External incentives and internal states guide goal-directed behavior via the differential recruitment of the nucleus accumbens and the medial prefrontal cortex.

Goal-directed behavior is governed by internal physiological states and external incentives present in the environment (e.g. hunger and food). While the role of the mesocorticolimbic dopamine (DA) system in behavior guided by environmental incentives has been well studied, the effect of relevant physiological states on the function of this system is less understood. The current study examined the role of the medial prefrontal cortex (mPFC) and the nucleus accumbens (NAcc) in the kind of food-reinforced behaviors known to be sensitive to the internal state produced by food deprivation conditions. Operant lever-press reinforced on fixed ratio 1 (FR1) and progressive ratio (PR) schedules was tested after temporary inactivation of, or DA receptor blockade in, the prelimbic mPFC or NAcc core of rats with differing levels of food deprivation (0, 12 and 36-h). Food deprivation increased PR breakpoints, as well as the number of lever-presses emitted on the FR1 schedule. Both temporary inactivation and DA blockade of NAcc reduced breakpoints across deprivation conditions, while temporary inactivation and DA blockade of mPFC reduced breakpoints only in food-deprived rats. Neither manipulation of mPFC and NAcc had any effect on behavior reinforced on the FR1 schedule. Thus, mPFC and NAcc were differentially relevant to the behaviors tested-NAcc was recruited when the behavioral cost per reinforcer was rising or high regardless of food deprivation conditions, while mPFC was recruited when food-deprived animals behaved through periods of sparse reinforcement density in order to maximize available gain.

Effects of food deprivation on goal-directed behavior, spontaneous locomotion, and c-Fos immunoreactivity in the amygdala.

Previous work in our laboratory has shown that food deprivation and food presentation produce different patterns of neuronal activity (as measured by c-Fos immunoreactivity) in the medial prefrontal cortex and nucleus accumbens of rats. Since the amygdala has been implicated in both motivational and reinforcement processes and has neuronal connections to both the prefrontal cortex and nucleus accumbens, it was of interest to assess amygdaloid c-Fos immunoreactivity during similar manipulations of food deprivation and presentation. In the current study, c-Fos counts in both basolateral and central amygdalar nuclei were observed to increase in rats 12- and 36-h food deprived (relative to 0-h controls)-an effect reversed by the presentation of either a small or large meal (2.5 or 20g of food). In another experiment, rats working on a progressive ratio schedule of reinforcement exhibited elevated break-points as a function of food deprivation, a result consistent with the view that the feeding manipulations increased the subjects' level of motivation. In contrast, food deprivation reduced the spontaneous locomotor activity of rats, presumably as a result of an inherent energy-conservation strategy when no food is readily available. These data suggest that the state of food deprivation is associated with: (a) enhanced behavioral output only when food is attainable (increased goal-directed behavior, but decreased spontaneous activity), and (b) increased synaptic engagement in neuronal circuits involved in affective valuation and related decision-making (increased c-Fos counts in the amygdala).

Alcohol consumption is preferred to water in rats pretreated with intravenous cocaine.

Clinical and anecdotal reports suggest a high incidence of alcohol administration during cocaine use, potentially as a means to diminish the aversive effects of cocaine that follow the initial positive drug effects. We have previously shown that in the operant runway, oral ethanol significantly reduces the approach-avoidance retreats that develop in response to IV cocaine. The current study was intended to test whether rats given the same dose of IV cocaine administered in our previous study would choose to consume ethanol rather than water in a two bottle choice paradigm. We also examined whether significant serum levels of the psychoactive cocaine metabolite, cocaethylene, were found in our animals that may account for the preference for ethanol. Animals pretreated with cocaine drank significantly more ethanol than did animals pretreated with saline. There were no measurable levels of cocaethylene at 10 or 40 min post-cocaine and extremely low values at the 20-min time point, indicating that cocaethylene formation does not reinforce the co-administration of cocaine and alcohol in rats. These data demonstrate that the presence of cocaine serves as a primary factor in the preference for alcohol in thirsty rats, potentially to reduce the well-documented negative/anxiogenic properties of cocaine.

Ethanol consumption reduces the adverse consequences of self-administered intravenous cocaine in rats.

RATIONALE: Human drug users report that the initial positive effects of cocaine are followed by a dysphoric state characterized by anxiety and drug-craving. As a means of presumably attenuating these negative aftereffects, 50-90% of cocaine users choose to co-administer ethanol during cocaine binges. This co-administration reportedly prolongs the "high" and diminishes the "low" associated with cocaine use. OBJECTIVE: The current study was intended to assess whether this phenomenon could be modeled in the animal laboratory. We have previously shown that animals running a straight alley for an intravenous cocaine reward develop a unique approach-avoidance "conflict" behavior that is characterized by stop and retreat behaviors as the subjects approach the goal box. The retreats are thought to reflect the concurrent positive (reward) and negative (anxiety) associations with the goal box and can be dose-dependently reduced by pretreatment with diazepam, which presumably attenuates the anxiety stemming from the conflict. METHODS: To test the role of ethanol in reducing cocaine-induced anxiety, rats were trained to run a straight-arm alley for a single daily injection of cocaine (1.0 mg/kg IV). RESULTS: Rats that had the opportunity to then drink either an 8% or a 4% sucrose-ethanol solution immediately following their daily runway trial came to exhibit fewer retreats than rats that did not drink ethanol following their cocaine injection. CONCLUSIONS: These results suggest that ethanol effectively reduces the development of approach-avoidance conflict in animals running an alley for IV cocaine, a result that may account for the prevalence of cocaine-ethanol co-administration in humans.

Benzodiazepine modulation of opiate reward.

Case studies reveal that opiate addicts often premedicate themselves with benzodiazepine (BDZ) tranquilizers prior to taking their opiate. It has been hypothesized that such actions occur because the BDZ enhances the euphoric or reinforcing properties of the opiate. The present study tested this hypothesis in the animal laboratory. Sprague-Dawley rats were used to examine whether the magnitude of conditioned preferences for a distinctive environment associated with intravenous heroin delivery would be augmented by intraperitoneal alprazolam pretreatment. Results demonstrated that alprazolam produced a leftward shift of the heroin dose-response curve in the conditioned place preference test. The data obtained are consistent with the view that BDZs can augment the affective response to heroin in laboratory animals.

Dopamine antagonism attenuates the unconditioned incentive value of estrous female cues.

The role of dopaminergic transmission in the incentive-motivational processes involved in the generation of male sexual behavior was examined. Three groups of sexually naïve Long-Evans male rats traversed a straight alley for one of three goalbox targets: an empty goalbox, a nonestrous female, or an estrous female. A Plexiglas partition within the goalbox allowed for the perception of visual, auditory, and olfactory cues, but prevented physical contact. Baseline run times revealed that subjects returned to the goalbox significantly faster for an estrous female than for a nonestrous female, replicating our earlier work on the inherent incentive value of primary female cues. When subjects were then pretreated with the dopamine receptor antagonist, haloperidol (0.0, 0.075, or 0.15 mg/kg), they expressed decreased sexual motivation as reflected by increased run times for estrous female targets. Subjects' run times for the empty goalbox condition were unaffected by haloperidol, suggesting that the drug did not reliably impair motoric capacity. Results support the contention that central dopaminergic systems are involved in the regulation of the positive, unconditioned incentive value of estrous female cues.

Haloperidol challenge during copulation prevents subsequent increase in male sexual motivation.

Male rats manifest an increase in sexual motivation following sexual experience. The current experiment was devised to investigate the role of dopamine in this process by assessing whether sexual behavior occurring in the presence of the dopamine receptor antagonist, haloperidol, would continue to alter the subjects' subsequent sexual motivation. Four groups of male Long-Evans rats (total N=34) traversed an operant runway once per day for one of two goalbox targets: a nonestrous or estrous female. Following establishment of baseline run times (10 trials), all males received one ejaculation with a receptive female in a separate testing environment. Subjects were pretreated with vehicle or one of three doses of haloperidol (0.05, 0.075, 0.10 mg/kg) 45 min prior to being paired with the receptive female. All subjects successfully achieved ejaculation under these conditions. Subjects were then re-tested within the runway for their motivation to approach the two types of female targets (10 trials). Vehicle-treated subjects expressed the expected increase in sexual motivation following sexual experience, while haloperidol treatment dose-dependently attenuated this effect. Subjects that received the highest haloperidol dose subsequently manifested increased run times and intra-runway "retreat" behaviors, suggesting that female cues may have become associated with an aversive sexual experience. These results are consistent with the view that dopamine systems play a role in the rewarding or reinforcing consequences of male sexual behavior.

Evidence for opponent-process actions of intravenous cocaine.

The present experiment was devised to test a prediction of the Opponent-Process Theory of drug action. This theory presumes that the initial affective experience of a subject treated with cocaine would be diametrically different immediately after administration compared to some point later in time when the positive impact of the drug had subsided. A conditioned place-preference procedure was employed in which a novel environment was paired with the effects of cocaine either immediately after, 5 min after, or 15 min after an intravenous injection of 0.75 mg/kg cocaine. It was hypothesized that animals would come to prefer environments associated with the immediate positive effects of cocaine and avoid environments associated with the drug's subsequent negative effects. The results confirmed this hypothesis. While the 0-min delay and 5-min delay groups exhibited conditioned preferences for the cocaine-paired environment, the 15-min delay group came to avoid the side of the preference apparatus paired with cocaine. These data, therefore, serve as additional support for an Opponent-Process account of cocaine's actions.

Sexual motivation in the male rat: the role of primary incentives and copulatory experience.

Two experiments explored the motivational impact of primary incentive female cues on the operant behavior of sexually naive and experienced male rats. In the first experiment, a straight-arm runway was used to assess the subjects' motivation to approach a goalbox containing either male or female "targets." Twelve sexually naive Long-Evans males ran for: (1) an empty goalbox; (2) a male conspecific; (3) an ovariectomized (OVX) female; (4) an OVX female given estradiol; (5) or an OVX female treated with estradiol and progesterone. A perforated Plexiglas partition in the goalbox prevented the subject males from physically interacting with the targets, although olfactory, visual, and auditory cues were accessible. We hypothesized that subjects would manifest shorter run times (reflecting greater motivation) when the goalbox contained a receptive/proceptive female as opposed to a nonreceptive female target. Subjects' run times were ordered depending on the nature of the target (from slowest to fastest): empty goalbox, male conspecific, OVX female, OVX + estradiol female, and OVX + estradiol + progesterone female. As predicted, subjects ran significantly faster for a receptive/proceptive female than for a nonreceptive female, indicating that sexually naive males are inherently motivated by female precopulatory cues. In the second experiment, 30 sexually naive male subjects ran for a goalbox containing either a nonestrous (OVX) or an estrous (OVX + estradiol + progesterone) female. Following six trials, 10 males were allowed one intromission with a receptive female, 10 males experienced one ejaculation, and 10 remained sexually naive. Only those males having experienced an ejaculation subsequently decreased their run times for both nonestrous and estrous females, indicating that sexual reinforcement produced by ejaculation, but not intromission, further enhances the motivational impact of female incentive cues.

Haloperidol does not attenuate conditioned place preferences or locomotor activation produced by food- or heroin-predictive discriminative cues.

The present study examined whether a discriminative cue previously predictive of food or heroin reinforcement could activate and direct behavior in an environment that had never been paired with primary reinforcement. Olfactory cues, predicting the availability (S+) or unavailability (S-) of either heroin (0.1 mg/kg i.v.) or food (45 mg Noyes food pellets) reinforcement in the goal box of a straight-arm runway, were later tested in a separate environment for their ability to elicit locomotion (activate behavior) or induce a conditioned place preference (direct behavior). Presentation of the S+, but not the S-, resulted in a reliable increase in spontaneous locomotor activity that was not blocked by pretreatment with the dopamine receptor antagonist, haloperidol. Similarly, subjects displayed a preference for a novel location in which the S+, but not the S-, was placed. This preference was also unaltered by pretreatment with haloperidol. These data suggest that discriminative cues can profoundly affect behavior, even in environments that have themselves never been associated with primary reinforcement. Additionally, the conditioned motivational quality of these cues is unaltered by treatment with the same dopamine receptor antagonist shown in previous work to attenuate the primary reinforcing properties of heroin and food.

Intraaccumbens raclopride attenuates amphetamine-induced locomotion, but fails to prevent the response-reinstating properties of food reinforcement.

It has been well established that the presentation of a single reinforced trial in the midst of extinction results in a reinstatement of the previously reinforced operant response. In previous experiments, we have shown that systemically applied raclopride (a selective dopamine D2 receptor antagonist) dose dependently blocked the response-reinstating properties of food reinforcement, while SCH39166 (a selective dopamine D1 receptor antagonist) did not (11). The current experiments investigated the possible role of the nucleus accumbens in these actions of raclopride. In the first of two experiments, hungry rats were trained to traverse a straight runway for food reinforcement, a response that was then weakened through a series of extinction trials. On a single treatment trial, subjects were infused with one of three doses of intraaccumbens raclopride (0.0, 2.5, or 5.0 microg/0.5 microl/side) just prior to a reinforced trial. Twenty-four hours later, a single test trial was run in an unbaited runway. The results demonstrate that the prior day's reinforced trial produced a reinstatement of operant runway performance that was unaltered by intraaccumbens applications of raclopride. Two days later, the same animals were tested in a second experiment investigating the effects of intraaccumbens raclopride on amphetamine-induced locomotion. Subjects were pretreated with 1.0 mg/kg s.c. amphetamine prior to a 90-min locomotor activity session. The following day, subjects were again pretreated with amphetamine, but this time with a challenge dose of raclopride. Results demonstrate that the same raclopride doses that produced no effect in the response-reinstating experiment produced, in the same rats, a dose-dependent attenuation in amphetamine-induced locomotion. These data suggest that dopamine D2 receptors in the nucleus accumbens may not, in and of themselves, be necessary for the response-reinstating effects of food reinforcement.

Naloxone blocks reinforcement but not motivation in an operant runway model of heroin-seeking behavior.

The present study examined the effects of opiate receptor antagonism on both the motivation to seek heroin and the reinforcing consequences of heroin administration. Subjects were trained to discriminate between olfactory cues predicting either the delivery of intravenous heroin reinforcement (S+) or saline (S-). Subjects were then tested in the presence of the opiate receptor antagonist, naloxone (0.5, 1.0, or 3.0 mg/kg intraperitoneally). Naloxone had no effect on either S+ or S- trials. However, 24 hr later on the first posttreatment trial, subjects that had received heroin in the presence of naloxone (on the previous trial) now traversed the alley more slowly when presented with the S+. These data suggest that although the motivation to seek heroin was not disrupted by naloxone, the reinforcing consequences of heroin administration were.

Amphetamine infusions into the prefrontal cortex attenuate the sensitization to amphetamine.

1. Adult male rats were implanted with chronic medial prefrontal cortex cannulae. 2. On each of 5 consecutive days, rats received bilateral 0.5 ml intra-prefrontal cortex injections of either 5 mg d-amphetamine or saline, followed by a subcutaneous injection of either 1 mg/kg d-amphetamine or saline. 3. Immediately after the drug treatments each rat was placed into a photocell equipped locomotor activity chamber for 60 min. 4. Administration of d-amphetamine into the prefrontal cortex did not block the acute locomotor response to subcutaneous d-amphetamine nor did it in itself produce an increase in locomotor activity. However, prefrontal cortex amphetamine treatments did attenuate the sensitized locomotor effects of subcutaneous amphetamine that developed over trials/days. 5. Dopamine in the prefrontal cortex may be involved in the development of amphetamine-induced behavioral sensitization.

Haloperidol does not affect motivational processes in an operant runway model of food-seeking behavior.

The present experiment examined the effects of dopamine receptor antagonism on subjects' motivation to seek food. Rats were trained to discriminate between 2 olfactory cues predicting either the presence (S+) or absence (S-) of food reinforcement in the goal box of a straight-arm runway. Rats learned to traverse the alley quickly when presented with the S+ and much more slowly when presented with the S-. Haloperidol pretreatment was unable to alter this pattern of behavior (i.e., rats still ran quickly when presented with the scent that predicted food availability). Thus, it seems that the same dopamine antagonist treatments that have been shown to disrupt food reinforcement do not prevent the food-seeking behavior produced by presentation of food-predictive cues.

A role for D2, but not D1, dopamine receptors in the response-reinstating effects of food reinforcement.

Although the reinforcing properties of food are reduced in the presence of dopamine antagonist drugs, controversy exists about the relative roles of D1 vs D2 receptor subtypes in the actions of these drugs. The current experiment compared the effects of raclopride (a selective D2 receptor antagonist) and SCH 39166 (a selective D1 receptor antagonist) in the response-reinstating effects of food reinforcement. Hungry rats were trained to run a straight-alley for food reinforcement during single daily trials. The operant was then extinguished during consecutive daily non-reinforced trials. Subjects were then injected with one of four doses of raclopride (0.0, 1.0, 0.5, and 0.25 mg/kg, i.p.) or SCH 39166 (0.0, 1.0, 0.5, and 0.1 mg/kg i.p.) 30 min prior to a single reinforced treatment trial. Twenty-four h later, a test trial was conducted in an unbaited runway. The single reinforced trial in the midst of extinction was observed to reinstate operant runway performance. Raclopride, but not SCH 39166, dose-dependently attenuated this reinstatement. Motor control groups ruled out the possibility that these results were due to differential residual motor effects of the drugs. Results suggest that D2, but not D1, dopamine receptors, are involved in the response-reinstating properties of food reinforcement.

Concurrent positive and negative goalbox events produce runway behaviors comparable to those of cocaine-reinforced rats.

Rats traversing a straight-alley for reinforcing stimuli typically exhibit faster running times as training proceeds. In previous work from this laboratory, animals running for a reinforcement consisting of intravenous infusions of cocaine, unexpectedly demonstrated a progressive increased time to enter the goalbox over trials. Closer observation revealed that the animals were exhibiting a unique retreat behavior (i.e., stopping their forward advance toward the goalbox and returning toward the startbox). It was hypothesized that the retreat behavior reflected an inherent conflict that originated from concurrent positive and negative associations with the goalbox. Such associations were attributed to cocaine's dual and well documented reinforcing and anxiogenic effects. To test this idea, the present study compared the runway behavior of animals that concurrently received food and mild foot shock in the goalbox to the behavior of the other animals running for cocaine. Results demonstrated that food + shock reinforced animals took longer to enter the goalbox and made more retreats than a control group that received only food in the goalbox. Both these effects were reversed by pretreatment with the anticonflict, anxiolytic drug, diazepam. The behavior pattern of animals that received the combination of food and footshock was found to strongly resemble that of IV cocaine-reinforced rats, a result consistent with the notion that chronic cocaine administration has both positive and negative consequences.

Reinstatement of drug-seeking behavior produced by heroin-predictive environmental stimuli.

The current study examined whether stimuli predictive of heroin availability were capable of inducing a relapse of drug-seeking behavior in an operant runway task. Olfactory stimuli (orange and almond food extract) served as discriminative cues about the availability (S+) or unavailability (S-) of heroin reinforcement (a single 0.1 mg/kg IV infusion) in the goal box of a straight arm runway. Following discrimination training, the running response was extinguished in the absence of the olfactory cues. On a single trial, the discriminative stimuli were then tested for their ability to reinstate running behavior prior to presentation of the heroin reinforcer. Subjects presented with the S+ on test day took significantly less time to traverse the alley than they did on the final day of extinction, while those subjects presented with the S- on test day continued to run slowly. These results demonstrate, in an animal model of drug self-administration, that environmental discriminative cues can produce a relapse in drug seeking behavior following a period of abstinence. The response-reinstating properties of the S+ odor were unaltered by pretreatment with any of three doses of haloperidol (0.0, 0.15 or 0.3 mg/kg IP), suggesting that the motivating properties of heroin-predictive stimuli or cues remain intact during dopamine receptor antagonism.

Effects of haloperidol in a response-reinstatement model of heroin relapse.

The present study employed an animal model of drug relapse in which previously extinguished heroin self-administration behavior was reinstated following a single reinforced trial. Male albino rats were trained to traverse a straight-alley for a reinforcer consisting of a single IV injection of 0.06 mg/kg diacetylmorphine (heroin). Once the alley-running had been established, the heroin reinforcer was removed and the operant behavior permitted to extinguish over trials. On treatment day, animals were injected 45 min prior to testing with 0.0, 0.075, 0.10, 0.15 or 0.3 mg/kg of the dopamine receptor antagonist, haloperidol. A single trial was then conducted during which some animals continued to experience extinction conditions while others were injected with the heroin reinforcer upon entry into the goal box. The effects of these manipulations were determined during an additional single test trial conducted 24 h later when the subjects were no longer drugged. While heroin produced a reliable reinstatement in operant responding, this effect was dose-dependently prevented by pretreatment with haloperidol. These data suggest that dopamine receptor antagonism alters the reinforcing consequences of heroin administration as measured by heroin's ability to reinstate operant behavior following a prolonged period of nonreinforced responding.

Haloperidol differentially affects reinforcement and motivational processes in rats running an alley for intravenous heroin.

The role of drug-paired environmental stimuli in opiate self-administration was investigated by exposing animals to discrete cues that were predictive of the availability or unavailability of heroin reinforcement. Rats were trained to traverse a straight arm runway for a reinforcement consisting of a single 0.1 mg/kg intravenous infusion of heroin delivered upon entrance to the goal box. On each trial, one of two discriminative olfactory stimuli (orange and almond) was used: one which signaled the availability of heroin in the goal box (S+), and one which signaled its absence (S-). The effect of dopamine (DA) receptor antagonism on reinforcement and motivational processes was investigated by pretreating subjects with 0.0, 0.15 or 0.30 mg/kg of the DA receptor antagonist drug, haloperidol. Haloperidol had no effect on operant runway performance (i.e. goal time) in any condition. However, 24 h later, on the first post-treatment trial, those haloperidol animals that received heroin in the goal box on the previous trial (i.e. the S+ condition) ran reliably more slowly than subjects that received vehicle on the previous S+ trial. These results suggest that haloperidol does not affect the motivational properties of stimuli which predict the availability of heroin, while it does diminish the reinforcing effects of actually receiving heroin.

Repeated stimulation of the ventral tegmental area sensitizes the hyperlocomotor response to amphetamine.

The locomotor-activating effects of amphetamine have been reported to increase with repeated drug administration. Although the precise underlying mechanisms for this behavioral sensitization effect remain unknown, many investigators have suggested a role for the mesolimbic dopaminergic system that emanates from cell bodies in the ventral tegmental area (VTA) of the midbrain. To test this hypothesis, the present study examined the effects of repeated electrical stimulation of the VTA (in place of repeated amphetamine administration) on the hyperlocomotor actions of d-amphetamine. Locomotor activity induced by 0.75 mg/kg SC amphetamine was assessed during two 90-min tests, one before and one after a 14-day treatment regimen during which animals experienced daily 15-min sessions of intracranial VTA stimulation. Each session involved the delivery of 600 trains of 0.5 s 60-Hz sine-wave stimulation applied at one of four intensities: 0, 15, 30, or 45 microA. An additional comparison group of rats self-administered 30 microA of VTA stimulation. Data analysis revealed that both the self-stimulation and the high current groups were reliably more active posttreatment compared to pretreatment. No such sensitization-like effects were observed in any of the other treatment groups. These results are consistent with the hypothesis that repeated activation of VTA neurons can produce a sensitization to the behavioral effects of d-amphetamine.