Neurobiology of Learning and Memory 109 (2014) 56–61

Contents lists available at ScienceDirect

Neurobiology of Learning and Memory

journal homepage: www.elsevier .com/ locate /ynlme

Exposure to predator odor influences the relative use of multiplememory systems: Role of basolateral amygdala

1074-7427/$ - see front matter � 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.nlm.2013.11.015

⇑ Corresponding author. Address: Department of Psychology, Texas A&MUniversity, College Station, TX 77843, United States. Fax: +1 203 432 7172.

E-mail address: markpackard@tamu.edu (M.G. Packard).

Kah-Chung Leong, Mark G. Packard ⇑Department of Psychology, Texas A&M University, United States

a r t i c l e i n f o

Article history:Received 19 June 2013Revised 31 October 2013Accepted 24 November 2013Available online 10 December 2013

Keywords:AmygdalaMemoryStressLearningAnxietyTMTPredator odor

a b s t r a c t

In a dual-solution plus-maze task in which both hippocampus-dependent place learning and dorsolateralstriatal-dependent response learning provide an adequate solution, the relative use of multiple memorysystems can be influenced by emotional state. Specifically, pre-training peripheral or intra-basolateral(BLA) administration of anxiogenic drugs result in the predominant use of response learning. The presentexperiments were designed to extend these findings by examining whether exposure to a putativelyethologically valid stressor would also produce a predominant use of response learning. In experiment1, adult male Long-Evans rats were exposed to either a predator odor (trimethylthiazoline [TMT], a com-ponent of fox feces) or distilled water prior to training in a dual-solution water plus maze task. On a probetrial 24 h following task acquisition, rats previously exposed to TMT predominantly displayed responselearning relative to control animals. In experiment 2, rats trained on a single-solution plus maze task thatrequired the use of response learning displayed enhanced acquisition following pre-training TMT expo-sure. In experiment 3, rats exposed to TMT or distilled water were trained in the dual-solution task andreceived post-training intra-BLA injections of the sodium channel blocker bupivacaine (1.0% solution,0.5 ll) or saline. Relative to control animals, rats exposed to TMT predominantly displayed responselearning on the probe trial, and this effect was blocked by neural inactivation of the BLA. The findingsindicate that (1) the use of dorsal striatal-dependent habit memory produced by emotional arousal gen-eralizes from anxiogenic drug administration to a putatively ecologically valid stressor (i.e. predatorodor), and (2) the BLA mediates the modulatory effect of exposure to predator odor on the relative useof multiple memory systems.

� 2013 Elsevier Inc. All rights reserved.

1. Introduction

Several neurobehavioral studies employing the use of reversibleand irreversible lesion techniques in animals have demonstrateddouble dissociations between the roles of the hippocampal systemand the dorsal striatum in ‘‘cognitive’’ and stimulus–response ‘‘ha-bit’’ learning and memory, respectively (for reviews see Packard &Knowlton, 2002; White & McDonald, 2002). For example, in a dual-solution plus-maze task that can be acquired using either place orresponse learning, neural inactivation of the hippocampus impairsplace learning, whereas inactivation of the dorsolateral striatumimpairs response learning (Packard & McGaugh, 1996; for reviewsee Packard, 2009a).

In view of evidence that multiple memory systems can be acti-vated in parallel, recent studies have focused on potential factorsthat may influence the relative use of these memory systems in agiven learning situation. One factor that been examined in this

context is the memory modulatory effect of emotional arousal,specifically in the form of stress and/or anxiety. For example, in adual-solution plus-maze task, peripheral pre-training administra-tion of anxiogenic drugs (i.e. noradrenergic alpha-2 receptor antag-onists yohimbine or RS 79948-197) can bias rats towards the use ofdorsolateral striatal-dependent response learning (Packard & Win-gard, 2004). Consistent with evidence that the basolateral amyg-dala (BLA) exerts a memory modulatory influence that is linkedto emotional arousal (for review see McGaugh, 2004), intra-BLAinjections of an anxiogenic dose of RS-79948-197 also produces apredominant use of response learning (Packard & Wingard, 2004;Wingard & Packard, 2008).

Whereas previous research investigated the effects of emotionalarousal on the relative use of multiple memory systems byemploying pharmacological manipulations (i.e. anxiogenic drugtreatment), the present experiments were designed to extendthese findings by employing exposure to predator odor, a puta-tively ethologically valid stressor. It is well established that expo-sure to predator odor induces stress and anxiety in rodents(Griffith, 1919, 1920). Rats that have been exposed to cat fur/odorexhibit a range of fear/anxiety-like avoidance behaviors such as

K.-C. Leong, M.G. Packard / Neurobiology of Learning and Memory 109 (2014) 56–61 57

freezing, hiding, and decreased stimulus contact (Blanchard,Blanchard, Wiess, & Meyers, 1990; Dielenberg & McGregor,2001). In addition, exposure to 2,5-dihydro-2,4,5-trimethylthiazo-line (TMT), a sulfur-containing compound that is specific to red foxfeces, also induces fear/anxiety in laboratory rats (Burwash, Tobin,Woolhouse, & Sullivan, 1998; Morrow, Elsworth, & Roth, 2002;Vernet-Maury, 1980; Vernet-Maury, Polak, & Demael, 1984; forreview see Fendt, Endres, Lowry, Apfelbach, & McGregor, 2005).Several studies have examined the effect of exposure to TMT onperformance of various learning and memory tasks (for reviewsee Takahashi, Nakashima, Hong, & Watanabe, 2005) and TMThas been used extensively as an unconditioned stimulus (US) infear conditioning tasks (for review see Takahashi, Chan, & Pilar,2008).

In experiment 1 we examined the effect of pre-training expo-sure to TMT on the relative use of ‘‘place’’ and ‘‘response’’ learningin a dual-solution water plus-maze. In this task rats are trained toswim from the same start arm to a hidden escape platform that isalways located in the same goal arm. This dual-solution task can beacquired by learning to make the same body turn response at thechoice point (i.e. response learning), or alternatively by learningthe spatial location of the escape platform (i.e. place learning).The relative use of these strategies can be assessed after acquisi-tion on a probe trial in which rats are started from the oppositestart arm. On the probe trial, rats that swim to the same spatiallocation in which the platform was located during training aredesignated ‘‘place’’ learners, whereas rats that make the body turnresponse reinforced during training are designated ‘‘response’’learners’’. Response learning in the plus-maze is dorsolateral stria-tal-dependent, whereas place learning is hippocampus-dependent(Packard, 1999; Packard & McGaugh, 1996; for review see Packard,2009b).

In experiment 2 we investigated the effect of pre-training TMTexposure on acquisition of a dorsolateral striatal-dependent single-solution plus-maze task that required the use of response learning.In this task, rats are started from varying start arms (North, South)and are trained to always make the same body turn response (e.g.turn left) at the maze choice point. Finally, in experiment 3 weexamined the effect of post-training neural inactivation of theBLA on the ability of TMT exposure to influence the relative useof place and response learning.

2. General methods

2.1. Subjects

Subjects (n = 98) were experimentally naïve adult male CharlesRiver Long-Evans rats (weighing 300–400 g). Animals were housedindividually in a climate-controlled vivarium with ad libitum accessto food and water. Experiments were conducted during the lightphase cycle of a 12:12 h light–dark cycle (lights on at 7 a.m.).

2.2. Apparatus

The water plus-maze apparatus used was identical to that usedin our previous studies (e.g. Leong, Goodman, & Packard, 2012;Packard & Gabriele, 2009). A clear Plexiglas plus-maze (43.2 cmin height, 26 cm in arm-width, and 59.1 cm in length) was insertedinto a black circular water maze (1.73 m in diameter, 45 cm inheight). The maze was filled with water (25 �C) to a depth ofapproximately 21 cm. A submerged Plexiglas escape platform(14 � 14 � 20 cm) was located at the end of a maze arm that varieddepending on the specific training protocol. The PVC holding con-tainers used for either predator odor or distilled water exposure(45 � 30 � 25 cm) was placed underneath a ventilation hood.

2.3. Odorant exposure and drugs injections

The method of exposure to the predator odor 2,3,5-trimethyl-3-thiazoline (TMT) was similar to previous studies (Endres & Fendt,2008; Fendt, Endres, & Apfelbach, 2003). Consistent with severalprevious studies examining the behavioral effects of TMT exposure(e.g. Galliot, Levaillant, Beard, Millot, & Pourie, 2010; Hacquemand,Jacquot, & Brand, 2012; Morrow, Roth, & Elsworth, 2000) distilledwater was used for the control group. TMT (5 ll) or distilled water(5 ll) was deposited onto circular filter paper (4.7 cm diameter)and placed on the wall of the holding container 10 cm from thebottom. Rats were placed into the appropriate TMT or control (dis-tilled water) container for 5 min immediately prior to training.

In experiment 3, bilateral intra-BLA infusions (0.5 ll/side) ofbupivicaine (1% solution, Abbott Laboratories) or saline wereadministered via a microsyringe pump with an electronic timer(Sage Instruments) through 10 ll Hamilton syringes connected toan polyethylene tubing (PE 10) and injection needle (16 mmlength, 30 gauge). Bupivicaine acts as a sodium channel blocker,hence providing temporary inactivation of the region via the block-ade of action potential conductance. Infusions were administeredover a period of 52 s. Following this period, injection needles wereleft in the guide cannula for an additional 60 s to allow fordiffusion.

2.4. Surgery

Prior to surgery rats were anesthetized with vapor isofluraneand then implanted with bilateral guide cannulae using standardstereotaxic procedure. The guide cannulae (23 gauge, 15 mm inlength) were inserted into the brain overlying the BLA and heldin place using jeweler’s screws and dental acrylic. The stereotaxiccoordinates for the BLA guide cannulae placements wereAP = �2.2 mm, ML = ±4.7 mm, DV = �7.0 mm. These coordinatesare the same as our previous studies that have examined the roleof intra-BLA drug infusions on plus-maze behavior (e.g. Packard& Gabriele, 2009; Packard & Wingard, 2004; Wingard & Packard,2008). Following surgery animals were given 8 days of post-oper-ative recovery prior to beginning behavioral training.

2.5. Histology

Following the completion of behavioral procedures rats weresacrificed with a 1 ml injection of pentobarbital sodium and phe-nytoin sodium (Euthasol Euthanasia Solution, Virbac Corporation,Texas). Rats were then perfused with physiological saline followedby 10% formaldehyde-saline solution. Brains were removed andsectioned at 20 lm through the cannula tract region using a cryo-stat, and were subsequently mounted on slides and stained withcresyl violet. The location of the injection needle tips were con-firmed using a standard rat brain atlas (Paxinos & Watson, 1997),and were located in the basolateral amygdala ranging from�1.80 to �3.14 mm from bregma (Fig. 1). Five animals were re-moved from the data analyses due to cannula placements thatwere located outside of the BLA.

It should be noted that although the injection needle tips werelocated in the BLA, the possibility of drug spread into other amyg-dala nuclei (e.g. central nucleus) cannot be completely ruled out.Nonetheless, converging evidence suggests that the memory mod-ulatory effects of emotional arousal are mediated by the BLA. Thus,lesions of the BLA, but not the central nucleus block amygdalainfluences on hippocampus-dependent memory (Roozendaal &McGaugh, 1996). In addition, post-training drug administrationmodulates memory when infused into the BLA, but not the centralnucleus (Roozendaal & McGaugh, 1997). Finally, the influence ofthe amygdala on hippocampal long-term potentiation is mediated

Fig. 1. Illustrated brain sections from rats indicating the location of infusion needleplacements (filled circles shown with overlap) in the BLA ranging from �1.80 mmto �3.14 mm anterior–posterior from bregma (unlabelled diagrams from Atlas ofPaxinos & Watson, 1997).

58 K.-C. Leong, M.G. Packard / Neurobiology of Learning and Memory 109 (2014) 56–61

by the BLA, and not the central nucleus (Akirav & Richter-Levin,2002).

2.6. Behavioral procedures

2.6.1. Experiment 1Immediately following pre-training exposure to TMT (n = 15) or

distilled water (n = 15) rats were transported to the behavioraltesting room. Animals were trained in a dual-solution water

plus-maze task for 2 consecutive days (6 trials/day). On each trial,animals were placed into the start-arm of the maze (i.e. southarm), facing the maze wall and were given 60 s to swim to a hiddenplatform located in the goal-arm (i.e. east arm). The start-arm andgoal-arm remained fixed throughout the training period. The oppo-site arm from the start-arm was blocked off with a Plexiglas bar-rier. After reaching the platform rats remained there for 10 sbefore being removed and placed in an adjacent opaque holdingcontainer for a 30 s inter-trial interval. If the rat made a full bodyturn into the incorrect arm (i.e. west arm) the trial was scored asan error. Following training, rats received a probe trial on the thirdday to assess their relative use of place and response learning. Noexposure to predator odor was given prior to the probe trial. On theprobe trial, the start-arm was shifted to the opposite arm (i.e. northarm), with the arm directly opposite blocked off (i.e. south arm).Rats that turned left at the choice point and entered the east armon the probe trial (i.e. approached the same spatial location thatthe hidden platform was located in during training) were desig-nated as place learners. Rats that made a right turn at the choicepoint and entered the west arm on the probe trial (i.e. made thesame body turn to swim to the hidden platform as during training)were designated as response learners.

2.6.2. Experiment 2Immediately following pre-training exposure to TMT (n = 11) or

distilled water (n = 11) (days 1 through 3) rats were transported tothe behavioral testing room. Animals were then trained in a single-solution response water plus-maze task similar to previous studiesfrom our lab (Leong et al., 2012). In this task rats were trained for 5consecutive days (6 trials/day). On each trial, rats were placed inthe start arm (north or south) facing the maze wall and were given60 s to swim to a hidden escape platform located in another arm(east or west). The sequence of the start arm varied dependingon day. On odd days, the start arm sequence was NSSNNS and oneven days, the start arm sequence was SNNSSN. The escape plat-form was always placed in the arm in which a right body turn atthe maze choice point led to the platform. If the rat failed to findthe escape platform in 60 s, the experimenter manually guidedthe rat to the platform. Upon climbing onto the platform, rats re-mained there for 10 s before being removed from the maze andplaced in an opaque holding container for a 30 s inter-trial interval.On each trial, a correct response was scored if the rat made a fullbody turn into the correct arm in which the escape platform waslocated. A full body turn into the wrong arm resulted in the trialbeing scored as an error.

2.6.3. Experiment 3Immediately following exposure to TMT (n = 29) or distilled

water (n = 12) on days 1 and 2, rats were transported to the behav-ioral testing room. Rats that received pre-training TMT exposurewere divided into two additional groups, receiving post-trainingintra-BLA infusions of either bupivicaine (n = 16) or vehicle saline(n = 13). Rats exposed to pre-training distilled water receivedpost-training intra-BLA infusions of saline (n = 12). Importantly,we have previously demonstrated that intra-BLA infusion of bupiv-icaine does not itself effect acquisition of either place or responselearning in the water plus-maze (Packard & Gabriele, 2009).

Experiment 3 replicated the behavioral procedure used for thedual-solution water maze task in experiment 1 with the exceptionthat animals were trained for 4 trials/day (a sufficient number oftrials to produce learning). Thus, rats were trained to swim fromthe same start arm to the same goal arm on all trials (4 trials/day) on days 1 and 2. Immediately following training on days 1and 2 rats received post-training intra-BLA infusions of either bup-ivicaine (0.5 ll/side) or saline. Rats were then given a probe trial on

100

TMT

K.-C. Leong, M.G. Packard / Neurobiology of Learning and Memory 109 (2014) 56–61 59

day 3 and were designated as either ‘‘place’’ or ‘‘response’’ learnersbased on their probe trial behavior.

0 1 2 3 4 525

50

75 dH2O

Day

% C

orre

ct

Fig. 3. Enhancing effects of pre-training exposure to TMT on acquisition of responselearning in the forced-response water plus-maze task.

3. Results

3.1. Experiment 1

3.1.1. TMT exposure biases rats towards response learning in a dualsolution plus-maze task

A two-way one-repeated measures ANOVA (Group � Day) re-vealed a main effect of Day (F(1,28) = 8.29, p < 0.05), indicating thatanimals in both groups learned the task over two days. There wasno significant difference between groups (F(1,28) = 0.25, n.s.), indi-cating that there was no effect of pre-training exposure to TMT ontask acquisition.

The effect of pre-training exposure to TMT on the use of ‘‘place’’or ‘‘response’’ learning on the subsequent day 3 probe trial isshown in Fig. 1. A v2 analysis was performed to determine if therewas a difference in the relative use of place or response learning onthe probe trial. Rats exposed to distilled water displayed an abso-lute preference for the use of place learning on the probe trial (11place rats, 4 response rats with the analysis showing a significanttrend (v2 = 3.27, p = 0.07). In sharp contrast, rats exposed to TMTpre-training displayed a significant use of response learning strat-egy on the probe trial (3 place rats, 12 response rats; v2 = 5.4,p < 0.05). These findings indicate that in the dual-solution waterplus-maze task in which place and response learning can both pro-vide an adequate solution, pre-training exposure to TMT influ-enced the type of strategy adopted by rats during the probe trial.Specifically, relative to control rats, rats that had received pre-training TMT exposure were biased towards the use of responselearning (Fig. 2).

3.2. Experiment 2

3.2.1. Exposure to TMT enhances response learning in a single-solutionplus maze task

The effect of pre-training exposure to TMT on acquisition of thesingle-solution response learning task is shown in Fig. 2. A two-way repeated measures ANOVA (Group � Day) computed on per-centage correct on days 1–5 revealed a main effect of Group,(F(1,20) = 5.83, p < 0.05), and of Day (F(4,80) = 12.33, p < 0.01).There was no significant Group � Day interaction (F(4,80) = 0.13,n.s.). These results indicate that rats from both groups showed sig-nificant improvement in response learning over the 5 day trainingperiod, and relative to rats that received pre-training exposure todistilled water, rats that received pre-training exposure to TMTdisplayed facilitated task acquisition (Fig. 3).

TMT dH2O0

2

4

6

8

10

12 ResponsePlace

Drug Group

Num

ber o

f rat

s

Fig. 2. Number of rats in each experimental group that exhibited place or responselearning on the day 3 probe trial. Rats received pre-training exposure to TMT ordistilled water.

3.3. Experiment 3

3.3.1. Intra-BLA bupivicaine infusions block the TMT-induced biastowards response learning in a dual solution plus-maze task

Similar to experiment 1, a two-way one-repeated measures AN-OVA (Group � Day) revealed a main effect of Day (F(1,35) = 23.53,p < 0.01), indicating all groups of animals learned the task over twodays. Additionally, there was no significant difference betweengroups (F(1,35) = 0.40, n.s.), indicating that there was no effect ofpre-training exposure to TMT on learning of this task. The effectof post-training intra-BLA infusions of bupivicaine on the abilityof TMT exposure to influence the relative use of place and responselearning on the day 3 probe trial is shown in Fig. 3. v2 analyses re-vealed that control rats exposed to distilled water prior to trainingand receiving post-training intra-BLA infusions of saline displayeda significant trend towards the predominant use of place learningon the probe trial, (9 place rats, 3 response rats; v2 = 3.00,p = 0.08). In contrast, v2 analysis revealed that rats exposed toTMT prior to training and receiving post-training intra-BLA infu-sions of vehicle saline displayed a significant use of response learn-ing on the day 3 probe trial (2 place rats, 11 response rats;v2 = 6.23, p < 0.05). This finding replicates the bias towards theuse of response learning that was produced by pre-training TMTexposure in experiment 1. However, when bupivicaine was infusedpost-training into the BLA immediately after TMT exposure, therewas no significant difference in the type of learning strategy usedon the subsequent day 3 probe trial (7 place rats, 6 response rats;v2 = 1.00, n.s.). Taken together, these findings indicate that pre-training TMT exposure biases rats towards the use of responselearning on a subsequent probe trial, and that intra-BLA infusionsof bupivicaine block this effect (Fig. 4).

TMT-SAL TMT-BP dH2O-SAL0123456789

1011

ResponsePlace

Group

No.

of R

ats

Fig. 4. Number of rats in each experimental group that exhibited place or responselearning on the day 3 probe trial. Rats received pre-training exposure to TMT ordistilled water and received either post-training injections of bupivacaine or salineinto BLA.

60 K.-C. Leong, M.G. Packard / Neurobiology of Learning and Memory 109 (2014) 56–61

4. Discussion

The present findings indicate that in a dual-solution plus-mazetask that can be acquired using both place and response learning,pre-training exposure to the predator odor 5-dihydro-2,4,5-trim-ethylthiazoline (TMT) biases rats towards the use of responselearning. In addition, in a single solution plus-maze task that re-quires the use of response learning, pre-training exposure to TMTenhances task acquisition. Extensive evidence indicates that expo-sure to TMT is negatively emotionally arousing to animals, induc-ing anxiety/fear-like effects that have been assessed via a variety ofbehavioral measures including freezing, defecation, and approachlatency (e.g. Vernet-Maury et al., 1984; Burwash et al., 1998; Hot-senpillar & Williams, 1997; Morrow et al., 2002; for review seeFendt et al., 2005). Thus, as has been previously observed followingdrug-induced anxiety (e.g. Elliott & Packard, 2008; Packard &Gabriele, 2009; Packard & Wingard, 2004), exposure to a putativelyecologically valid stressor (i.e. predator odor) can also influence therelative use of multiple memory systems. Overall, the findings pro-vide further evidence supporting the hypothesis that robust emo-tional arousal induced by stress/anxiety leads to the facilitationand preferential use of habit memory (for reviews see Packard,2009b; Packard & Goodman, 2012).

The precise neural basis of TMT-induced anxiety/stress has yetto be fully established, although several studies suggest a potentialrole for the amygdaloid complex. For example, temporary inactiva-tion of the BLA and medial amygdala blocks TMT-induced freezingbehavior (Muller & Fendt, 2006). Exposure to TMT increases activ-ity in the bed nucleus of the stria terminalis as well as the centralnucleus of the amygdala (Day, Masini, & Campeau, 2004), provid-ing further support for the role of the amygdaloid complex inTMT-induced anxiety. Although electrolytic lesions of the lateralamygdala (Wallace & Rosen, 2001) blocked TMT-induced freezingbehavior, this effect was not observed following cell-body sparingneurotoxic lesions of this area (Pagani & Rosen, 2009; Wallace &Rosen, 2001).

We have previously observed that the BLA mediates the abilityof the anxiogenic drug RS 79948-197 to bias rats towards the useof response learning in a dual-solution plus-maze task, as well asfacilitate response learning in a single-solution plus-maze task(Packard & Gabriele, 2009; Packard & Wingard, 2004; Wingard &Packard, 2008). These previous plus-maze findings are consistentwith extensive evidence implicating the BLA in mediating themodulatory effects of emotional arousal on memory (for reviewsee McGaugh, 2004). In the present study (experiment 3), neuralinactivation of the BLA also prevented the bias towards responselearning that is produced by pre-training TMT exposure in thedual-solution plus-maze task. Taken together, these findings sug-gest that TMT exposure and anxiogenic drug administration influ-ence the relative use of place and response learning via a commonmechanism that involves the BLA.

It is important to note that although the BLA mediates thememory modulatory influence of anxiogenic drug administrationand TMT exposure on plus-maze behavior, the functional integrityof this brain region is not necessary for the acquisition of eitherplace or response learning (Packard & Gabriele, 2009). Rather, thememory modulatory role of the BLA appears to involve activationof efferent projections that modulate memory processes occurringin other brain structures (Packard, Cahill, & McGaugh, 1994;McGaugh, 2004). In the case of place and response learning inthe plus-maze, evidence indicates that these two types of learningare mediated by the hippocampus and dorsolateral striatum,respectively (e.g. Packard & McGaugh, 1996; for review seePackard, 2009a). One possibility is that TMT exposure activatesBLA efferents and biases rats towards the use of dorsolateral

striatal-dependent response learning by directly influencing synap-tic plasticity within the striatum. Consistent with this hypothesis,pre-training predator exposure increases c-fos mRNA expressionin the dorsolateral striatum in rats that display a procedural learn-ing strategy in a water radial-arm maze task (VanElzakker et al.,2011). Alternatively, TMT exposure may indirectly favor the useof dorsolateral striatal-dependent response learning by impairinghippocampus-dependent place learning. Consistent with the latterhypothesis, both anxiogenic drug administration (Packard & Gabri-ele, 2009; Wingard & Packard, 2008) and exposure to a predator(presence of a cat), can impair hippocampus-dependent spatiallearning (Park, Campbell, & Diamond, 2002; Park, Zoladz, Conrad,Fleshner, & Diamond, 2008; but see also Diamond, Campbell, Park,Halonen, & Zoladz, 2007; Galliot et al., 2010) and retrieval (Dia-mond et al., 2006). Moreover, this stress-induced impairment ofhippocampus-dependent memory appears to involve a modulatoryinfluence of the amygdala. Thus, pre-training exposure to a stressregimen (restraint and tail-shock) impairs hippocampus-depen-dent learning in a spatial water maze task, and this effect isblocked by lesions of the amygdala (Kim, Lee, Han, & Packard,2001). Consistent with the present findings using TMT exposure,a pre-training stress regimen also enhanced the relative use of dor-solateral striatal-dependent memory in a dual-solution task (Kimet al., 2001). Finally, it has been suggested that the impairmentof hippocampus-dependent spatial memory following pre-trainingpredator exposure is mediated by stress-induced increases inphosphorylated CaMKII within the BLA (Zoladz et al., 2012).

In sum, our findings indicate that pre-training exposure to thepredator odor TMT biases animals towards the use of responselearning/habit memory in a dual-solution plus-maze task, as wellas facilitates acquisition of response learning in a single-solutionplus-maze task. This effect of TMT appears to be mediated, at leastin part, by the BLA, providing further evidence of a role for thisbrain region in emotional modulation of the relative use of multi-ple memory systems. An anxiety/stress induced bias towards theuse of habit memory has also been recently observed in humanstudies (e.g. Schwabe & Wolf, 2009; Schwabe et al., 2007), and ithas been suggested that this modulatory influence of emotionalarousal on the relative use of multiple memory systems may haveimplications for understanding the role of learning and memoryprocesses in several human psychopathologies (for reviews seeGoodman, Leong, & Packard, 2012; Packard, 2009b; Schwabe, Dick-inson, & Wolf, 2011).

Acknowledgment

Research supported by National Science Foundation GrantIBN-03122212 (M.P.).

References

Akirav, I., & Richter-Levin, G. (2002). Mechanisms of amygdala modulation ofhippocampal plasticity. Journal of Neuroscience, 22, 9912–9921.

Blanchard, R. J., Blanchard, D. C., Wiess, S. M., & Meyers, S. (1990). The effects ofethanol and diazepam on reactions to predatory odors. PharmacologyBiochemistry and Behavior, 35, 775–780.

Burwash, M. D., Tobin, M. E., Woolhouse, A. D., & Sullivan, T. P. (1998). Laboratoryevaluation of predator odors for eliciting an avoidance response in roof rats(Rattus rattus). Journal of Chemical Ecology, 24, 49–66.

Day, H. E., Masini, C. V., & Campeau, S. (2004). The pattern of brain c-fos mRNAinduced by a component of fox odor, 2,5-dihydro-2,4,5-trimethylthiazoline(TMT), in rats, suggests both systemic and processive stress characteristics.Brain Research, 1025, 139–151.

Diamond, D. M., Campbell, A. M., Park, C. R., Halonen, J., & Zoladz, P. R. (2007). Thetemporal dynamics model of emotional memory processing: A synthesis on theneurobiological basis of stress-induces amnesia, flashbulb and traumaticmemories, and the Yerkes–Dodson Law. Neural Plasticity, 1–33.

Diamond, D. M., Campbell, A. M., Park, C. R., Woodson, J. C., Conrad, C. D.,Bachstetter, A. D., & Mervis, R. F. (2006). Influence of predator stress on the

K.-C. Leong, M.G. Packard / Neurobiology of Learning and Memory 109 (2014) 56–61 61

consolidation versus retrieval of long-term spatial memory and hippocampalspinogenesis. Hippocampus, 16, 571–576.

Dielenberg, R. A., & McGregor, I. S. (2001). Defensive behavior in rats towardspredatory odors: A review. Neuroscience & Biobehavioral Reviews, 25, 597–609.

Elliott, A. E., & Packard, M. G. (2008). Intra-amygdala anxiogenic drug infusion priorto retrieval biases rats towards the use of habit memory. Neurobiology ofLearning and Memory, 90, 616–623.

Endres, T., & Fendt, M. (2008). Inactivation of the lateral septum blocks fox odor-induced fear behavior. NeuroReport, 19, 667–670.

Fendt, M., Endres, T., & Apfelbach, R. (2003). Temporary inactivation of the bednucleus of the stria terminalis but not of the amygdala blocks freezing inducedby trimethylthiazoline, a component of fox feces. Journal of Neuroscience, 23,23–28.

Fendt, M., Endres, T., Lowry, C. A., Apfelbach, R., & McGregor, I. S. (2005). TMT-induced autonomic and behavioral changes and the neural basis of itsprocessing. Neuroscience & Biobehavioral Reviews, 29, 1145–1156.

Galliot, E., Levaillant, M., Beard, E., Millot, J.-L., & Pourie, G. (2010). Enhancement ofspatial learning by predator odor in mice. Involvement of amygdala andhippocampus. Neurobiology of Learning and Memory, 93, 196–202.

Goodman, J., Leong, K. C., & Packard, M. G. (2012). Emotional modulation of multiplememory systems: Implications for the neurobiology of post-traumatic stressdisorder. Reviews in Neurosciences, 23, 627–643.

Griffith, C. R. (1919). A possible case of instinctive behavior in the white rat. Science,50, 166–167.

Griffith, C. R. (1920). The behavior of white rats in the presence of cats.Psychobiology, 2, 19–28.

Hacquemand, R., Jacquot, L., & Brand, G. (2012). Postnatal exposure to predator odor(TMT) enhances spatial learning in mice adulthood. Behavioural Brain Research,229, 113–117.

Hotsenpillar, G., & Williams, J. L. (1997). A synthetic predator odour (TMT) enhancesconditioned analgesia and fear when paired with a benzodiazepine receptorinverse agonist (FG-7142). Psychobiology, 25, 83–88.

Kim, J. J., Lee, H. J., Han, J. S., & Packard, M. G. (2001). Amygdala is critical for stress-induced modulation of hippocampal long-term potentiation and learning. TheJournal of Neuroscience, 21, 5222–5228.

Leong, K. C., Goodman, J., & Packard, M. G. (2012). Buspirone blocks the enhancingeffect of the anxiogenic drug RS 79948–197 on consolidation of habit memory.Behavioural Brain Research, 234, 299–302.

McGaugh, J. L. (2004). The amygdala modulates the consolidation of memories ofemotionally arousing experiences. Annual Review of Neuroscience, 27, 1–28.

Morrow, B. A., Elsworth, J. D., & Roth, R. H. (2002). Fear-like biochemical andbehavioral responses in rats to the predator odor, TMT, are dependent on theexposure environment. Synapse, 46, 11–18.

Morrow, B. A., Roth, R. H., & Elsworth, J. D. (2000). TMT, a predator odor, elevatesmesoprefrontal dopamine metabolic activity and disrupts short-term workingmemory. Brain Research Bulletin, 52, 519–523.

Muller, M., & Fendt, M. (2006). Temporary inactivation on the medial andbasolateral amygdala differentially affects TMT-induced fear behavior in rats.Behavioural Brain Research, 167, 57–62.

Packard, M. G. (1999). Glutamate infused posttraining into the hippocampus orcaudate-putamen differentially strengthens place and response learning. PNAS,96, 12881–12886.

Packard, M. G. (2009a). Exhumed from thought: Basal ganglia and response learningin the plus-maze. Behavioural Brain Research, 199, 24–31.

Packard, M. G. (2009b). Anxiety, cognition, and habit: A multiple memory systemsperspective. Brain Research, 1293, 121–128.

Packard, M. G., Cahill, L., & McGaugh, J. L. (1994). Amygdala modulation ofhippocampal-dependent caudate nucleus-dependent memory processes.Proceedings of the National Academy of Sciences, 91, 8477–8481.

Packard, M. G., & Gabriele, A. (2009). Peripheral anxiogenic drug injectionsdifferentially affect cognitive and habit memory: Role of basolateralamygdala. Neuroscience, 164, 457–462.

Packard, M. G., & Goodman, J. (2012). Emotional arousal and multiple memorysystems in the mammalian brain. Frontiers in Behavioral Neuroscience, 6, 1–9.

Packard, M. G., & Knowlton, B. J. (2002). Learning and memory functions of the BasalGanglia. Annual Review of Neuroscience, 25, 563–593.

Packard, M. G., & McGaugh, J. L. (1996). Inactivation of hippocampus or caudatenucleus with lidocaine differentially affects expression of place and responselearning. Neurobiology of Learning and Memory, 65, 65–72.

Packard, M. G., & Wingard, J. C. (2004). Amygdala and ‘‘emotional’’ modulation ofthe relative use of multiple memory systems. Neurobiology of Learning andMemory, 82, 243–252.

Pagani, J. H., & Rosen, J. B. (2009). The medial hypothalamic defensive circuit and2,5-dihyro-2,4,5-trimathylthiazoline (TMT) induced fear: Comparison ofelectrolytic and neurotoxic lesions. Brain Research, 1286, 133–146.

Park, C. R., Campbell, A. M., & Diamond, D. M. (2002). Chronic psychosocial stressimpairs learning and memory and increases sensitivity to yohimbine in adultrats. Biological Psychiatry, 15, 994–1004.

Park, C. R., Zoladz, P. R., Conrad, C. D., Fleshner, M., & Diamond, D. M. (2008). Acutepredator stress impairs the consolidation and retrieval of hippocampal-dependent memory in male and female rats. Learning & Memory, 15, 271–280.

Paxinos, G., & Watson, C. (1997). The rat brain in stereotaxic coordinates (4th ed.).Orlando, FL: Academic Press.

Roozendaal, B., & McGaugh, J. L. (1996). Amygdaloid nuclei lesions differentiallyaffect glucocorticoid-induced memory enhancement in an inhibitory avoidancetask. Neurobiology of Learning and Memory, 65, 1–8.

Roozendaal, B., & McGaugh, J. L. (1997). Glucocorticoid receptor agonist andantagonist administration into the basolateral but not central amygdalamodulates memory storage. Neurobiology of Learning and Memory, 67, 176–179.

Schwabe, L., Dickinson, A., & Wolf, O. T. (2011). Stress, habits, and drug addiction:A psychoneuroendocrinological perspective. Experimental and ClinicalPsychopharmacology, 19, 53–63.

Schwabe, L., Oitzl, M. S., Philippsen, C., Richter, S., Bohringer, A., Wippich, W., &Schachinger, H. (2007). Stress modulates the use of spatial versus stimulus-response learning strategies in humans. Learning and Memory, 14, 109–116.

Schwabe, L., & Wolf, O. T. (2009). Stress prompts habit behavior in humans. TheJournal of Neuroscience, 29, 7191–7198.

Takahashi, L. K., Chan, M. M., & Pilar, M. L. (2008). Predator odor fear conditioning:Current perspectives and new directions. Neuroscience and BiobehavioralReviews, 32, 1218–1227.

Takahashi, L. K., Nakashima, B. R., Hong, H., & Watanabe, K. (2005). The smell ofdanger: A behavioral and neural analysis of predator odor-induced fear.Neuroscience and Biobehavioral Reviews, 29, 1157–1167.

VanElzakker, M. B., Zoladz, P. R., Thompson, V. M., Park, C. R., Halonen, J. D., Spencer,R. L., & Diamond, D. M. (2011). Influence of pre-training predator stress on theexpression of c-fos mRNA in the hippocampus, amygdala, and striatumfollowing long-term spatial memory retrieval. Frontiers in BehavioralNeuroscience, 5, 1–13.

Vernet-Maury, E., Polak, E. H., & Demael, A. (1984). Structure-activity relationship ofstress-inducing odorants in the rat. Journal of Chemical Ecology, 10, 1007–1018.

Vernet-Maury, E. (1980). Trimethyl-thiazoline in fox feces: A natural alarmingsubstance for the rat. In H. van der Starre (Ed.), Olfaction and Taste. WashingtonDC: IRL Press.

Wallace, K. J., & Rosen, J. B. (2001). Neurotoxic lesions of the lateral nucleus of theamygdala decrease conditioned fear but not unconditioned fear of a predatorodor: Comparison with electrolytic lesions. Journal of Neuroscience, 21,3619–3627.

White, N. M., & McDonald, R. J. (2002). Multiple parallel memory systems in thebrain of the rat. Neurobiology of Learning and Memory, 77, 125–184.

Wingard, J. C., & Packard, M. G. (2008). The amygdala and emotional modulation ofcompetition between cognitive and habit memory. Behavioral Brain Research,193, 126–131.

Zoladz, P. R., Park, C. R., Halonen, J. D., Salim, S., Alzoubi, K. H., Srivareerat, M.,Fleshner, M., Alkadhi, K. A., & Diamond, D. M. (2012). Differential expression ofmolecular markers of synaptic plasticity in the hippocampus, prefrontal cortex,and amygdala in response to spatial learning, predator exposure, and stress-induced amnesia. Hippocampus, 22, 577–589.