dependent [2,42]. Another issue is the apparent immunity of cued learning to sleep loss. Studies of sleep and cued fear conditioning have used standard “delay” conditioning, in which the cue and the unconditioned stimulus co-terminate and thus overlap in time. Lesion studies have shown that delay fear conditioning is amygdalabut not hippocampus-dependent [21,35]. Separating the cue and unconditioned stimulus by a time interval, a procedure called trace conditioning (involving a memory “trace” of the conditioned stimulus which must persist until the unconditioned stimulus for association to be possible), makes cued fear learning dependent on normal hippocampal function [20,28]. Thus, we characterized the effects of SD on trace conditioning (for comparison with our prior study of delay-conditioning) to test the hypothesis that changing experimental parameters to make cued fear conditioning hippocampus-dependent also makes it susceptible to sleep loss. Aside from parameters of the learning paradigm, another variable that can be modiﬁed in these studies is the type of sleep loss. While continuous extended SD is found in some military personnel and transportation workers, chronic sleep restriction (SR) in which short periods of sleep are possible is a more common pattern in the general population. Recent SR and sleep fragmentation studies in animals have shown suppressed spatial maze learning (which has many parallels to contextual learning) and impaired hippocampal neurogenesis and synaptic plasticity [19,43]. Building on our previous study of continuous SD, we tested for effects of incomplete SD on contextual conditioning. Rats were subjected to 3 days of SR, allowing 4 h of uninterrupted sleep during the normal sleep period (lights on). Similar daily SR in humans leads to significant confusion and reduced vigilance . In the present studies, all SD and SR occurred prior to conditioning. While studies using post-training reduced sleep have been important in relating sleep to learning and memory, particularly for issues of consolidation and retention, chronic pre-training sleep disruption more closely models the typical clinical situation. 2. Materials and methods 2.1. Subjects Male Sprague–Dawley rats (275–300 g) were purchased from Harlan (Indianapolis, IN). Rats were single housed and had ad libitum access to food and water. Animals were treated in accordance with the guidelines established in the “Principles of Laboratory Animal Care” (NIH) and procedures were approved by the University of New Orleans Institutional Animal Care and Use Committee. Lights in the housing and SD/SR rooms came on at 7:00 h and went off at 19:00 h. 2.2. Sleep deprivation/restriction For continuous SD, rats underwent 72 h of the platform-over-water method, as described by Ruskin et al.  (Fig. 1). Brieﬂy, rats remained on a 5-cm diameter platform surrounded by water; if an animal fell asleep it would slip partially or wholly off the platform into the water, waking it. Control rats remained in their home cage. We use the phrase “continuous SD” to refer to an animal being on the platform-overwater continuously, as opposed to discrete sessions on the platform-over-water as in SR (see below). This method for reducing sleep in animals affects paradoxical sleep preferentially over slow-wave sleep, so animals will have some slow-wave sleep (see Section 4); thus, continuous SD in this paradigm is not synonymous with total SD. For SR, we used a schedule that deprived rats of sleep for 20 h per day for 3 days, allowing four uninterrupted hours of sleep in the home cage during the animal’s normal sleep time (mid-day; Fig. 1). Control rats were moved to novel cages for 20 h per day for 3 days, to control for handling and environmental novelty. Thus, rats in all groups were returned to their home cage for 4 h at mid-day. Food and drinking water were available ad libitum in the platform-over-water chambers. A separate group of animals was videotaped during home cage periods of the SR schedule for later visual scoring of behavior. 2.3. Conditioning and testing At the end of the SD or SR schedules, animals were fear conditioned (Fig. 1). SD and control rats underwent a trace conditioning procedure. Brieﬂy, rats were placed
Fig. 1. Schematic of sleep deprivation/restriction and training/testing schedules. The vertical dimension represents time of day (zeitgeber hours indicated at left); days are represented horizontally. Black regions at left indicate lights-off period of each day. Gray areas indicate periods spent on the platform-over-water; at other times, free sleep is possible. (A) Seventy-two hours SD schedule for rats. (B) Threeday SR schedule for rats, with three 20-h no-sleep periods. “trn”: fear training, “con”: contextual memory test, “cue”: cued memory test. “d1”: day 1, “d2”: day 2, etc.
in a 25 cm × 25 cm × 18.5 cm clear Plexiglas chamber, with a metal bar ﬂoor and lid with a built-in speaker (Freeze Monitor, San Diego Instruments, San Diego, CA). After 180 s in the chamber, three tone–footshock pairings were presented. The cue was a 30 s 80 dB tone ending 28 s before onset of shock (2 s, 0.8 mA); a 120-s intertrial period followed. Rats were removed from the chamber 30 s after the last pairing. A second control group of home cage-housed rats underwent a procedure identical except for the absence of footshock (“no shock” group). Twenty-four hours later, rats were returned to the conditioning chamber for 6 min for assessment of contextually conditioned freezing and defecation; no tone or shock was presented during this test. Two hours later, rats were again placed in the conditioning chamber for 6 min for assessment of cue-conditioned freezing. To avoid fear responses induced by the context itself, during this test the chamber was modiﬁed by the insertion of red Plexiglas panels so that the context was novel in terms of the ﬂoor type, chamber shape, and wall color/brightness. The cue was presented once, during the fourth minute. Measurement of freezing behavior (and presentation of stimuli during training) was accomplished with the Freeze Monitor software. Fecal boli were not counted in the cue test: on a time scale of minutes to hours, fecal boli are a limited resource within an animal, such that repeated measurement of defecation becomes problematic. Because of defecation in the context test 2 h prior to the cue test, and treatment group-related differences in contextual defecation (leading to differing amounts of available boli at cue testing), defecation in the cue test was considered to be too confounded to justify quantiﬁcation. SR rats and their controls underwent fear conditioning as described above, except no cue was presented. Three shock presentations occurred at 120-s intervals, except in the “no shock” group. The cue-conditioning test was not performed. All SR and SD animals were allowed to sleep ad libitum in the home cage after conditioning (Fig. 1). In all experiments, the conditioning chamber was cleaned between tests. 2.4. Analysis For all experiments, freezing data were subjected to analysis of variance, whereas defecation data, which were not normally distributed, were subjected to Kruskal–Wallis analysis of variance on ranks.
3. Results In trace conditioning experiments, home-cage control rats showed substantial amounts of freezing and defecation in the context test, whereas 72 h SD rats showed minimal freezing and defecation (Fig. 2). SD rats during the context test did not signiﬁcantly differ from no-shock control rats, which do not have conditioned fear. In the cue test of trace-conditioned rats, freezing was minimal prior to the tone in all groups, but was robust only in home cage control rats in the minute during which the tone sounded (Fig. 3). As in the context test, the freezing scores of SD rats were low during the cue presentation and did not signiﬁcantly differ from those of no-shock control rats. This effect on cued fear is in contrast to the lack of effect found in previous studies using delay, as opposed to trace, conditioning [3,9,17,27,39].
D.N. Ruskin, G.J. LaHoste / Behavioural Brain Research 191 (2008) 67–71
Fig. 4. Behavior during home cage periods during sleep restriction protocol. The measure is the number of minutes without movement per 30-min period during the 4-h free-sleep periods on d2 and d3. Number of subjects is four.
Fig. 2. Sleep deprivation effects on contextual fear. Treatment had a signiﬁcant effect on freezing (F2,21 = 25.8, p < 0.001) and defecation (H2 = 15.6, p < 0.001) in the context test. *p < 0.05, **p < 0.001 vs. home cage group (Newman–Keuls test for freezing; Dunn’s test for defecation). Number of subjects applies to top and bottom panels.
During the 72-h SR schedule there were two 4-h home cage periods (Fig. 1). Visual scoring of movement during these periods showed that activity subsided greatly after the ﬁrst 30 min, and that rats spent the bulk of the remaining time sleeping (Fig. 4). Three days of SR signiﬁcantly reduced both conditioned freezing and defecation in the context test compared to control (Fig. 5). SR rats did not signiﬁcantly differ from no-shock controls: all had minimal freezing
Fig. 5. Sleep restriction effects on contextual fear. Treatment had a signiﬁcant effect on freezing (F2,20 = 15.0, p < 0.001) and defecation (H2 = 9.5, p < 0.01). *p < 0.05, **p < 0.001 vs. control group (Newman–Keuls test for freezing; Dunn’s test for defecation). Number of subjects applies to top and bottom panels.
and defecation. Though not directly compared, SR and SD protocols appeared to be comparably effective in impairing contextual fear (Figs. 2, 5). 4. Discussion Fig. 3. Sleep deprivation effects on cued fear. “m1” and “m4” indicate the ﬁrst and fourth minutes of the test (prior to and during cue presentation, respectively). There were signiﬁcant effects of time (minutes 1 vs. 4; F1,21 = 36.3, p < 0.001) and treatment (F2,21 = 14.8, p < 0.001) on freezing in the cue test, as well as a signiﬁcant interaction (F2,21 = 12.8, p < 0.001). Post hoc Newman–Keuls test are indicated for the interaction. §§ p < 0.001 vs. m1, **p < 0.001 vs. home cage m4.
Reduced sleep prior to training completely blocked learned contextual fear, expressed by two measures. The conditioned defecation measure conﬁrmed that the lack of freezing in SD and SR animals in contextual fear tests was not an artifact of hyperactivity, supporting prior studies of animal SD in which
D.N. Ruskin, G.J. LaHoste / Behavioural Brain Research 191 (2008) 67–71
behavioral expression of aversive [18,38,41,43] or non-aversive memory  did not depend on freezing. Reduced sleep affected cued fear in a learning paradigm known to involve the hippocampus (see below). This pattern of SD/SR effects further supports the concept that reduced sleep degrades hippocampal, but not amygdaloid, function. The negative impact of reduced sleep on the hippocampus is also demonstrated in animal studies of spatial learning in maze tasks, and does not appear to be secondary to a hypothalamic–pituitary–adrenal stress response [38,41]. Recent studies have also shown that SD impairs human spatial memory consolidation [15,32]. Our data do not speak to the sleep stage involved in hippocampal processes. The platform-over-water method should, in theory, prevent only paradoxical sleep, during which muscle tone is lost. In practice, however, slow-wave sleep is also reduced and probably fragmented. With the rat weight and platform size used in the present study, electroencephalographic recordings showed that paradoxical sleep is virtually abolished, but slow-wave sleep is also reduced, by ∼65% . Thus, the SD/SR interference with hippocampal function described here likely involves loss of paradoxical sleep, but we cannot rule out involvement of slow-wave sleep. Cued fear conditioning can be sensitive to SD depending on the speciﬁc protocol. If the conditioned and unconditioned stimulus coterminate (delay-conditioning), then SD does not affect cued fear, as shown by prior studies; if an interval is imposed between conditioned stimulus offset and unconditioned stimulus onset (trace conditioning), SD impairs cued fear (present data). Lesion studies have shown the hippocampal dependence of trace fear conditioning [28,36] consistent with its vulnerability to lack of sleep. Although the present study does not include a delay-conditioned group, the similar methodologies in our prior studies of rat delay-conditioning [27,39] make the comparison reasonable. Our data suggest that the successful formation of cued fear can require sleep but only under conditions known to depend on a functional hippocampus. SD reduces unconditioned freezing to footshock in rats . We found a similar effect during delay-conditioning (three tone–shock pairings) in mice after 24-h SD (unpublished data). This effect seems to involve hyperactivity, and diminished behavioral inhibition in aversive circumstances . Reduced unconditioned freezing during training in a recently sleep-deprived animal, however, cannot be interpreted as indicating a lack of learning during training, as SD animals show normal cue-induced delay-conditioned freezing during later testing (present data; [3,17,27,39]). There is other evidence that conditioned and unconditioned freezing behavior are dissociable [30,31]. Impaired hippocampal-based spatial learning could explain the paradoxical enhancement of maze learning in rats by SD in several early papers [4,5,24]. These studies used water mazes consisting of multiple T-intersections. This type of maze can be successfully completed using either spatial allocentric or procedural (striatal-based) egocentric strategies, and in such a situation spatial and procedural learning (and their neural substrates) compete with one another [8,11,19,33]. An impairment of the hippocampus by SD may allow for augmented maze performance by unopposed procedural learning in a sequential T-maze task. Four hours of free sleep time per day was not sufﬁcient to reinstate normal contextual conditioning. These ﬁndings are in accord with studies of spatial learning using several days of chronic SR or 1 day of sleep fragmentation [19,43]. In rats, one episode of continuous SD as short as 6 h impairs spatial memory . This surprising effect of a seemingly brief epoch without sleep is explained by the normal sleep/wake pattern of small rodents, in which sleep and waking periods are much less consolidated than in humans. In one recent study, laboratory rodents had approximately 70 sleeping bouts during the active (dark) phase of the day, and over 100 waking
bouts during the rest (light) phase . Future experiments will be necessary to characterize SR effects on cued (delay and trace) fear conditioning. Effects of reduced sleep on contextual, trace, and spatial learning relate to underlying effects on hippocampal plasticity. SD or sleep fragmentation interferes with long-term potentiation at hippocampal CA1 and dentate gyrus synapses, in a manner not secondary to hypothalamic–pituitary–adrenal stress [22,27,43]. This effect is caused by reduced synaptic transmission via the N-methylD-aspartate (NMDA) subtype of glutamate receptor: SD changes surface expression and/or subunit composition of NMDA receptors [9,22,26]. It is worth noting that one form of trace cued conditioning (namely, short-interval trace) does not depend on hippocampal NMDA receptors, although it does depend on ongoing hippocampal activity . Signiﬁcant questions remain about how lack of sleep leads to altered hippocampal NMDA receptors, and why the same effect apparently (based on behavioral data) does not occur in the amygdala, a structure in which NMDA-dependent long-term potentiation clearly occurs . Also, it remains to be determined if NMDA receptor alterations cause the cognitive effects of sleep loss related to regions other than the hippocampus and amygdala (such as diminished attention and extinction, related to prefrontal cortex). Apart from these issues, the identiﬁcation of the NMDA receptor as the agent underlying SD-related memory problems raises the possibility of pharmacological reversal. Indeed, bath application of glycine, a requisite NMDA receptor co-agonist, restored normal long-term potentiation in hippocampi from SD rats . This ﬁnding suggests that, in vivo, pharmacological enhancement of NMDA receptor function could prevent or reverse some of the diminished cognition caused by lack of sleep. Acknowledgement Supported by DARPA grant DAAD 19-02-1-0042 (GJL). References  Albert I, Cicala GA, Siegel J. The behavioral effects of REM sleep deprivation in rats. Psychophysiology 1970;6:550–60.  Antoniadis EA, McDonald RJ. Amygdala, hippocampus and discriminative fear conditioning to context. Behav Brain Res 2000;108:1–19.  Bueno OFA, Lobo LL, Oliveira MGM, Gugliano EB, Pomarico AC, Tuﬁk S. Dissociated paradoxical sleep deprivation effects on inhibitory avoidance and conditioned fear. Physiol Behav 1994;56:775–9.  Bunch ME, Cole A, Frerichs J. The inﬂuence of twenty-four hours of wakefulness upon the learning and retention of a maze problem in white rats. Comp Psychol 1937;23:1–11.  Bunch ME, Frerichs JB, Licklider JR. An experimental study of maze learning ability after varying periods of wakefulness. J Comp Psychol 1938;26:499– 514.  Burman MA, Gewurtz JC. Hippocampal activity, but not plasticity, is required for early consolidation of fear conditioning with a short trace interval. Eur J Neurosci 2007;25:2483–90.  Caruso CC. Possible broad impacts of long work hours. Ind Health 2006;44:531–6.  Chang Q, Gold PE. Switching memory systems during learning: changes of patterns in brain acetylcholine release in the hippocampus and striatum in rats. J Neurosci 2003;23:3001–5.  Chen C, Hardy M, Zhang J, LaHoste GJ, Bazan NG. Altered NMDA receptor trafﬁcking contributes to sleep deprivation-induced hippocampal synaptic and cognitive impairments. Biochem Biophys Res Commun 2006;340:435–40.  Dametto M, Suchecki D, Bueno OFA, Moreira KM, Tuﬁk S, Oliveria MGM. Social stress does not interact with paradoxical sleep deprivation-induced memory impairment. Behav Brain Res 2002;129:171–8.  Daniel JM, Lee CD. Estrogen replacement in ovariectomized rats affects strategy selection in the Morris water maze. Neurobiol Learn Mem 2004;82: 142–9.  Dawson D, Reid K. Fatigue, alcohol and performance impairment. Nature 1997;388:235.  Dinges DF, Pack F, Williams K, Gillen KA, Powell JW, Ott GE, et al. Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4–5 h per night. Sleep 1997;20:267–77.
D.N. Ruskin, G.J. LaHoste / Behavioural Brain Research 191 (2008) 67–71  Faletti MG, Maruff P, Collie A, Darby DG, McStephen M. Qualitative similarities in cognitive impairment associated with 24 h of sustained wakefulness and a blood alcohol concentration of 0.05%. J Sleep Res 2003;12:265–74.  Ferrara M, Iaria G, De Gennaro L, Guarigilia C, Curcio G, Tempesta D, et al. The role of sleep in the consolidation of route learning in humans: a behavioural study. Brain Res Bull 2006;71:4–9.  Goosens KA, Maren S. Long-term potentiation as a substrate for memory: evidence from studies of amygdaloid plasticity and Pavlovian fear conditioning. Hippocampus 2002;12:592–9.  Graves LA, Heller EA, Pack AI, Abel T. Sleep deprivation selectively impairs memory consolidation for contextual fear conditioning. Learn Mem 2003;10:168–76.  Guan Z, Peng X, Fang J. Sleep deprivation impairs spatial memory and decreases extracellular signal-regulated kinase phosphorylation in the hippocampus. Brain Res 2004;1018:38–47.  Hairston IS, Little MTM, Scanlon MD, Barakat MT, Palmer TD, Sapolsky RM, et al. Sleep restriction suppresses neurogenesis induced by hippocampusdependent learning. J Neurophysiol 2005;94:4224–33.  Huerta PT, Sun LD, Wilson MA, Tonegawa S. Formation of temporal memory requires NMDA receptors within CA1 pyramidal neurons. Neuron 2000;25:473–80.  Kim JJ, Fanselow MS. Modality-speciﬁc retrograde amnesia of fear. Science 1992;256:675–7. ¨  Kopp C, Longordo F, Nicholson JR, Luthi A. Insufﬁcient sleep reversibly alters bidirectional synaptic plasticity and NMDA receptor function. J Neurosci 2006;26:12456–65.  Lamond N, Dawson D. Quantifying the performance impairment associated with fatigue. J Sleep Res 1999;8:255–62.  Licklider JCR, Bunch ME. Effects of enforced wakefulness upon the growth and the maze-learning performance of white rats. J Comp Psychol 1946;39:339–50.  Martinez-Gonzalez D, Obermeyer W, Fahy JL, Riboh M, Kalin NH, Benca RM. REM sleep deprivation induces changes in coping responses that are not reversed by amphetamine. Sleep 2004;27:609–17.  McDermott CM, Hardy MN, Bazan NG, Magee JC. Sleep deprivation-induced alterations in excitatory synaptic transmission in the CA1 region of the rat hippocampus. J Neurophysiol 2006;570:553–65.  McDermott CM, LaHoste GJ, Davis CP, Chen C, Bazan NC, Magee JC. Sleep deprivation causes behavioral, synaptic, and membrane excitability alterations in hippocampal neurons. J Neurosci 2003;23:9687–95.  McEchron MD, Bouwmeester H, Tseng W, Weiss C, Disterhoft JF. Hippocampectomy disrupts auditory trace fear conditioning and contextual fear conditioning in the rat. Hippocampus 1998;8:638–46.
 Mochizuki T, Crocker A, McCormack S, Yanagisawa M, Sakurai T, Scammell TE. Behavioral state instability in orexin knock-out mice. J Neurosci 2004;24:6291–300.  Oliveira LC, Broiz AC, de Macedo CE, Landeira-Fernandez J, Brand˜ao ML. 5-HT2 receptor mechanisms of the dorsal periaqueductal gray in the conditioned and unconditioned fear in rats. Psychopharmacology 2007;191:253–62.  Oliveira LC, Nobre MJ, Brand˜ao ML, Landeira-Fernandez J. Role of amygdala in conditioned and unconditioned fear generated in the periaqueductal gray. Neuroreport 2004;15:2281–5.  Orban P, Rauchs G, Balteau E, Degueldre C, Luxen A, Maquet P, et al. Sleep after spatial learning promotes covert reorganization of brain activity. Proc Natl Acad Sci USA 2006;103:7124–9.  Packard MG. Glutamate infused post-training into the hippocampus or caudate-putamen differentially strengthens place and response learning. Proc Natl Acad Sci USA 1999;96:12881–6. ¨ R, Tobler I. Sleep depri Palchykova S, Winsky-Sommerer R, Meerlo P, Durr vation impairs object recognition in mice. Neurobiol Learn Mem 2006;85: 263–71.  Phillips RG, LeDoux JE. Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behav Neurosci 1992;106:274–85.  Quinn JJ, Oommen SS, Morrison GE, Fanselow MS. Post-training excitotoxic lesions of the dorsal hippocampus attenuate forward trace, backward trace, and delay fear conditioning in a temporally speciﬁc manner. Hippocampus 2002;12:495–504.  Roehrs T, Burduvali E, Bonahoom A, Drake C, Roth T. Ethanol and sleep loss: a “dose” comparison of impairing effects. Sleep 2003;26:981–5.  Ruskin DN, Dunn KE, Billiot I, Bazan NG, LaHoste GJ. Eliminating the adrenal stress response does not affect sleep deprivation-induced acquisition deﬁcits in the water maze. Life Sci 2006;78:2833–8.  Ruskin DN, Liu C, Dunn KE, Bazan NG, LaHoste GJ. Sleep deprivation impairs hippocampus-mediated contextual learning but not amygdala-mediated cued learning in rats. Eur J Neurosci 2004;19:3121–4.  Selden NRW, Everitt BJ, Jarrard LE, Robbins TW. Complementary roles for the amygdala and hippocampus in aversive conditioning to explicit and contextual cues. Neuroscience 1991;42:335–50.  Smith C, Rose GM. Evidence for a paradoxical sleep window for place learning in the Morris water maze. Physiol Behav 1996;59:93–7.  Sutherland RJ, McDonald RJ. Hippocampus, amygdala, and memory deﬁcits in rats. Behav Brain Res 1990;37:57–79.  Tartar JL, Ward CP, McKenna JT, Thakkar M, Arrigoni E, McCarley RW, et al. Hippocampal synaptic plasticity and spatial learning are impaired in a rat model of sleep fragmentation. Eur J Neurosci 2006;23:2739–48.