Neurobiology of Learning and Memory 118 (2015) 55–63

Contents lists available at ScienceDirect

Neurobiology of Learning and Memory

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

Dopamine depletion in either the dorsomedial or dorsolateral striatumimpairs egocentric Cincinnati water maze performance while sparingallocentric Morris water maze learning

http://dx.doi.org/10.1016/j.nlm.2014.10.0091074-7427/� 2014 Elsevier Inc. All rights reserved.

⇑ Corresponding authors at: Cincinnati Children’s Research Foundation, Divisionof Neurology (MLC 7044), 3333 Burnet Ave., Cincinnati, OH 45229, United States.Fax: +1 (513) 636 3912.

E-mail addresses: Charles.Vorhees@cchmc.org (C.V. Vorhees), Michael.Wil-liams@cchmc.org (M.T. Williams).

Amanda A. Braun a, Robyn M. Amos-Kroohs a, Arnold Gutierrez a, Kerstin H. Lundgren b, Kim B. Seroogy b,Matthew R. Skelton a, Charles V. Vorhees a,⇑, Michael T. Williams a,⇑a Division of Neurology, Department of Pediatrics, Cincinnati Children’s Research Foundation, University of Cincinnati College of Medicine, Cincinnati, OH 45229, United Statesb Department of Neurology, University of Cincinnati College of Medicine, Cincinnati, OH 45267, United States

a r t i c l e i n f o a b s t r a c t

Article history:Received 15 January 2013Revised 24 October 2014Accepted 29 October 2014Available online 13 November 2014

Keywords:Dorsolateral striatumDorsomedial striatumDopamineEgocentricAllocentric

Both egocentric route-based learning and spatial learning, as assessed by the Cincinnati water maze(CWM) and Morris water maze (MWM), respectively, are impaired following an 80% dopamine (DA) lossin the neostriatum after 6-hydroxydopamine (6-OHDA) administration in rats. The dorsolateral striatum(DLS) and the dorsomedial striatum (DMS) are implicated in different navigational learning types, namelythe DLS is implicated in egocentric learning while the DMS is implicated in spatial learning. This exper-iment tested whether selective DA loss through 6-OHDA lesions in the DMS or DLS would impair one orboth types of navigation. Both DLS and DMS DA loss significantly impaired route-based CWM learning,without affecting spatial or cued MWM performance. DLS 6-OHDA lesions produced a 75% DA loss in thisregion, with no changes in other monoamine levels in the DLS or DMS. DMS 6-OHDA lesions produced a62% DA loss in this region, without affecting other monoamine levels in the DMS or DLS. The results indi-cate a role for DA in DLS and DMS regions in route-based egocentric but not spatial learning and memory.Spatial learning deficits may require more pervasive monoamine reductions within each region beforedeficits are exhibited. This is the first study to implicate DLS and DMS DA in route-based egocentricnavigation.

� 2014 Elsevier Inc. All rights reserved.

1. Introduction environment is fluid and represented in a common coordinate

Impairments in navigational ability are present in numeroushuman disorders where they impair the quality of life and increasedependency (Aguirre & D’Esposito, 1999; Iaria, Palermo,Committeri, & Barton, 2009; Livingstone & Skelton, 2007; Sanders,Holtzer, Lipton, Hall, & Verghese, 2008; Weniger & Irle, 2006). Suc-cessful navigation requires complex interactions among multipledistinct, but parallel cognitive processes that can be subdivided intoegocentric (self-oriented path integration and route-based) andallocentric (map-based) wayfinding. Route-based navigationinvolves a representation of space connected by ‘‘nodes’’ or choicepoints representing successive decision points in a grid or pathway(Aguirre & D’Esposito, 1999; Byrne, 1982). In the allocentricprocess, the navigator’s spatial orientation to distal cues in the

map system external to the navigator (Byrne, 1982; Garber, 2000).Considerable behavioral, anatomical, and electrophysiological

evidence suggests that the neostriatum is an important modulatorin both egocentric and allocentric learning (Braun, Graham,Schaefer, Vorhees, & Williams, 2012; Cook & Kesner, 1988; Devan,McDonald, & White, 1999; Devan & White, 1999; Jog, Kubota,Connolly, Hillegaart, & Graybiel, 1999; McDonald & White, 1994;McGeorge & Faull, 1989; Mizumori, Puryear, & Martig, 2009;Mizumori, Yeshenko, Gill, & Davis, 2004; Packard, 2009; Packard,Hirsh, & White, 1989; Penner & Mizumori, 2012; Potegal, 1969,1972; Ragozzino, Leutgeb, & Mizumori, 2001; Taube, 1998;Whishaw & Dunnett, 1985; Whishaw, Mittleman, Bunch, &Dunnett, 1987). The neostriatum is a heterogeneous structure withanatomical subregions for different functions. The dorsomedial stri-atum (DMS) receives primary inputs from multiple sensory andassociation areas, such as the hippocampus and medial prefrontalcortex, and while lesions in this area have widespread effects, theyoften produce impairments in allocentric learning (Colombo, Davis,& Volpe, 1989; Devan & White, 1999; Devan et al., 1999; McGeorge& Faull, 1989; Whishaw et al., 1987). For example, DMS lesions or

56 A.A. Braun et al. / Neurobiology of Learning and Memory 118 (2015) 55–63

DMS dopamine (DA) depletion result in allocentric learning andplace strategy deficits in the Morris water maze (MWM) and T-maze, respectively (Devan & White, 1999; Devan et al., 1999; Lex,Sommer, & Hauber, 2011). Sensory and motor cortices have majorprojections to the dorsolateral striatum (DLS), that are associatedwith egocentric or response learning and stimulus–response habitformation (Devan & White, 1999; McGeorge & Faull, 1989;Packard & McGaugh, 1996; Palencia & Ragozzino, 2005; Reading,Dunnett, & Robbins, 1991; White, 1997; Yin & Knowlton, 2004;Yin & Knowlton, 2006; Yin, Knowlton, & Balleine, 2004, 2006). How-ever, this heterogeneity of function within the neostriatum may notbe fully preserved in regard to egocentric learning. Excitotoxiclesions of the DMS and DLS each result in a severe learning impair-ment in a 14-unit T-maze procedural learning task, implicatingboth regions in egocentric learning (Pistell et al., 2009).

The focus of the present experiments was to elucidate theregionally-specific role of neostriatal DA in egocentric and allocen-tric navigation. DA in the neostriatum influences both glutamater-gic afferents and striatal medium spiny neuronal efferents thatmodulate striatal output (Penner & Mizumori, 2012). Previously,we showed that widespread neostriatal 6-hydroxydopamine(6-OHDA)-induced DA reduction impaired learning in both theallocentric MWM and route-based Cincinnati water maze (CWM)(Braun et al., 2012). While DMS DA has been implicated in allocen-tric T-maze learning strategy (Lex et al., 2011), it has not beentested for involvement in either route-based or allocentric naviga-tion. Moreover, the role of DA in DLS-mediated route-based or allo-centric navigation has yet to be tested. Accordingly, we testedgroups of animals given selective 6-OHDA injections in either theDMS or DLS and evaluated them in the CWM and MWM, respec-tively (test order was examined previously (Broening, Morford,Inman-Wood, Fukumura, & Vorhees, 2001; Skelton et al., 2009)compared with sham-operated controls. Motivation and swim-ming ability were assessed to control for potential performancechanges not associated with learning.

2. Methods

2.1. Animals

Adult male Sprague–Dawley CD IGS rats (225–250 g at the timeof arrival) were purchased from Charles River Laboratories,Raleigh, NC (strain 001). Animals were pair-housed in polypropio-nate cages (46 � 24 � 20 cm) containing woodchip bedding for atleast a 1-week acclimation period prior to surgery. Animals hadfree access to food and water, were housed in an environmentallycontrolled vivarium (21 ± 1 �C), and were on a 14 h light–darkcycle (lights on at 600 h). All procedures were in compliance withthe Institutional Animal Care and Use Committee and the vivariumis fully accredited by the Association for the Assessment andAccreditation of Laboratory Animal Care.

2.2. Surgery

Rats were anesthetized with 2–4% isoflurane (IsoThesia; ButlerAnimal Health Supply, Dublin, OH) with continuous administrationvia a nose cone throughout surgery. Rats were placed in a motor-ized, computer-controlled stereotaxic apparatus (StereoDrive,Stoelting Co., Wood Dale, IL), and were given bilateral injectionsof 6-OHDA (Sigma, St. Louis, MO) using a 26 gauge 10 ll HamiltonGastight syringe (Reno, NV). Coordinates were based on the Paxi-nos and Watson brain atlas (Paxinos, Watson, Pennisi, & Topple,1985). For the DLS lesions, a volume of 3 ll [4 lg/ll 6-OHDA in0.2% ascorbic acid saline solution] was injected over 9 min (frombregma: AP: +0.2 mm; ML: ±3.5 mm; from skull: DV: �4.8 mm),

with the needle left in place for 1 min following injection. For theDMS lesions, a volume of 0.4 ll [30 lg/ll] was injected in each siteover 4 min (from bregma: AP: +1.0 mm; ML: ±1.7 mm; DV:�5.0 mm; and AP: �0.4 mm; ML: ±2.6 mm; DV: �4.5 mm), withthe needle left in place for 5 min following completion of injection.Control animals (SHAM) received an identical amount of saline in0.2% ascorbic acid vehicle (VEH) using the same procedure for itsparticular group. Following surgery, animals were given 0.1 mlbuprenorphine hydrochloride to minimize pain. Animals wereallowed to recover for 2 weeks before the beginning of testing.The number of animals represented in each group is given in thefigure legends.

2.3. Behavioral testing

2.3.1. Straight channelOne day prior to CWM testing, animals were tested for swim-

ming ability in a 244 cm long � 15 cm wide � 51 cm high waterfilled (38 cm deep) straight channel for 4 consecutive trialswith a maximum time limit of 2 min/trial (Herring, Schaefer,Gudelsky, Vorhees, & Williams, 2008; Vorhees et al., 2008). Straightchannel swimming served three functions: (a) to acclimate animalsto swimming, (b) to teach that escape was possible by climbing onthe submerged platform at the opposite end of the channel, and (c)to determine if animals had comparable swimming ability.

2.3.2. Cincinnati water mazeThe CWM is a nine-unit multiple T water maze (21 ± 1 �C) as

described previously (Vorhees, 1987; Vorhees, Weisenburger,Acuff-Smith, & Minck, 1991; Vorhees et al., 2008). Animals hadto locate a submerged escape platform; the room was illuminatedwith infrared lighting in order to eliminate visual cues; a videocamera was mounted above the maze sensitive to light in the infra-red range and fed to a monitor in another room. Two trials/day(5 min limit/trial) were given. If an animal failed to find the escapewithin 5 min on trial-1 of each day, there was at least a 5 minintertrial interval (ITI) before trial-2. If they found the escape ontrial-1 in less than 5 min, trial-2 was given immediately. Animalsreaching the time limit were removed from the maze from wher-ever they were when the time limit was reached. Latency to escapeand number of errors (defined as head and shoulder entry in a stemor arm of a T or reentry into the start channel) were recorded. Tocorrect for animals that stopped searching, they were given anerror score equal to the number of errors +1 made by the animalthat found the escape and made the most errors in <5 min. Animalsthat never found the platform were removed from analysis. Datafor the CWM were analyzed in 2-day (4 trials) blocks similar tothe 4-trial blocks used to analyze MWM data.

2.3.3. Morris water maze hidden platformTo test spatial navigational learning, MWM hidden platform

testing began the day following CWM completion (Morris, 1981).Animals were placed in a 244 cm diameter tank of water(21 ± 1 �C) and were required to find a submerged platform(10 cm diameter) in a stationary position with pseudo-random-ized, balanced cardinal and ordinal start positions. For 6 days, ratswere given 4 trials/day with a 2 min trial limit and an ITI of 15 s (onthe platform). If a rat failed to find the platform within the timelimit, it was placed on the platform. On the 7th day, a 30 s probetrial was given from a novel start position with the platformremoved. Data were collected using video tracking software (Any-Maze, Stoelting Co., Wood Dale, IL).

2.3.4. Morris water maze cuedCued MWM testing began the day following the hidden platform

testing and was conducted over two days. A yellow plastic ball was

A.A. Braun et al. / Neurobiology of Learning and Memory 118 (2015) 55–63 57

attached to the top of a brass rod mounted in the center of the sub-merged platform (10 cm diameter) to mark its location. On each day,rats were given 4 trials with the locations of the platform and start-ing positions randomized (2 min trial limit with an ITI of 15 s on theplatform +15–20 s to reposition the platform). Latency was recorded(AnyMaze could not track rats under these lighting conditions).

2.4. Tissue collection

Tissue collection took place following the completion of testing.Animals were brought to an adjacent suite and decapitated. Brainswere removed and the neostriatum dissected and further seg-mented into the DMS and DLS. Brain regions were rapidly frozenfor later monoamine assay as described (Williams et al., 2007).

2.5. Monoamine assays

Monoamines were assayed via high performance liquid chroma-tography with electrochemical detection (HPLC-ECD). Frozen tis-sues were weighed, thawed, and sonicated in appropriatevolumes of 0.1 N perchloric acid (Fisher Scientific, Pittsburgh, PA).Samples were centrifuged for 14 min at 13,000 RCF at 4 �C. Thesupernatant sample was transferred to a new vial for injection ona Supelco Supelcosil™ LC-18 column (150 � 4.6 mm, 3 lm;Sigma–Aldrich Co., St. Louis, MO). The HPLC system consisted of aWaters 717plus autosampler (Waters Corp., Milford, MA), ESA584 pump, and Coulochem III electrochemical detector. The poten-tial settings were�150 mV for E1 and +250 mV for E2, with a guardcell potential set at +350 mV. MD-TM mobile phase (ESA, Inc.) wasused and consisted of 75 mM sodium dihydrogen phosphate(monohydrate), 1.7 mM 1-octanesulfonic acid sodium salt, 100 ll/l triethylamine, 25 lM EDTA, and 10% acetonitrile, with a final pHof 3.0. The pump flow rate was set at 0.7 ml/min, and the sampleswere run at 28 �C. Standards for DA, 3,4-dihydroxyphenylaceticacid (DOPAC), homovanillic acid (HVA), norepinephrine (NE), 5-HT, and 5-hydroxyindoleacetic acid (5-HIAA) (all obtained fromSigma–Aldrich Co., St. Louis, MO) were prepared in 0.1 N perchloricacid. All neurotransmitters were run on a single chromatogram.

2.6. Immunohistochemistry

Using the same surgical procedures as in behavioral testing, adifferent set of rats were given a unilateral injection of 6-OHDAin the DLS (N = 3) or DMS (N = 3) and a VEH injection on the con-tralateral side using the coordinates described. Two weeks aftersurgery the animals were brought into an adjacent suite, perfusedtranscardially with 4% paraformaldehyde, and the brains dissected,postfixed, and sunk in sucrose overnight. Brains were sectioned (at30-lm thickness) on a microtome, and the free-floating sectionsprocessed for tyrosine hydroxylase (TH) immunohistochemistryas previously described (Hemmerle, Dickerson, Herman, &Seroogy, 2014), using mouse monoclonal anti-TH primary antibody(MAB318, diluted 1:8000; EMD Millipore, Telecuma, CA), biotinyl-ated horse anti-mouse IgG secondary antibody (BA-2000, diluted1:200; Vector Laboratories, Burlingame, CA), and ABC Elite Kitreagents (Vector Laboratories) with diaminobenzidine as chroma-gen. Neostriatal immunostaining for TH was analyzed for theregional specificity of 6-OHDA injections as indicated by TH deple-tion in the DMS or DLS. Sections were viewed and scanned at 20�on an Aperio AT2 slide scanner and uploaded to Aperio eSlide Man-ager (Leica Biosystems, Buffalo Grove, IL).

2.7. Statistical analysis

DLS and DMS groups were tested separately, therefore data fromeach experiment were analyzed independently. Data were analyzed

using mixed linear ANOVA models (SAS Proc Mixed, SAS Institute9.2, Cary, NC). The covariance matrix for each dataset was checkedusing best fit statistics. In most cases, the best fit was to the autore-gressive-1 covariance structure. Kenward–Rodger adjusted degreesof freedom were used. Measures taken repetitively on the same ani-mal, such as day or block, were within-subject factors. Significantinteractions were analyzed using simple-effect slice ANOVAs ateach level of the repeated measure factor. Biochemical data wereanalyzed using two-tailed t-tests. Significance was set at p 6 0.05.Data are presented as least square (LS) mean ± LS SEM.

3. Results

3.1. Immunohistochemistry

Representative sections from one animal that received a DLS orDMS lesion are shown (Fig. 1). Unilateral 6-OHDA injection in theDLS resulted in DLS-specific loss of TH immunostaining (Fig. 1B).No reduction of striatal TH immunoreactivity was observed inthe contralateral DLS with VEH injection (Fig. 1A).

The striatal regional specificity of the DMS injections of 6-OHDAwas also confirmed by TH immunostaining. Unilateral 6-OHDAinjections in the DMS resulted in DMS-specific loss of TH immuno-staining (Fig. 1D) compared with the contralateral DMS injectedwith VEH (Fig. 1C).

3.2. Dorsolateral striatal 6-OHDA lesions

3.2.1. Straight channelNo difference in time to swim the straight channel was

observed across trials between DLS 6-OHDA-treated and SHAManimals (LS mean ± LS SEM across trials: 6-OHDA: 13.8 ± 1.1 s;SHAM: 12.3 ± 1.2 s).

3.2.2. Cincinnati water maze6-OHDA-treated animals had significantly increased latencies

to find the platform compared with SHAM animals(F(1,37.7) = 5.75, p 6 0.05; Fig. 2A) with significantly longer laten-cies observed from block-3 through block-8 (treatment � block:F(8,177) = 2.18, p 6 0.05; Fig. 2B). DLS 6-OHDA-treated animalscommitted significantly more errors compared with SHAM animals(F(1,39) = 6.56, p 6 0.05; Fig. 2A) with significantly more errorsobserved during block-3 through block-6 and block-8 (treat-ment � block: F(8,177) = 2.24, p 6 0.05; Fig. 2C).

3.2.3. Morris water maze6-OHDA-treated animals showed no difference in MWM perfor-

mance compared with SHAM animals. No significant differencewas found in latency to find the platform (Fig. 3), path length, orcumulative distance to the platform. Swim speed did not differbetween groups (6-OHDA: 0.41 ± 0.07 m/s; SHAM: 0.47 ± 0.09 m/s). Initial heading error and average heading error were not signif-icantly different between 6-OHDA-treated animals and SHAM con-trols. During the probe trial, 6-OHDA-treated animals were notaffected on the number of platform crossovers (6-OHDA:0.66 ± 0.28; SHAM: 0.5 ± 0.26) or average distance from the plat-form site compared with SHAM animals (6-OHDA: 0.83 ± 0.06 m;SHAM: 0.85 ± 0.06 m). For cued platform trials, there was no signif-icant latency difference between 6-OHDA-treated animals andSHAM controls (averaged across days and trials: 6-OHDA:28.76 ± 3.31 s; SHAM: 23.48 ± 3.54 s).

3.2.4. Monoamine assessment6-OHDA injection caused a 75% decrease in DLS DA

compared with SHAM animals (t(22) = 11.2, p 6 0.001; Fig. 4A)

Fig. 1. Immunohistochemistry. VEH injections in the DLS (A) compared with contralateral DLS 6-OHDA injections (B). DLS 6-OHDA selectively destroyed TH neurons in theDLS. TH loss was limited to the DMS following unilateral 6-OHDA injections in the DMS (D) compared with contralateral DMS VEH injections (C).

Fig. 2. DLS Cincinnati water maze latency and errors. There was a main effect of lesion: DLS lesioned animals had a longer latency and made more errors than SHAMs (A).Across days, DLS lesioned animals had significantly longer latencies during blocks 3–8 (B) and made significantly more errors during blocks 3–6 and block 8, compared withSHAMs. N = 12/group. ⁄ p 6 0.05, ⁄⁄ p 6 0.01.

58 A.A. Braun et al. / Neurobiology of Learning and Memory 118 (2015) 55–63

with significant decreases in DA metabolites (DOPAC: t(22) = 6.36,p 6 0.001; HVA = t(20) = 6.06, p 6 0.001) and utilization ratios(DOPAC/DA = t(22) = 5.38, p 6 0.001; overall turnover ratio:t(20) = 3.10, p 6 0.001) (Table 1). NE and 5-HT levels in the DLS

were not altered in 6-OHDA-treated animals compared with SHAMcontrols (Fig. 4B and C, respectively). To determine if DLS 6-OHDAinjections affected the DMS this region was also analyzed. DAconcentrations (Fig. 4D), metabolites and turnover in the DMS after

Fig. 3. DLS Morris water maze latency. 6-OHDA lesions in the DLS did not have asignificant effect on latency (main effect or day � lesion interaction) in the MWM tofind the hidden platform compared with SHAMs.

A.A. Braun et al. / Neurobiology of Learning and Memory 118 (2015) 55–63 59

6-OHDA DLS injection were not significantly different comparedwith SHAM animals. NE (Fig. 4E) and 5-HT (Fig. 4F) levels were alsonot different.

3.3. Dorsomedial striatal 6-OHDA lesions

3.3.1. Straight channelNo difference in time to swim the straight channel was

observed between 6-OHDA-treated animals and SHAM controls(LS mean ± LS SEM across trials: 6-OHDA: 18.53 ± 2.52 s; SHAM:17.33 ± 2.22 s).

3.3.2. Cincinnati water maze6-OHDA-treated rats had increased latency to find the platform

compared with SHAM animals (F(1,19.4) = 4.36, p 6 0.05; Fig. 5A),but the treatment � block interaction was not significant (Fig. 5B).A trend toward significantly more errors overall in 6-OHDA-treatedanimals was also seen (F(1,19.8) = 4.04, p 6 0.10; Fig. 5A), howeveron blocks 4–6 and block 9 lesioned animals made significantlymore errors (treatment � block: F(8,109) = 2.59, p 6 0.05; Fig. 5C)compared with SHAM control animals.

3.3.3. Morris water maze6-OHDA-treated animals showed no difference in MWM perfor-

mance compared with SHAM animals. No significant differenceswere found in latency to reach the platform (Fig. 6), path length,or cumulative distance to the platform. Swim speed was not signif-icantly different between the groups (6-OHDA: 0.29 ± 0.01 m/s;SHAM: 0.28 ± 0.01 m/s). Initial and average heading errors werenot significantly altered. During the probe trial, 6-OHDA-treatedrats showed no significant effect on the number of platform cross-ings (6-OHDA: 1.00 ± 0.58; SHAM: 1.00 ± 0.38), or on average dis-tance from the platform site (6-OHDA: 0.93 ± 0.06 m; SHAM:0.73 ± 0.07 m; t(13) = 2.07, p 6 0.10). For cued platform trials, therewas no significant latency difference between 6-OHDA-treated ani-mals and SHAM animals (averaged across days and trials: 6-OHDA:27.66 ± 9.30 s; SHAM: 36.27 ± 8.21 s).

3.3.4. Monoamine assessmentFollowing 6-OHDA injection in the DMS, DA concentrations

were decreased 62% compared with SHAM controls (t(14) = 8.87,p 6 0.001; Fig. 7A). 6-OHDA injection in the DMS also decreasedDA metabolites (DOPAC: t(14) = 3.86, p 6 0.01; HVA: t(14) = 2.74,p 6 0.05) and increased turnover (DOPAC/DA: t(14) = 2.42,p 6 0.01; HVA/DA: t(14) = 5.29, p 6 0.001; overall turnover:

t(14) = 3.47, p 6 0.01) compared with SHAM controls (Table 2).DMS NE (Fig. 7B) and 5-HT (Fig. 7C) were not altered followingDMS 6-OHDA injection compared with SHAM injections.

DA concentrations following 6-OHDA DMS injection were notsignificantly different in the DLS compared with SHAM controls(Fig. 7D). DA metabolites and turnover and NE (Fig. 7E) and 5-HT(Fig. 7F) were not significantly altered in the DLS following 6-OHDA DMS injection compared with SHAM controls.

4. Discussion

6-OHDA injections in the DLS reduced DA levels by 75% andresulted in CWM route-based navigation deficits, but had no effecton hidden platform allocentric learning in the MWM. DMS DAdepletion of 62% also resulted in route-based CWM navigationaldeficits, without altering MWM-based allocentric learning. Thesedeficits were independent of motivational or motoric impairments(no differences in straight channel, cued platform MWM, or swimspeed in the MWM). The DMS has been implicated in other behav-iors, such as modulation of expected reward value, initiationbehavior, and goal-directed behavior (Calaminus & Hauber, 2009;Fouquet et al., 2013; Kim, Lee, & Jung, 2013; Mizumori et al.,2009; Penner & Mizumori, 2012; White, 1997). These other pro-cesses are likely not contributing to the egocentric impairmentseen in the present study as the other tested behaviors were unal-tered. For each of the striatal subregions, the 6-OHDA injectiondamage was limited to the targeted area, precluding potentialeffects from the other region to explain the route-based naviga-tional deficits. NE and 5-HT were not altered, regardless of whichstriatal subregion was lesioned, leaving DA loss to account forthe observed learning impairments. It is unlikely that brain regionsoutside of the neostriatum were involved, as whole neostriatal DAloss has not been shown to affect monoamine levels in other brainregions associated with learning (Braun et al., 2012). The observedchanges in DA metabolites and turnover are consistent with whatothers have found for these regions (Aguiar et al., 2008; Braunet al., 2012; Chen et al., 2007; Henze et al., 2005; Tadaieskyet al., 2008).

The neostriatum has long been associated with egocentriclearning (Braun et al., 2012; Cook & Kesner, 1988; Devan &White, 1999; Devan et al., 1999; Mizumori, Canfield, & Yeshenko,2005; Mizumori et al., 2004; Packard, 2009; Packard & Knowlton,2002; Packard & McGaugh, 1996; Packard et al., 1989; Palencia &Ragozzino, 2005; Penner & Mizumori, 2012; Potegal, 1969, 1972;Taube, 1998; White, 1997; White & McDonald, 2002; Yin &Knowlton, 2004; Yin & Knowlton, 2006; Yin et al., 2004, 2006).While a separation of function between the DMS and DLS in regardto allocentric learning tasks has been observed, both regions haveindependently been implicated in egocentric learning. Glutamatein the DLS has a modulatory role in egocentric response learningin a T-maze and post-training silencing of the DLS inhibits egocen-tric response (Packard & McGaugh, 1996; Palencia & Ragozzino,2005). Recently, Etienne et al. showed that pharmacological inhibi-tion of the DMS reduced route-based (direction-based) learning,but not allocentric or cued learning in rhesus macaques (Etienne,Guthrie, Gross, & Boraud, 2012). Excitotoxic lesions of either theDMS or DLS impaired procedural learning in a 14-unit T-maze(Pistell et al., 2009). While both Pistell et al. and this study impli-cate the DMS and DLS in complex egocentric learning, the 14-unitT-maze and the CWM have several differences that distinguish theCWM as a test of route-based egocentric navigation rather than asa task of procedural memory as is the 14 unit T-maze. For example,in the CWM there are no spatial cues available, it uses water as themotivator, and animals are not forced into making a left–rightresponse choice during navigation. The 14 unit T-maze has spatial

Fig. 4. DLS monoamine levels. 6-OHDA lesions in the DLS significantly decreased DA levels in the DLS by 75% (A), with no change in NE (B), or 5-HT (C) DLS levels, comparedwith SHAMS. DMS DA (D), NE (E), and 5-HT (F) levels were not altered following DLS DA depletion. ⁄⁄⁄ p 6 0.001.

Table 1Dorsolateral striatal (DLS) dopamine metabolite levels (3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA)) and turnover rate after DLS 6-OHDA injection.

DLS Lesion DOPAC (pg/mg) HVA (pg/mg) DOPAC/DA HVA/DA (DOPAC + HVA)/DA

6-OHDA 499.9 ± 92.4*** 232.4 ± 49.7*** 0.18 ± 0.01*** 0.08 ± 0.01 0.25 ± 0.02**

SHAM 1308.2 ± 85.9 721.7 ± 60.8 0.11 ± 0.004 0.06 ± 0.005 0.17 ± 0.008

** p 6 0.01.*** p 6 0.001.

60 A.A. Braun et al. / Neurobiology of Learning and Memory 118 (2015) 55–63

cues available, uses shock for the motivator, guillotine doors closeoff previously visited arms, and there is a forced left–right responseto navigate correctly. Testing in the CWM lasts for 18 days (2 trials/day) allowing animals to demonstrate long-term learning andmemory ability, whereas testing in the 14 unit T-maze lasts for1 day with 15 trials.

How DA signaling in the DMS and DLS modulates egocentriclearning is currently unknown. The DMS and DLS possess both allo-centric place cells and neurons that fire only to specific egocentricresponse movements such as turns, forward movement, and headdirection (Lavoie & Mizumori, 1994; Ragozzino et al., 2001;Wiener, 1993). The egocentric response cells in the DLS and/or

Fig. 5. DMS Cincinnati water maze latency and errors. (A) 6-OHDA treated animals in the DMS had significantly longer latencies to find the platform, and a trend towardmaking significantly more errors compared with controls over the duration of testing. DMS lesioned animals had significantly longer latencies than SHAMs (B), and madesignificantly more errors than SHAM controls on blocks 4–6 and block 9 (C). N = 7/6-OHDA; 9/SHAM. + p 6 0.1, ⁄ p 6 0.05.

Fig. 6. DMS Morris water maze latency. 6-OHDA lesions in the DMS did not have asignificant effect on latency (main effect or day � lesion interaction) to find thehidden platform in the MWM compared with SHAMs.

Fig. 7. DMS Monoamine levels. 6-OHDA injection in the DMS significantlydecreased DA levels in the DMS by 62% (A), with no change in NE (B), or 5-HT (C)DMS levels compared with SHAMS. DLS DA (D), NE (E), and 5-HT (F) levels were notaltered following DMS 6-OHDA lesions. ⁄⁄⁄ p 6 0.001.

A.A. Braun et al. / Neurobiology of Learning and Memory 118 (2015) 55–63 61

DMS could be influenced by DA projections and compromised fol-lowing DA loss. Striatal DA could also be influencing egocentriclearning through its direct regulation of glutamatergic input tomedium spiny neurons, inputs that are necessary for initial ego-centric learning in the DLS (Palencia & Ragozzino, 2005; Sesack,Carr, Omelchenko, & Pinto, 2003).

The finding that allocentric learning was spared following DLSDA loss is consistent with the literature that shows this area isnot necessary for this type of learning, regardless of lesion type(Devan et al., 1999; Mizumori et al., 2009; Packard, 2009; Yin &Knowlton, 2004; Yin et al., 2006). Electrolytic and excitotoxiclesions of the DMS cause deficits in hidden platform MWM learn-ing (Devan & White, 1999; Devan et al., 1999). DA in the posteriorDMS has been implicated in place learning, however not specifi-cally in MWM-based allocentric learning (Lex et al., 2011). Inthe latter study, when given a choice between solving a T-mazeusing an egocentric response strategy or an allocentric placestrategy subsequent to posterior DMS DA depletion, a signifi-cantly higher proportion of animals utilized an egocentricresponse (83%) compared with SHAM controls (50%) early in test-ing. During later phases of learning no differences were observedbetween groups. As it is more common for animals to utilize aplace strategy during early training and transition into the

response strategy following continued training, the inferencewas that DA-depleted animals exhibited a deficit in allocentricperformance during the phase of acquisition when place learningnormally dominates. No overall learning deficit was observed inthat both groups learned the task; only the strategy used initiallydiffered.

Table 2Dorsomedial striatal (DMS) dopamine metabolite levels (3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA)) and turnover rate after DMS 6-OHDA injection.

DMS lesion DOPAC (pg/mg) HVA (pg/mg) DOPAC/DA HVA/DA (DOPAC + HVA)/DA

6-OHDA 695.4 ± 111.8** 346.9 ± 29.9* 0.25 ± 0.03* 0.12 ± 0.01*** 0.38 ± 0.04**

SHAM 1324.8 ± 114.2 482.2 ± 36.7 0.18 ± 0.01 0.06 ± 0.004 0.24 ± 0.01

* p 6 0.05.** p 6 0.01.

*** p 6 0.001.

62 A.A. Braun et al. / Neurobiology of Learning and Memory 118 (2015) 55–63

Differences between mazes may explain the lack of effect in thecurrent study for place learning compared with the Lex et al.(2011) study. The T-maze is more rudimentary than the CWM,making it easier to solve. While the T-maze gives a choice betweentwo strategies, the CWM and MWM are configured such that onlyone strategy or the other is effective. The CWM is tested underinfrared light eliminating spatial cues, and animals do not developan egocentric learning strategy in the MWM using the testing pro-tocol herein (Morris, 1981). While animals in the Lex et al. (2011)study resorted to a response strategy over a place learning strategyin the T-maze, animals in the present study learned at the samerate as controls when given only the option of allocentric learningin the MWM, but had deficits when given only the option of ego-centric learning in the CWM. Because the CWM is a more complexegocentric learning task, it was able to uncover the involvement ofDA in the DMS for this type of learning.

It is unlikely that greater DA loss would have resulted in aMWM allocentric learning impairment. Allocentric learning defi-cits in the MWM require a threshold of about 60% neostriatal DAdepletion (Braun et al., 2012; Da Cunha et al., 2003; De Leonibuset al., 2007; Lindner et al., 1999; Miyoshi et al., 2002; Mura &Feldon, 2003; Whishaw & Dunnett, 1985), a level of reductionexceeded in the present experiments. Since DA loss in the DLSand DMS lesioned groups surpassed this level of reduction, it sug-gests that more widespread neostriatum DA loss is necessarybefore allocentric learning deficits are observed rather than greatersubregional loss. In agreement with this, genetically DA-deficientmice unable to show allocentric MWM learning exhibit a restora-tion of learning following DA supplementation to either the DMSor DLS; this also indicates that allocentric learning does not dependon DA signaling in a single striatal subregion (Darvas & Palmiter,2010). Taken together with previous studies, it appears that allo-centric learning deficits following DMS lesions require a lesion ofmore than DA alone.

Limitations to the present study include: attempts to depleteDA further without causing 6-OHDA damage to other regions wereunsuccessful; that we tested only CWM egocentric learning and itis possible that other tasks might show different effects; that nav-igation is undoubtedly the product of complex interactions amongdifferent neurotransmitters and receptors in different regions, suchthat isolating only the role of DA in two neostriatal subregions isnecessarily artificial; and test order may have contributed to thefindings since there may have been positive transfer from theCWM to the MWM that, if it occurred, would have benefitedMWM performance and reduced apparent effects on allocentricnavigation. This is unlikely though, as mice lacking neostriatalDA are impaired in strategy-switching, and excitotoxic lesions ofthe DMS increase perseverative behavior in both rats and marmo-set monkeys (Castane, Theobald, & Robbins, 2010; Clarke, Robbins,& Roberts, 2008; Darvas & Palmiter, 2010; Rogers, Baunez, Everitt,& Robbins, 2001). Future studies are needed, however, to furtherclarify each of these points.

While neostriatal DA has been shown to be involved in egocen-tric navigation (Anguiano-Rodriguez, Gaytan-Tocaven, & Olvera-Cortes, 2007; Braun et al., 2012), this is the first experiment toimplicate both DLS and DMS DA as modulatory factors in egocentric

route-based navigation. This study is also the first to directly impli-cate the DMS in route-based learning. Taken together with our pre-vious data where DA was depleted throughout the neostriatum(Braun et al., 2012), the findings support the view that neostriatalDA involvement in allocentric learning requires contributions fromboth the DLS and DMS. Conversely, the DLS and DMS can each influ-ence egocentric learning independently.

Acknowledgments

This work was supported by NIH project grant ES015689 andtraining grant ES07051 and the Gardner Family Center for Parkin-son’s Disease and Movement Disorders.

References

Aguirre, G. K., & D’Esposito, M. (1999). Topographical disorientation: A synthesisand taxonomy. Brain, 122(Pt 9), 1613–1628.

Anguiano-Rodriguez, P. B., Gaytan-Tocaven, L., & Olvera-Cortes, M. E. (2007).Striatal serotonin depletion facilitates rat egocentric learning via dopaminemodulation. European Journal of Pharmacology, 556, 91–98.

Braun, A. A., Graham, D. L., Schaefer, T. L., Vorhees, C. V., & Williams, M. T. (2012).Dorsal striatal dopamine depletion impairs both allocentric and egocentricnavigation in rats. Neurobiology of Learning and Memory, 97, 402–408.

Broening, H. W., Morford, L. L., Inman-Wood, S. L., Fukumura, M., & Vorhees, C. V.(2001). 3,4-methylenedioxymethamphetamine (ecstasy)-induced learning andmemory impairments depend on the age of exposure during early development.Journal of Neuroscience, 21, 3228–3235.

Byrne, R. W. (1982). Geographical knowledge and orientation. In Normality andpathology in cognitive functions (pp. 239–264). London: Academic.

Calaminus, C., & Hauber, W. (2009). Modulation of behavior by expected rewardmagnitude depends on dopamine in the dorsomedial striatum. NeurotoxicityResearch, 15, 97–110.

Castane, A., Theobald, D. E., & Robbins, T. W. (2010). Selective lesions of thedorsomedial striatum impair serial spatial reversal learning in rats. BehaviouralBrain Research, 210, 74–83.

Chen, H., Jing, F. C., Li, C. L., Tu, P. F., Zheng, Q. S., & Wang, Z. H. (2007). Echinacosideprevents the striatal extracellular levels of monoamine neurotransmitters fromdiminution in 6-hydroxydopamine lesion rats. Journal of Ethnopharmacology,114, 285–289.

Clarke, H. F., Robbins, T. W., & Roberts, A. C. (2008). Lesions of the medial striatum inmonkeys produce perseverative impairments during reversal learning similarto those produced by lesions of the orbitofrontal cortex. Journal of Neuroscience,28, 10972–10982.

Colombo, P. J., Davis, H. P., & Volpe, B. T. (1989). Allocentric spatial and tactilememory impairments in rats with dorsal caudate lesions are affected bypreoperative behavioral training. Behavioral Neuroscience, 103, 1242–1250.

Cook, D., & Kesner, R. P. (1988). Caudate nucleus and memory for egocentriclocalization. Behavioral and Neural Biology, 49, 332–343.

Da Cunha, C., Wietzikoski, S., Wietzikoski, E. C., Miyoshi, E., Ferro, M. M., Anselmo-Franci, J. A., et al. (2003). Evidence for the substantia nigra pars compacta as anessential component of a memory system independent of the hippocampalmemory system. Neurobiology of Learning and Memory, 79, 236–242.

Darvas, M., & Palmiter, R. D. (2010). Restricting dopaminergic signaling to eitherdorsolateral or medial striatum facilitates cognition. Journal of Neuroscience, 30,1158–1165.

De Leonibus, E., Pascucci, T., Lopez, S., Oliverio, A., Amalric, M., & Mele, A. (2007).Spatial deficits in a mouse model of Parkinson disease. Psychopharmacology(Berlin), 194, 517–525.

Devan, B. D., McDonald, R. J., & White, N. M. (1999). Effects of medial and lateralcaudate-putamen lesions on place- and cue-guided behaviors in the watermaze: Relation to thigmotaxis. Behavioural Brain Research, 100, 5–14.

Devan, B. D., & White, N. M. (1999). Parallel information processing in the dorsalstriatum: Relation to hippocampal function. Journal of Neuroscience, 19,2789–2798.

Etienne, S., Guthrie, M., Gross, C., & Boraud, T. (2012). Dissociation of spatialnavigation strategies in primates by inhibiting various territories of thestriatum. Society for Neuroscience Abstracts, 38.

A.A. Braun et al. / Neurobiology of Learning and Memory 118 (2015) 55–63 63

Fouquet, C., Babayan, B. M., Watilliaux, A., Bontempi, B., Tobin, C., & Rondi-Reig, L.(2013). Complementary roles of the hippocampus and the dorsomedialstriatum during spatial and sequence-based navigation behavior. PLoS One, 8,e67232.

Garber, P. A. (2000). Evidence for the use of spatial, temporal, and social informationby some primate foragers. In S. B. Boinski & P. A. Garber (Eds.), On the move: Howand why animals travel in groups (pp. 261–298). Chicago: University of ChicagoPress.

Hemmerle, A. M., Dickerson, J. W., Herman, J. P., & Seroogy, K. B. (2014). Stressexacerbates experimental Parkinson’s disease. Molecular Psychiatry, 19,638–640.

Henze, C., Earl, C., Sautter, J., Schmidt, N., Themann, C., Hartmann, A., et al. (2005).Reactive oxidative and nitrogen species in the nigrostriatal system followingstriatal 6-hydroxydopamine lesion in rats. Brain Research, 1052, 97–104.

Herring, N. R., Schaefer, T. L., Gudelsky, G. A., Vorhees, C. V., & Williams, M. T. (2008).Effect of +-methamphetamine on path integration learning, novel objectrecognition, and neurotoxicity in rats. Psychopharmacology (Berlin), 199,637–650.

Iaria, G., Palermo, L., Committeri, G., & Barton, J. J. (2009). Age differences in theformation and use of cognitive maps. Behavioural Brain Research, 196, 187–191.

Jog, M. S., Kubota, Y., Connolly, C. I., Hillegaart, V., & Graybiel, A. M. (1999). Buildingneural representations of habits. Science, 286, 1745–1749.

Kim, H., Lee, D., & Jung, M. W. (2013). Signals for previous goal choice persist in thedorsomedial, but not dorsolateral striatum of rats. Journal of Neuroscience, 33,52–63.

Lavoie, A. M., & Mizumori, S. J. (1994). Spatial, movement- and reward-sensitivedischarge by medial ventral striatum neurons of rats. Brain Research, 638,157–168.

Lex, B., Sommer, S., & Hauber, W. (2011). The role of dopamine in the dorsomedialstriatum in place and response learning. Neuroscience, 172, 212–218.

Lindner, M. D., Cain, C. K., Plone, M. A., Frydel, B. R., Blaney, T. J., Emerich, D. F., et al.(1999). Incomplete nigrostriatal dopaminergic cell loss and partial reductions instriatal dopamine produce akinesia, rigidity, tremor and cognitive deficits inmiddle-aged rats. Behavioural Brain Research, 102, 1–16.

Livingstone, S. A., & Skelton, R. W. (2007). Virtual environment navigation tasks andthe assessment of cognitive deficits in individuals with brain injury. BehaviouralBrain Research, 185, 21–31.

McDonald, R. J., & White, N. M. (1994). Parallel information processing in the watermaze: Evidence for independent memory systems involving dorsal striatumand hippocampus. Behavioral and Neural Biology, 61, 260–270.

McGeorge, A. J., & Faull, R. L. (1989). The organization of the projection from thecerebral cortex to the striatum in the rat. Neuroscience, 29, 503–537.

Miyoshi, E., Wietzikoski, S., Camplessei, M., Silveira, R., Takahashi, R. N., & Da, C. C.(2002). Impaired learning in a spatial working memory version and in a cuedversion of the water maze in rats with MPTP-induced mesencephalicdopaminergic lesions. Brain Research Bulletin, 58, 41–47.

Mizumori, S. J., Canfield, J. G., & Yeshenko, O. (2005). Parallel and interrelated neuralsystems underlying adaptive navigation. Integrative and Comparative Biology, 45,547–554.

Mizumori, S. J., Puryear, C. B., & Martig, A. K. (2009). Basal ganglia contributions toadaptive navigation. Behavioural Brain Research, 199, 32–42.

Mizumori, S. J., Yeshenko, O., Gill, K. M., & Davis, D. M. (2004). Parallel processingacross neural systems: Implications for a multiple memory system hypothesis.Neurobiology of Learning and Memory, 82, 278–298.

Morris, R. G. (1981). Spatial localization does not require the presence of local cues.Learning and Motivation, 12, 239–260.

Mura, A., & Feldon, J. (2003). Spatial learning in rats is impaired after degenerationof the nigrostriatal dopaminergic system. Movement Disorders, 18, 860–871.

Aguiar, L. M., Macedo, D. S., Vasconcelos, S. M., Oliveira, A. A., de Sousa, F. C., & Viana,G. S. (2008). CSC, an adenosine A(2A) receptor antagonist and MAO B inhibitor,reverses behavior, monoamine neurotransmission, and amino acid alterationsin the 6-OHDA-lesioned rats. Brain Research, 1191, 192–199.

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

Packard, M. G., Hirsh, R., & White, N. M. (1989). Differential effects of fornix andcaudate nucleus lesions on two radial maze tasks: Evidence for multiplememory systems. Journal of Neuroscience, 9, 1465–1472.

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.

Palencia, C. A., & Ragozzino, M. E. (2005). The contribution of NMDA receptors in thedorsolateral striatum to egocentric response learning. Behavioral Neuroscience,119, 953–960.

Paxinos, G., Watson, C., Pennisi, M., & Topple, A. (1985). Bregma, lambda and theinteraural midpoint in stereotaxic surgery with rats of different sex, strain andweight. Journal of Neuroscience Methods, 13, 139–143.

Penner, M. R., & Mizumori, S. J. (2012). Neural systems analysis of decision makingduring goal-directed navigation. Progress in Neurobiology, 96, 96–135.

Pistell, P. J., Nelson, C. M., Miller, M. G., Spangler, E. L., Ingram, D. K., & Devan, B. D.(2009). Striatal lesions interfere with acquisition of a complex maze task in rats.Behavioural Brain Research, 197, 138–143.

Potegal, M. (1969). Role of the caudate nucleus in spatial orientation of rats. Journalof Comparative and Physiological Psychology, 69, 756–764.

Potegal, M. (1972). The caudate nucleus egocentric localization system. ActaNeurobiologiae Experimentalis (Warsaw), 32, 479–494.

Ragozzino, K. E., Leutgeb, S., & Mizumori, S. J. (2001). Dorsal striatal head directionand hippocampal place representations during spatial navigation. ExperimentalBrain Research, 139, 372–376.

Reading, P. J., Dunnett, S. B., & Robbins, T. W. (1991). Dissociable roles of the ventral,medial and lateral striatum on the acquisition and performance of a complexvisual stimulus-response habit. Behavioural Brain Research, 45, 147–161.

Rogers, R. D., Baunez, C., Everitt, B. J., & Robbins, T. W. (2001). Lesions of the medialand lateral striatum in the rat produce differential deficits in attentionalperformance. Behavioral Neuroscience, 115, 799–811.

Sanders, A. E., Holtzer, R., Lipton, R. B., Hall, C., & Verghese, J. (2008). Egocentric andexocentric navigation skills in older adults. Journals of Gerontology. Series A,Biological Sciences and Medical Sciences, 63, 1356–1363.

Sesack, S. R., Carr, D. B., Omelchenko, N., & Pinto, A. (2003). Anatomical substratesfor glutamate-dopamine interactions: Evidence for specificity of connectionsand extrasynaptic actions. Annals of the New York Academy of Sciences, 1003,36–52.

Skelton, M. R., Schaefer, T. L., Herring, N. R., Grace, C. E., Vorhees, C. V., & Williams,M. T. (2009). Comparison of the developmental effects of 5-methoxy-N, N-diisopropyltryptamine (Foxy) to (+/-)-3,4-methylenedioxymethamphetamine(ecstasy) in rats. Psychopharmacology (Berlin), 204, 287–297.

Tadaiesky, M. T., Dombrowski, P. A., Figueiredo, C. P., Cargnin-Ferreira, E., Da, C. C., &Takahashi, R. N. (2008). Emotional, cognitive and neurochemical alterations in apremotor stage model of Parkinson’s disease. Neuroscience, 156, 830–840.

Taube, J. S. (1998). Head direction cells and the neurophysiological basis for a senseof direction. Progress in Neurobiology, 55, 225–256.

Vorhees, C. V. (1987). Maze learning in rats: A comparison of performance in twowater mazes in progeny prenatally exposed to different doses of phenytoin.Neurotoxicology and Teratology, 9, 235–241.

Vorhees, C. V., Herring, N. R., Schaefer, T. L., Grace, C. E., Skelton, M. R., Johnson, H. L.,et al. (2008). Effects of neonatal (+)-methamphetamine on path integration andspatial learning in rats: Effects of dose and rearing conditions. InternationalJournal of Developmental Neuroscience, 26, 599–610.

Vorhees, C. V., Weisenburger, W. P., Acuff-Smith, K. D., & Minck, D. R. (1991). Ananalysis of factors influencing complex water maze learning in rats: Effects oftask complexity, path order and escape assistance on performancefollowing prenatal exposure to phenytoin. Neurotoxicology and Teratology, 13,213–222.

Weniger, G., & Irle, E. (2006). Posterior parahippocampal gyrus lesions in the humanimpair egocentric learning in a virtual environment. European Journal ofNeuroscience, 24, 2406–2414.

Whishaw, I. Q., & Dunnett, S. B. (1985). Dopamine depletion, stimulation orblockade in the rat disrupts spatial navigation and locomotion dependent uponbeacon or distal cues. Behavioural Brain Research, 18, 11–29.

Whishaw, I. Q., Mittleman, G., Bunch, S. T., & Dunnett, S. B. (1987). Impairments inthe acquisition, retention and selection of spatial navigation strategies aftermedial caudate-putamen lesions in rats. Behavioural Brain Research, 24,125–138.

White, N. M. (1997). Mnemonic functions of the basal ganglia. Current Opinion inNeurobiology, 7, 164–169.

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

Wiener, S. I. (1993). Spatial and behavioral correlates of striatal neurons in ratsperforming a self-initiated navigation task. Journal of Neuroscience, 13,3802–3817.

Williams, M. T., Herring, N. R., Schaefer, T. L., Skelton, M. R., Campbell, N. G., Lipton, J.W., et al. (2007). Alterations in body temperature, corticosterone, and behaviorfollowing the administration of 5-methoxy-diisopropyltryptamine (‘foxy’) toadult rats: A new drug of abuse. Neuropsychopharmacology, 32, 1404–1420.

Yin, H. H., & Knowlton, B. J. (2004). Contributions of striatal subregions to place andresponse learning. Learning and Memory, 11, 459–463.

Yin, H. H., & Knowlton, B. J. (2006). The role of the basal ganglia in habit formation.Nature Reviews Neuroscience, 7, 464–476.

Yin, H. H., Knowlton, B. J., & Balleine, B. W. (2004). Lesions of dorsolateral striatumpreserve outcome expectancy but disrupt habit formation in instrumentallearning. European Journal of Neuroscience, 19, 181–189.

Yin, H. H., Knowlton, B. J., & Balleine, B. W. (2006). Inactivation of dorsolateralstriatum enhances sensitivity to changes in the action-outcome contingency ininstrumental conditioning. Behavioural Brain Research, 166, 189–196.