A simple and reliable method for the positive identification of pavlovian conditioned cocaine effects in open-field behavior

Journal of Neuroscience Methods 73 (1997) 1–8

A simple and reliable method for the positive identification ofpavlovian conditioned cocaine effects in open-field behavior

Robert Carey *, Junmin Gui

VA Medical Center and SUNY Health Science Center, Syracuse, NY, USA

Received 9 September 1996; received in revised form 8 October 1996; accepted 13 October 1996

Abstract

The identification of the Pavlovian conditioning of the behavioral effects of cocaine using open-field behavior is oftenconfounded by the concurrent occurrence of behavioral habituation in control animals. Thus, differences in spontaneous activitybetween cocaine conditioned animals vs. control can be explained either by Pavlovian conditioning of the psychostimulant effectsof cocaine or by anti-habituation effects of cocaine. In a series of experiments we demonstrate that location of the animal withinthe open-field permits a positive identification of cocaine conditioning independent from habituation factors. In three separateexperiments, five daily paired 10 mg/kg cocaine treatments induced both increased locomotion as well as increased entries into thecentral zone in the open-field as compared with saline and cocaine unpaired control groups. Critically, in three experimentalreplications, animals which received the paired cocaine treatment exhibited statistically significant increases in central zone entriesin non-drug tests for conditioning both with respect to the saline and cocaine unpaired groups as well as to pre-conditioning levels.In contrast, the spontaneous locomotor behavior in the cocaine paired group on the conditioning test did not reliably increaseabove pre-conditioning levels but rather was only increased when compared with the reduced habituated activity levels in thesaline and cocaine unpaired groups. The conditioned increase in central zone entries induced by cocaine was equally robust at 4and 9 days post-conditioning but yet could be extinguished with repeated non-cocaine exposures to the open-field environment.© 1997 Elsevier Science B.V.

Keywords: Central zone; Cocaine; Habituation; Open-field; Pavlovian conditioning; Spontaneous activity

1. Introduction

For many years, Pavlovian conditioning of drugeffects was focused exclusively upon drug induced auto-nomic system responses (Pavlov, 1927; Eikelboom andStewart, 1982). With the emergence of behavioral phar-macology, the behavioral domain of Pavlovian drugconditioning has been expanded to include drug in-duced motoric and reward effects (Pickens andDougherty, 1971; Stewart and Eikelboom, 1987; Careyand Damianopoulos, 1994). One of the earliest reportsof the conditioning of drug induced locomotor stimula-

tion was that of amphetamine induced hyperlocomo-tion (Tilson and Rech, 1973). Many subsequent reportshave confirmed this basic observation and extended itto other psychostimulant drugs (Beninger and Hahn,1983; Herz and Beninger, 1987; Barr et al., 1983; Stew-art and Druhan, 1993). In Pavlovian conditioning,however, the conditioned stimulus, in theory, is as-sumed to be neutral vis-a-vis the elicitation of the targetdrug response. In conditioning locomotor stimulationwith psychostimulant drugs, the animals are typicallyplaced in an open-field environment wherein the situa-tional cues elicit a variety of spontaneous exploratorybehaviors (e.g. rearing, locomotion, etc.). When ani-mals are treated with a moderate dose of psychostimu-lant drug, the effect on behavior is not a completelyunique set of behaviors but rather a modulation of

* Corresponding author. Present address: 800 Irving Avenue, Re-search (151), VA Medical Center, Syracuse, NY 13 210, USA. Tel.:+1 315 4767461 (ext. 2824); fax: +1 315 4765348.

0165-0270/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved.PII S 0 1 6 5 -0270 (96 )02203 -0

R. Carey, J. Gui / Journal of Neuroscience Methods 73 (1997) 1–82

ongoing behavior (Damianopoulos and Carey, 1994;Stewart and Vezina, 1988). Thus, the unconditioneddrug stimulus enhances the unconditioned responseelicited by the open-field environment which also servesas the conditioned stimulus. Furthermore, if animalsare repeatedly placed in the same open-field environ-ment without drug, a marked behavioral habituation tothe test environment can develop. The significance ofthese aspects of drug conditioned locomotion for aPavlovian conditioning framework is that the open-fieldserving as a conditioned stimulus (CS) relative to thedrug unconditioned stimulus (UCS) is not neutral andin addition, it undergoes habituation.

This CS-UCS overlap as well as habituation factorscreate a major problem of interpretation for studies ofthe drug conditioning of locomotor behavior. The ha-bituation issue is seemingly intractable. Specifically, theunpaired drug treatment group is exposed to the testenvironment in the non-drug state and therefore, habit-uates to the test environment. In contrast, the pairedtreatment group experiences the test environment in thedrug state and the drug treatment may block habitua-tion. Furthermore, it is also possible for habituation tooccur in the drug state but not transfer to the non-drugstate. In either circumstance, the paired animals wouldbe less habituated to the test environment than theunpaired animals. As a consequence of this possiblehabituation differential, the paired animals would bemore active than the unpaired animals in the test forconditioning. Thus, a difference in activity betweenpaired and unpaired groups obtained in a non-drugconditioning test using the paired/unpaired designcould be the result of either stimulant drug conditioningor it could be due to differential habituation effects.

In a previous paper, we devised a complex paradigm(Damianopoulos and Carey, 1994) in which potentialdrug induced anti-habituation effects upon locomotorbehavior could be differentiated from conditioned stim-ulant drug effects. In the present study, we report asimple behavioral protocol for detecting cocaine behav-ioral conditioning in an open-field environment inwhich habituation effects are circumvented. Instead ofusing locomotion distance as the response variable, weused the frequency of an animal’s penetration of thecenter zone of the open-field as the response measure.In general, rats exhibit thigmotaxis in an open-fieldenvironment and only infrequently enter the centerzone of the field (Schwarting et al., 1993). In the case ofanimals treated with a modest dose of cocaine, weobserved that animals entered the center zone morefrequently. Since entrance into the center zone is abehavior which seemingly would not undergo habitua-tion, we evaluated whether this behavioral measurewhich is increased by cocaine would serve as an indica-tor response for the occurrence of cocaine conditioning.Our studies indicate that entrance into the center zone

does not habituate and that cocaine induces a reliableconditioned center zone response which can be extin-guished.

2. Animals

Naive male Sprague-Dawley rats from TaconicFarms (Germantown, NY), 6 months old and weighingapproximately 500 g at the start of the experimentswere used. Upon arrival, the animals were housed inindividual 25×17×17 cm wire mesh cages in a cli-mate-controlled room at 2292°C with a 12 h dark/light cycle. During the 1st week after arrival, allanimals were handled and weighed daily for 7 days.During the second week the animals received threeinjections (i.p.) of saline (1 ml/kg) in order to acclimatethe animals to the injection procedure. All experimentsoccurred during the 12 h light cycle.

3. Drugs

Cocaine hydrochloride (Mallinckrodt SpecialtyChemical, St. Louis, MO) was dissolved in sterile dis-tilled H2O in a concentration of 10 mg/ml (all injectionswere performed i.p.).

4. Apparatus

The layout of the test chamber is illustrated in Fig. 1.All of the behavioral tests were conducted in thissquare open-field compartment which was 60×60×45cm. A closed-circuit video camera (RCA TC7011U)was mounted 50 cm above the open-field box. Allsignals were analyzed by a video tracking system, theVideomex-V from Columbus Instruments (Columbus,OH) and the data was imported into a PC compatiblecomputer. The walls of the chamber were white and thefloor of the open-field box was covered by plain whitepaper which was changed after each animal. Ambientwhite noise (80 dB) was provided by an audio tapeplayer and was turned on immediately prior to place-ment of the animal in the test chamber and turned offupon removal from the test chamber. Testing was con-ducted under conditions of red light illumination toenhance the contrast between the subject and back-ground and to reduce the animal’s shadow. In order tobe sure that penetration of the central zone had oc-curred, the animal’s head was blackened by a markerpen and the camera only tracked this feature of therat’s body. A central zone (CZ) comprising one ninth ofthe floor area was programmed to be monitored by thevideo analyzer independently from the rest of the open-field. During each session, data was collected every 2.5

R. Carey, J. Gui / Journal of Neuroscience Methods 73 (1997) 1–8 3

min by the computer. A dot matrix printer (EpsonFX-286e) was placed outside the test room and wasconnected to the image analyzer by a parallel cable andthe computer screen tracings of the animal’s movementwere printed out every 2.5 min. The complete testprocedure was conducted automatically without thepresence of the experimenter in the test room. In addi-tion, a VHS VCR was also connected to the camera forthe purpose of recording supplementary behavioraldata and providing the ability for one to review andre-input the video tape signal to the image analyzer incase of any possible malfunction of either the analyzeror the computer during the experiments.

5. Design and procedure

In the first phase of the experiment (pre-condition-ing) each animal received a saline injection followed bya 10 min test in the open-field environment. This testserved two purposes. It provided a basis for which tomatch treatment groups on the dependent variables oflocomotion distance and central zone (CZ) entries.Secondly, it provided a measure of the animals’ initialresponse to the test environment so that it could bedetermined subsequently, whether the conditioning pro-cedure had elevated an animal’s response level above itsnon-habituated response to the environment.

Two days after completion of this behavioral baselinetesting the animals were subdivided into three groupswith statistically equivalent levels of locomotion andCZ entry scores. One group received 0.9% sterilesodium chloride (S), one group cocaine 10 mg/kg be-fore testing (Coc-P) and the third group also receivedcocaine 10 mg/kg 30 min after testing (Coc-UP). Allgroups received five successive daily tests of 20 min induration (conditioning phase). The S (n=7) and Coc-P(n=7) groups were injected immediately before place-ment into the test environment, whereas the Coc-UP(n=7) group were injected in their homecages 30 minafter completion of testing.

In separate experiments this same protocol was re-peated three times. Three replications were performedin order to determine (a) a durability of the condition-ing by testing for conditioning at 4 and 9 days post-conditioning treatment and (b) to determine if theobserved behavioral response would undergo extinc-tion. As a consequence of these considerations, onereplication involved a non-drug post-conditioning testfor conditioning at 4 days after completion of theconditioning treatment, a second replication involved atest for conditioning at 9 days post-treatment and thethird replication involved conducting a series of sixnon-drug conditioning tests for extinction on days 4–9post treatment. In all of these non-drug tests for condi-tioning, animals received saline injections immediatelyprior to placement in the test environment.

6. Statistical calculations

Treatment effects were statistically evaluated by one-way and two-way ANOVA for locomotion distance(meters) and CZ entries. If statistically significant treat-ment effects were obtained with two-way ANOVA thenspecific effects were evaluated using one-way ANOVA.In order to make group comparisons, post-hoc Dun-can’s multiple range tests were performed since this testincorporates the variability at all treatments. Pairedt-tests were used when applicable. A p value of lessthan 0.05 was used as the criterion for statistical signifi-cance.

7. Results

The initial phase of this experiment entailed an as-sessment of habituation of spontaneous behavior in theopen-field using locomotion distance and CZ entries asdependent variables. Accordingly, 20 animals weregiven five daily 20 min tests in the open-field andlocomotion distance and CZ entries were statisticallyevaluated by One Way ANOVA with sessions as theindependent variable. For locomotion distance there

Fig. 1. A diagram of the open-field test environment and the videotracking system. The broken lines with the open-field identify thecentral zone. The central zone lines are not physically marked on thetest box floor but rather the central zone is only differentiated fromthe rest of the open-field by the video analyzer.


was a statistically significant session effect (F(4,19)=4.9; pB0.01) but for CZ entries there was not a statis-tically significant session effect (F(4,19)=1.6; p\0.05).To further analyze this session effect, the first and lastsessions were compared using paired t tests with loco-motion distance and CZ entries as the dependent vari-ables. For locomotion distance, the distance traversedin session 5 was significantly less than session 1 (t=4.1;pB0.001) whereas for CZ entries, there was no statisti-cally significant difference between session 1 and 5(t=1.7; p\0.05). Thus, this analysis indicated thatlocomotion distance exhibited the typical habituationeffect associated with repeated tests in an open-fieldwhereas, central zone entries did not change statisticallyacross five 20 min test sessions. These findings pointedto the potential utility of a CZ entry measure as adependent behavioral variable with low sensitivity tohabituation processes.

In each of the three replications of the conditioningexperiments, the Coc-P treatment had a highly statisti-cally significant effect upon locomotion distance andcenter zone entries during the 5-day cocaine/saline con-ditioning treatment phase of the experiment. In each ofthe three replications of the conditioning procedure thestatistical analyses using two-way ANOVA methodsindicated that the treatment effects were highly signifi-cant statistically df(2,18)F=132.5, 278.4 and 190.3,respectively for distance and F=104.3, 40.3 and 208.2respectively for CZ entries. All of the F values werestatistically significant at p less than 0.001. Importantly,there were no statistically significant session or treat-ment by session interaction effects (p\0.05). Follow-up analyses using One-Way ANOVA procedures toexamine specific treatment effects revealed that theCoc-P group differed from the Coc-UP and S groups atthe p less than 0.001 level in each of the three replica-tions for both locomotion distance and CZ entries asdependent variables. Thus, the 10 mg/kg cocaine treat-ments generated a robust unconditioned drug responsewhich was manifested both in increased locomotoractivity and CZ entries.

Fig. 2 presents the results of the initial conditioningexperiment. As can be seen in Fig. 2, the number of CZentries in the three treatment groups prior to the condi-tioning procedure were virtually equivalent (F(2,18)=0.06; p\0.05). In contrast, following the conditioningtreatment the paired cocaine group exhibited a signifi-cantly higher number of CZ entries (F(2,18)=9.9; pB0.001). The increased number of CZ entries exhibitedby the paired cocaine group was not only greater thanthe two control groups but it was also higher than itsown preconditioning baseline using a paired t-test anal-ysis (t=2.73; pB0.02). In general, a similar result wasobtained when locomotion distance was used as thedependent variable. Prior to conditioning the threegroups were statistically indistinguishable (F(2,18)=

Fig. 2. The means and SEMs of central zone entries for the threetreatment groups S, Coc-P and Coc-UP. The upper graph presentsthe entry scores for each group in the pre-conditioning treatment testand the lower graph shows the entry scores for the groups on theirnon-drug test for conditioning which was conducted 4 days aftercompletion of the conditioning treatments. * Indicates that the Coc-Pgroup had a statistically significant (pB0.01) increased level of CZentries as compared with each of the two control groups.

0.15; p\0.05). Following the conditioning procedure,however, there was a statistically significant treatmenteffect (F(2,18)=7.2; pB0.01) with the paired cocainegroups having a higher overall level of activity than thetwo control groups. In contrast to CZ entries, however,when the locomotion distance of the paired group onthe conditioning test was compared with its pre-condi-tioning activity level then the difference was not statisti-cally significant (t=1.3; p\0.05). Fig. 3 presents acomparison of the pre- vs. post-conditioning activitylevel of the paired cocaine groups. As is evident in Fig.3, the paired group did not increase in activity above itspre-conditioning level. A comparison of these two testsyielded F(1,6)=0.05 and a p value of greater than 0.05.In addition to this absence of an increase of locomotionactivity above the pre-conditioning level, both of thecontrol groups exhibited statistically significant declinesin activity from the pre-conditioning to their post-con-ditioning level with t values equal to 5.5 and 3.8,respectively for the S and Coc-UP groups which werestatistically significant at the p less than 0.01 level. Interms of habituation, the S and Coc-UP groups exhib-


ited habituation but the Coc-P group did not. Thus, onthe basis of locomotion data it would not be possible todifferentiate between a conditioned cocaine effect vs. acocaine anti-habituation effect. The CZ entry data,however, provide positive support for a cocaine condi-tioned effect. In order to convey a more direct represen-tation of the cocaine conditioned effect on CZ entry,Fig. 4 presents representative tracings of the actualpaths traversed by animals from each treatment groupduring the first interval in the test for conditioning. Theprimary criterion used in the selection of these tracingswas that the locomotion distances be equivalent. As canbe seen in Fig. 4, the Coc-P animal distributes more ofits locomotion in the central aspect of the open-fieldrather than simply being more active than the otheranimals.

The second major question addressed in relation tothe use of CZ entry as an index of cocaine conditioningwas whether the conditioned response would persist fora substantial interval after the completion of the condi-tioning procedure. Fig. 5 presents the CZ entries forthree treatment groups before and after the condition-ing treatment when the test for conditioning was de-layed for 9 days post-conditioning. As can be seen inFig. 5, the results were essentially the same as for thefirst experiment. Prior to conditioning there were nostatistically significant differences among groups(F(2,18)=0.14; p\0.05) but after conditioning therewas a significant group effect (F(2,18)=9.87; pB0.01)with the Coc-P group having a higher level of CZentries than the other two groups. When a similarstatistical analysis was conducted using locomotion dis-tance as the dependent variable, however, the groupdifferences did not reach statistical significance(F(2,18)=2.76; p\0.05), perhaps reflecting a partialhabituation loss in the control groups as a result of aperiod of non testing. This finding further supports the

Fig. 4. Tracings of the actual routes traveled in the first 2.5 min of thepost-conditioning test for representative animals in each treatmentgroup matched for locomotion distance.

Fig. 3. The means and SEMs of locomotion distance for the Coc-Pgroup on the pre-conditioning test and the conditioning test. Thedistance scores on the two tests were not statistically different (p\0.05 as determined by a paired t-test).

value of CZ entry as a reliable behavioral indicator forcocaine conditioning. While the preceding studies cer-tainly are supportive of a conditioning basis for thehigher level of CZ entries in the paired cocaine group,a key dimension of conditioned behavior is that itundergoes extinction when tests are given in the pres-ence of the conditioned stimulus but without the un-


conditioned stimulus. It can be seen in Fig. 6, in thethird experimental replication, that six extinction trialseliminated the conditioned cocaine central zone re-sponse. The statistical analyses indicated that thegroups had statistically equivalent frequencies of CZentries prior to conditioning (F(1,18)=0.07; p\0.05)but on the first test for conditioning, the paired cocainegroup had a statistically significant higher incidence ofCZ entry level than the S group (F(1,18)=8.1; pB0.01). On the sixth test for conditioning following fiveextinction trials, however, the groups were again statis-tically equivalent (F(1,18)=1.2; p\0.05). Thus, thisthird experiment shows that the increase in CZ entriesinduced by cocaine can be extinguished by repeatednon-drug exposures to the test environment.

8. Discussion

The Pavlovian conditioning of psychostimulant drugeffects is a well established clinical phenomenon (Kran-zler and Bauer, 1989; McLellan et al., 1988; Muntaneret al., 1989; Newlin, 1992; O’Brien et al., 1992a,b,1993). At the basic science level, however, the study of

Fig. 6. Means and SEMs of central zone entries for the S and Coc-Pgroups in the extinction experiment. The upper graph presents theresults of the pre-conditioning test, the middle graph contains theresults of the first extinction test conducted 4 days after completion ofthe conditioning treatments and the lower graph presents the resultsof the sixth extinction test conducted 9 days after completion of theconditioning treatments. * Denotes pB0.01 statistical significance forthe higher level of CZ entries for the Coc-P group vs. the S group.

Fig. 5. Means and SEMs of central zone entries for the S, Coc-P andCoc-UP groups on the pre-conditioning test and the post-condition-ing test when the post-conditioning test was delayed for 9 days afterthe completion of the conditioning treatment. * Denotes pB0.01 forthe Coc-P group vs. the S and Coc-UP groups.

the Pavlovian conditioning of psychostimulant drugeffects has been investigated with limited success. Withdrugs such as cocaine it is a simple matter to identifythe drug response in terms of an increase in sponta-neous locomotor behavior. As indicated earlier, the useof spontaneous locomotor behavior as an index ofconditioning is complicated by the occurrence of habit-uation factors with repeated testing which impacts di-rectly upon the behavior used as an index ofconditioning. One way to circumvent this issue is toselect behavioral responses induced by the drug whicheither do not occur spontaneously or do not exhibithabituation. If drug induced behaviors are used whichdo not occur spontaneously then, one frequently mustuse fairly high doses of drug. Typically, such behavioralresponses need trained observers which are unwieldyand are subject to the problems associated with experi-


menter based judgements (Schiff, 1982). In addition,negative findings for conditioning can be misleadingbecause, with high doses of a drug treatment, there canbe state dependent drug effects which mask conditionedeffects when testing is conducted in the non-drug state(Damianopoulos and Carey, 1993).

The present method for studying cocaine conditionedbehavior obviates many of the difficulties in measure-ment as well as interpretation of the cocaine condi-tioned behavior. Critically, the conditioned behavioreffects are reliable, persistent and independent of habit-uation effects. Furthermore, the conditioned behaviorcan be extinguished. Thus, the present method offersopportunities to study in an animal behavior model anumber of basic variables pertinent to conditioning asthey impact upon cocaine conditioning. In addition,since the dependent variable is linked to a qualitativecomponent of locomotion namely, location within thetest environment, it becomes possible to investigatepharmacology of cocaine conditioned behavior usingdrugs which do not grossly impair locomotor behavior.

Beyond serving as a tool for cocaine conditioning,the finding that cocaine substantially increases a ratsreadiness to enter the center zone of an open-field isalso of interest in light of suggestions of a relationshipbetween center zone activity and emotionality (Rothand Katz, 1980). Indeed, there is evidence supportive oftwo seemingly contradictory factors which can enhancecentral zone activity in rodents. Once facilitator ofcentral zone activity is stress (Roth and Katz, 1980). Infact, it has been shown that central zone activity can beincreased either by behavioral stress induction usingrestraint or footshock prior to placement in the open-field (Lee et al., 1988) or by the administration of stressrelated hormones such as corticosterone (Oitzl and deKloet, 1992). In direct contrast to this evidence forstress induced center zone activity, there is an extensivepharmacological literature which shows that drugs withanxiolytic activity enhance center zone entries (Chopinand Briley, 1987; Stefanski et al., 1992; Kostowski etal., 1989). In the specific case of cocaine, the most likelyfactor which accounts for the increased central zoneactivity induced by cocaine relates to its physiologicalaction as a stressor (Borowsky and Kuhn, 1991; Kali-vas and Duffy, 1989; Moldow and Fischman, 1987;Sarnyai et al., 1992; Sorg, 1992). In addition to positiveevidence for cocaine stressor effects it is also the casethat cocaine induces anxiogenic rather than anxiolyticeffects (Yang et al., 1992). While it could be argued thatthe increased central zone activity induced by cocaine issimply a manifestation of cocaine induced hyperloco-motion, such a possibility fails to acknowledge thathyperlocomotion is most readily expressed as increasedlocomotion in the animals preferred path along theperiphery (Schwarting et al., 1993). Possibly then, thestressor effects of cocaine act permissively to create the

opportunity to engage in locomotion away from theperiphery. The present observation of cocaine condi-tioned center zone activity may be consistent withcocaine stressor effects but, on the other hand, thisconditioned effect would appear inconsistent with co-caine place preference conditioning (Nomikos andSpyraki, 1988; Spyraki et al., 1987; Koob, 1992). Un-fortunately, place preference studies have not incorpo-rated measures of central zone activity so that itremains to be determined whether place preferenceinduced by cocaine is also accompanied by an increasein central zone activity. In the context of cocaine in-duced reward effects, it is relevant to observe that whilestress hormones can be activated by aversive events(e.g. footshock) the hormones themselves do not neces-sarily have aversive properties. In a number of studies ithas been shown that administration of stress relatedhormones, such as corticosterone, can have rewardeffects and also can enhance the reward effects of drugssuch as cocaine (Deroche et al., 1993; Piazza et al.,1993, 1994; Marinelli et al., 1994; Goeders and Guerin,1994). In any effort to integrate cocaine stressor effectsand cocaine reward effects, it is useful to take intoaccount the observations of (Schachter and Singer,1962). In a series of experiments, they administeredepinephrine to human subjects and evaluated their sub-jective experience. Critically, it was shown thatepinephrine could induce positive or negative effectsdepending upon the context in which the epinephrinewas experienced. In extending this context based analy-sis to corticosterone, it can be postulated that corticos-terone has a non-specific activating effect which canenhance aversive properties of events such as foot-shocks or enhance the positive sensation seeking quali-ties associated with situations related to the explorationof a novel environment. Thus, the non-specific activa-tion by a hormone may obtain its hedonic sign from thecontext in which it is experienced.

Acknowledgements

This research was supported by NIDA grantRO1DA05366-10.

References

Barr, G.A., Sharpless, N.S., Cooper, S., Schiff, S.R., Paredes, W. andBridger, W.H. (1983) Classical conditioning, decay and extinctionof cocaine-induced hyperactivity and stereotypy, Life Sci., 33:1341–1351.

Beninger, R.J. and Hahn, B.L. (1983) Pimozide blocks establishmentbut not expression of amphetamine-produced environment-spe-cific conditioning, Science, 220: 1304–1306.

Borowsky, B. and Kuhn, C.M. (1991) Monoamine mediation ofcocaine-induced hypothalamo-pituitary-adrenal activation, J.Pharmacol. Exp. Ther., 256: 204–210.


Carey, R.J. and Damianopoulos, E.N. (1994) Conditioned cocaineinduced hyperactivity: An association with increased medial pre-frontal cortex serotonin, Behav. Brain Res., 62: 177–185.

Chopin, P. and Briley, M. (1987) Animal models of anxiety: the effectof compounds that modify 5-HT neurotransmission, Trends Phar-macol. Sci., 8: 383–388.

Damianopoulos, E.N. and Carey, R.J. (1993) Apomorphine sensitiza-tion effects: Evidence for environmentally contingent behavioralreorganization process, Pharmacol. Biochem. Behav., 45: 655–663.

Damianopoulos, E.N. and Carey, R.J. (1994) A new method to assessPavlovian conditioning of psychostimulant drug effects, J. Neu-rosci. Methods, 53: 7–17.

Deroche, V., Piazza, P.V., Deminiere, J.M., Le Moal, M. and Simon,H. (1993) Rats orally self-administer corticosterone, Brain Res.,622: 315–320.

Eikelboom, R. and Stewart, J. (1982) Conditioning of drug-inducedphysiological responses, Psychol. Rev., 89: 507–528.

Goeders, N.E. and Guerin, G.F. (1994) Non-contingent electric foot-shock facilitates the acquisition of intravenous cocaine self-admin-istration in rats, Psychopharmacology (Berlin), 114: 63–70.

Herz, R.S. and Beninger, R.J. (1987) Comparison of the ability of(+ )-amphetamine and caffeine to produce environment-specificconditioning, Psychopharmacology (Berlin), 92: 365–370.

Kalivas, P.W. and Duffy, P. (1989) Similar effects of daily cocaineand stress on mesocorticolimbic dopamine neurotransmission inthe rat, Biol. Psychiat., 25: 913–928.

Koob, G.F. (1992) Drugs of abuse: anatomy, pharmacology andfunction of reward pathways, Trends Pharmacol. Sci., 13: 177–184.

Kostowski, W., Plaznik, A. and Stefanski, R. (1989) Intra-hippocam-pal buspirone in animal models of anxiety, Eur. J. Pharmacol.,168: 393–396.

Kranzler, H. and Bauer, L. (1989) A laboratory procedure forevaluation of pharmacotherapy for cocaine dependence, NIDARes. Mono. Series, 95: 324–325.

Lee, E.H.Y., Lin, Y.P. and Yin, T.H. (1988) Effects of lateral andmedial septal lesions on various activity and reactivity measures inrats, Physiol. Behav., 42: 97–102.

Marinelli, M., Piazza, P.V., Deroche, V., Maccari, S., Le Moal, M.and Simon, H. (1994) Corticosterone circadian secretion differen-tially facilitates dopamine-mediated psychomotor effect of cocaineand morphine, J. Neurosci., 14: 2724–2731.

McLellan, A.T., Childress, A.R., Ehrman, R. and O’Brien, C.P.(1988) Opiate and cocaine related stimuli elicit craving and physi-ological responses in drug-abuse patients, presented at the meet-ing of the Society for Stimulus Properties of Drugs, Cape Cod,MA: June 28.

Moldow, R.L. and Fischman, A.J. (1987) Cocaine induced secretionof ACTH, beta-endorphin, and corticosterone, Peptides, 8: 819–822.

Muntaner, C., Cascella, N.G., Kumor, K.M., Nagoshi, C., Herning,R. and Jaffe, J. (1989) Placebo responses to cocaine administra-tion in humans: effects of prior administrations and verbal in-structions, Psychopharmacology (Berlin), 99: 282–286.

Newlin, D.B. (1992) A comparison of drug conditioning and cravingfor alcohol and cocaine. In M. Galanter (Ed.), Recent Develop-ments in Alcoholism, Vol. 10: Alcohol and Cocaine: Similaritiesand Differences, Plenum Press, New York, pp. 147–164.

Nomikos, G.G. and Spyraki, C. (1988) Cocaine-induced place condi-tioning: importance of route administration and other proceduralvariables, Psychopharmacology (Berlin), 94: 119–125.

O’Brien, C.P., Childress, A.R., McLellan, A.T. and Ehrman, B.(1992) Classical conditioning in drug-dependent humans, Ann.NY. Acad. Sci., 654: 400–415.

O’Brien, C.P., Childress, A.R., McLellan, A.T. and Ehrman, R.(1992) A learning model of addiction. In C.P. O’Brien and J.H.Jaffe (Eds.), Addictive States, Raven Press, New York, pp. 157–177.

O’Brien, C.P., Childress, A.R., McLellan, A.T. and Erhman, R.(1993) Developing treatment that address classical conditioning,NIDA Res. Mono. Series, 135: 71–91.

Oitzl, M.S. and de Kloet, E.R. (1992) Corticosterone modulatesexploration via central mineralocorticoid receptors, Soc. Neu-rosci. Abs., 18: 718.

Pavlov, I.P. (1927) Conditioned Reflex, Oxford University Press,London, pp. 35.

Piazza, P.V., Deroche, V., Deminiere, J.M., Maccari, S., Le Moal, M.and Simon, H. (1993) Corticosterone in the range of stress-in-duced levels possesses reinforcing properties: implications forsensation-seeking behaviors, Proc. Natl. Acad. Sci. USA, 90:11 738–11 742.

Piazza, P.V., Marinelli, M., Jodogne, C. and Deroche, V. (1994)Inhibition of corticosterone synthesis by Metyrapone decreasescocaine-induced locomotion and relapse of cocaine self-adminis-tration, Brain Res., 658: 259–264.

Pickens, R. and Dougherty, J. (1971) Conditioning of the activityeffects of drugs. In T. Thompson and C. Schuster (Eds.), StimulusProperties of Drugs, Appleton-Century-Crofts, New York, pp.39–50.

Roth, K.A. and Katz, R.J. (1980) Stress, behavioral arousal andopen-field activity: A re-examination of emotionality in the rat,Neurosci. Biobehav. Rev., 3: 247–263.

Sarnyai, Z., Biro, E., Penke, B. and Telegdy, G. (1992) The cocaine-induced elevation of plasma corticosterone is mediated by endoge-nous corticotropin-releasing factor (CRF) in rats, Brain Res., 589:154–156.

Schachter, S. and Singer, J.E. (1962) Cognitive social and physiologi-cal determinants of emotion state, Psychol. Rev., 69: 379–399.

Schiff, S.R. (1982) Conditioned dopaminergic activity, Biol. Psy-chiat., 17: 135–154.

Schwarting, R.K.W., Golderberg, R., Steiner, H., Fornaguera, J. andHuston, J.P. (1993) A video image analyzing system for open-fieldbehavior in the rat focusing on behavioral asymmetries, J. Neu-rosci. Methods, 49: 199–210.

Sorg, B.A. (1992) Mesocorticolimbic dopamine systems: cross-sensi-tization between stress and cocaine, Ann. NY. Acad. Sci., 654:136–144.

Spyraki, C., Nomikos, G.G. and Varonos, D.D. (1987) Intravenouscocaine-induced place preference attenuation by haloperidol, Be-hav. Brain Res., 26: 57–62.

Stefanski, R., Palejko, W., Kostowski, W. and Plaznik, A. (1992) Thecomparison of benzodiazepine derivatives and serotonergic ago-nists and antagonists in two animal models of anxiety, Neu-ropharmacology, 31: 1251–1258.

Stewart, J. and Druhan, J.P. (1993) Development of both condition-ing and sensitization of the behavioral activating effects of am-phetamine is blocked by the non-competitive NMDA receptorantagonist, MK-801, Psychopharmacology (Berlin), 110: 125–132.

Stewart, J. and Eikelboom, R. (1987) Conditioned drug effects,Plenum Press, New York, pp. 1.

Stewart, J. and Vezina, P. (1988) Conditioning and behavioral sensi-tization. In C.R. Barnes and P.W. Kalivas (Eds.), Sensitization inthe Nervous System, Telford Press, Caldwell, NJ, pp. 207–224.

Tilson, H.A. and Rech, R.A. (1973) Conditioned Drug effects andabsence of tolerance to d-amphetamine induced motor activity,Pharmacol. Biochem. Behav., 1: 149–153.

Yang, X.M., Gorman, A.L., Dunn, A.J. and Goeders, N.E. (1992)Anxiogenic effects of acute and chronic cocaine administration:neurochemical and behavioral studies, Pharmacol. Biochem. Be-hav., 41: 643–650.

.

.

Documents

A simple and reliable method for the positive identification of pavlovian conditioned cocaine effects in open-field behavior