Conditioned Reinforcement Test Research Articles

α4-Containing GABAA Receptors on DRD2 Neurons of the Nucleus Accumbens Mediate Instrumental Responding for Conditioned Reinforcers and Its Potentiation by Cocaine.

Extrasynaptic GABAA receptors (GABAARs) composed of α4, β, and δ subunits mediate GABAergic tonic inhibition and are potential molecular targets in the modulation of behavioral responses to natural and drug rewards. These GABAARs are highly expressed within the nucleus accumbens (NAc), where they influence the excitability of the medium spiny neurons. Here, we explore their role in modulating behavioral responses to food-conditioned cues and the behavior-potentiating effects of cocaine. α4-Subunit constitutive knock-out mice (α4-/-) showed higher rates of instrumental responding for reward-paired stimuli in a test of conditioned reinforcement (CRf). A similar effect was seen following viral knockdown of GABAAR α4 subunits within the NAc. Local infusion of the α4βδ-GABAAR-preferring agonist THIP (4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol; Gaboxadol) into the NAc had no effect on responding when given alone but reduced cocaine potentiation of responding for conditioned reinforcers in wild-type, but not α4-/- mice. Finally, specific deletion of α4-subunits from dopamine D2, but not D1, receptor-expressing neurons (DRD2 and DRD1 neurons), mimicked the phenotype of the constitutive knockout, potentiating CRf responding, and blocking intra-accumbal THIP attenuation of cocaine-potentiated CRf responding. These data demonstrate that α4-GABAAR-mediated inhibition of DRD2 neurons reduces instrumental responding for a conditioned reinforcer and its potentiation by cocaine and emphasize the importance of GABAergic signaling within the NAc in mediating the effects of cocaine.

Ablation of NMDA receptors in dopamine neurons disrupts attribution of incentive salience to reward-paired stimuli

Midbrain dopamine (DA) neurons play a crucial role in the formation of conditioned associations between environmental cues and appetitive events. Activation of N-methyl-d-aspartate (NMDA) receptors is a key mechanism responsible for the generation of conditioned responses of DA neurons to reward cues. Here, we tested the effects of the cell type-specific inactivation of NMDA receptors in DA neurons in adult mice on stimulus-reward learning. Animals were trained in a Pavlovian learning paradigm in which they had to learn the predictive value of two conditioned stimuli, one of which (CS+) was paired with the delivery of a water reward. Over the course of conditioning, mutant mice learned that the CS+ predicted reward availability, and they approached the reward receptacle more frequently during CS+ trials than CS- trials. However, conditioned responses to the CS+ were weaker in the mutant mice, possibly indicating that they did not attribute incentive salience to the CS+. To further assess whether the attribution of incentive salience was impaired by the mutation, animals were tested in a conditioned reinforcement test. The test revealed that mutant mice made fewer instrumental responses paired with CS+ presentation, confirming that the CS+ had a weaker incentive value. Taken together, these results indicate that reward prediction learning does occur in the absence of NMDA receptors in DA neurons, but the ability of reward-paired cues to invigorate and reinforce behavior is attenuated.

Cannabinoid receptor-1 signaling contributions to sign-tracking and conditioned reinforcement in rats.

Endocannabinoids (eCBs) are critical gatekeepers of dopaminergic signaling, and disrupting cannabinoid receptor-1 (CB1) signaling alters DA dynamics to attenuate cue-motivated behaviors. Prior studies suggest that dopamine (DA) release plays a critical role in driving sign-tracking. Here, we determine whether systemic injections of rimonabant, a CB1 receptor inverse agonist, during Pavlovian lever autoshaping impair the expression of sign-tracking. We next examine whether rimonabant blocks the reinforcing properties of the Pavlovian lever cue in a conditioned reinforcement test. In Exp. 1, we trained rats in Pavlovian lever autoshaping prior to systemic rimonabant injections (0, 1, 3mg/kg) during early and late Pavlovian lever autoshaping sessions. In Exp. 2, we trained rats in Pavlovian lever autoshaping prior to systemic rimonabant injections (0, 1mg/kg) during a conditioned reinforcement test. Rimonabant dose-dependently decreased lever contact and probability, and increased sign-tracker's latency to approach the lever cue early in Pavlovian training. With extended training, many previously goal-tracking and intermediate rats shifted to lever approach, which remained dose-dependently sensitive to rimonabant. Rimonabant attenuated cue-evoked food cup approach early, but not late, in conditioning, and did not affect pellet retrieval or consumption. The inserted lever cue served as a robust conditioned reinforcer after Pavlovian lever autoshaping, and 1mg/kg rimonabant blocked conditioned reinforcement. Together, our results suggest that CB1 signaling mediates two critical properties of incentive stimuli; their ability to attract (Exp. 1) and their ability to reinforce (Exp. 2) behavior.

Individual Differences in the Attribution of Incentive Salience to a Pavlovian Alcohol Cue.

Individual differences exist in the attribution of incentive salience to conditioned stimuli associated with food. Here, we investigated whether individual differences also manifested with a Pavlovian alcohol conditioned stimulus (CS). We compiled data from five experiments that used a Pavlovian autoshaping paradigm and tests of conditioned reinforcement. In all experiments, male, Long-Evans rats with unrestricted access to food and water were acclimated to 15% ethanol. Next, rats received Pavlovian autoshaping training, in which a 10 s presentation of a retractable lever served as the CS and 0.2 mL of 15% ethanol served as the unconditioned stimulus (US). Finally, rats underwent conditioned reinforcement tests in which nose-pokes to an active aperture led to brief presentations of the lever-CS, but nose-pokes to an inactive aperture had no consequence. Rats were categorized as sign-trackers, goal-trackers and intermediates based on a response bias score that reflected their tendencies to sign-track or goal-track at different times during training. We found that distinct groups of rats either consistently interacted with the lever-CS (“sign-trackers”) or routinely approached the port during the lever-CS (“goal-trackers”) across a majority of the training sessions. However, some individuals (“shifted sign-trackers”) with an early tendency to goal-track later shifted to comparable asymptotic levels of sign-tracking as the group identified as sign-trackers. The lever-CS functioned as a conditioned reinforcer for sign-trackers and shifted sign-trackers, but not for goal-trackers. These results provide evidence of robust individual differences in the extent to which a Pavlovian alcohol cue gains incentive salience and functions as a conditioned reinforcer.

Conditioned-reinforcement tests of schedule-induced drinking

ABSTRACTThe purpose of this experiment was to test the role of food to establish a neutral stimulus as a conditioned reinforcer of licking in the schedule-induced drinking (SID) procedure. Throughout the experiment, one group of food-deprived rats received food-tone pairings and another group received food only, according to a fixed time 30-s schedule. Successive experimental phases tested the effects of food-tone pairings on rates of licking a water tube that produced a drop of water and the same tone that was paired with food, on the resistance to extinction of licking, and on the establishment of a new response (button-pressing) reinforced by the tone. Higher frequencies of licking, greater resistance to extinction of licking and higher rates of button pressing were observed for the rats that received food-tone pairings in comparison with the rats that received food only. The results are interpreted as supporting an account that SID is an operant behaviour reinforced by food.

The tendency to sign-track predicts cue-induced reinstatement during nicotine self-administration, and is enhanced by nicotine but not ethanol.

Some individuals are particularly responsive to reward-associated stimuli ("cues"), including the effects of these cues on craving and relapse to drug-seeking behavior. In the cases of nicotine and alcohol, cues may acquire these abilities via the incentive-enhancing properties of the drug. To determine the interaction between cue-responsivity and nicotine reinforcement, we studied the patterns of nicotine self-administration in rats categorized based on their tendency to approach a food-predictive cue ("sign-trackers") or a reward-delivery location ("goal-trackers"). In a second experiment, we determined whether nicotine and ethanol altered the incentive value of a food cue. Rats were classified as sign- or goal-trackers during a Pavlovian conditioned approach paradigm. Rats then self-administered intravenous nicotine (0.03mg/kg infusions) followed by extinction and cue-induced reinstatement tests. We also tested the effects of nicotine (0.4mg/kg base s.c.) or ethanol (0.7g/kg i.p.) on the approach to, and reinforcing efficacy of, a food cue. Sign-trackers showed greater reinstatement in response to a nicotine cue. Further, nicotine enhanced sign-tracking but not goal-tracking to a food cue and also enhanced responding for the food cue during the conditioned reinforcement test. Conversely, ethanol reduced sign-tracking and increased goal-tracking, but had no effect on conditioned reinforcement. Our studies demonstrate that the tendency to attribute incentive value to a food cue predicts enhanced cue-induced reinstatement. Additionally, the incentive value of food cues is differentially modulated by nicotine and ethanol, which may be related to the reinforcing effects of these drugs.

Distinct effects of ventral tegmental area NMDA and acetylcholine receptor blockade on conditioned reinforcement produced by food-associated cues

Stimuli paired with rewards acquire reinforcing properties to promote reward-seeking behavior. Previous work supports the role of ventral tegmental area (VTA) nicotinic acetylcholine receptors (nAChRs) in mediating conditioned reinforcement elicited by drug-associated cues. However, it is not known whether these cholinergic mechanisms are specific to drug-associated cues or whether VTA cholinergic mechanisms also underlie the ability of cues paired with natural rewards to act as conditioned reinforcers. Burst firing of VTA dopamine (DA) neurons and the subsequent phasic DA release in the nucleus accumbens (NAc) plays an important role in cue-mediated behavior and in the ability of cues to acquire reinforcing properties. In the VTA, both AChRs and N-methyl-d-aspartate receptors (NMDARs) regulate DA burst firing and phasic DA release. Here, we tested the role of VTA nAChRs, muscarinic AChRs (mAChRs), and NMDARs in the conditioned reinforcement elicited by a food-associated, natural reward cue. Subjects received 10 consecutive days of Pavlovian conditioning training where lever extension served as a predictive cue for food availability. On day 11, rats received bilateral VTA infusion of saline, AP-5 (0.1 or 1μg), mecamylamine (MEC: 3 or 30μg) or scopolamine (SCOP: 3 or 66.7μg) immediately prior to the conditioned reinforcement test. During the test, nosepoking into the active (conditioned reinforced, CR) noseport produced a lever cue while nosepoking on the inactive (non-conditioned reinforced, NCR) noseport had no consequence. AP-5 robustly attenuated conditioned reinforcement and blocked discrimination between CR and NCR noseports at the 1-μg dose. MEC infusion decreased responding for both CR and NCR while 66.7-μg SCOP disrupted the subject’s ability to discriminate between CR and NCR. Together, our data suggest that VTA NMDARs and mAChRs, but not nAChRs, play a role in the ability of natural reward-associated cues to act as conditioned reinforcers.

The attribution of incentive salience to Pavlovian alcohol cues: a shift from goal-tracking to sign-tracking.

Environmental stimuli that are reliably paired with alcohol may acquire incentive salience, a property that can operate in the use and abuse of alcohol. Here we investigated the incentive salience of Pavlovian alcohol cues using a preclinical animal model. Male, Long-Evans rats (Harlan) with unrestricted access to food and water were acclimated to drinking 15% ethanol (v/v) in their home-cages. Rats then received Pavlovian autoshaping training in which the 10 s presentation of a retractable lever served as the conditioned stimulus (CS) and 15% ethanol served as the unconditioned stimulus (US) (0.2 ml/CS; 12 CS presentations/session; 27 sessions). Next, in an operant test of conditioned reinforcement, nose pokes into an active aperture delivered presentations of the lever-CS, whereas nose pokes into an inactive aperture had no consequences. Across initial autoshaping sessions, goal-tracking behavior, as measured by entries into the fluid port where ethanol was delivered, developed rapidly. However, with extended training goal-tracking diminished, and sign-tracking responses, as measured by lever-CS activations, emerged. Control rats that received explicitly unpaired CS and US presentations did not show goal-tracking or sign-tracking responses. In the test for conditioned reinforcement, rats with CS-US pairings during autoshaping training made more active relative to inactive nose pokes, whereas rats in the unpaired control group did not. Moreover, active nose pokes were positively correlated with sign-tracking behavior during autoshaping. Extended training may produce a shift in the learned properties of Pavlovian alcohol cues, such that after initially predicting alcohol availability they acquire robust incentive salience.

Individual variation in the motivational and neurobiological effects of an opioid cue.

A discrete cue associated with intravenous injections of cocaine acquires greater control over motivated behavior in some rats ('sign-trackers', STs) than others ('goal-trackers', GTs). It is not known, however, if such variation generalizes to cues associated with other drugs. We asked, therefore, whether a discrete cue (a light) associated with the intravenous administration of an opioid drug (the short-acting mu receptor agonist, remifentanil) acquires incentive motivational properties differently in STs and GTs, as indicated by tests of Pavlovian conditioned approach and conditioned reinforcement. Consistent with studies using cocaine, STs approached a classically conditioned opioid cue more readily than GTs, and in a test of conditioned reinforcement worked more avidly to get it. Interestingly, STs and GTs did not differ in the acquisition of a conditioned orienting response. In addition, the performance of conditioned approach behavior, but not conditioned orientation, was attenuated by pretreatment with the dopamine receptor antagonist, flupenthixol, into the core of the nucleus accumbens. Lastly, food and opioid cues engaged similar amygdalo-striatal-thalamic circuitry to a much greater extent in STs than GTs, as indicated by Fos expression. Taken together, these data demonstrate that, similar to food and cocaine cues: (1) a discrete opioid cue attains greater incentive motivational value in STs than GTs; (2) the attribution of incentive motivational properties to an opioid cue is dopamine dependent; and (3) an opioid cue engages the so-called 'motive circuit' only if it is imbued with incentive salience.

Responding for conditioned reinforcement in C57BL/6 and CD-1 mice, and Sprague-Dawley rats: Effects of methylphenidate and amphetamine

Characterization of responding for conditioned reinforcement in mice is important to implement genetic tools in examining the neurobiological mechanisms underlying reward-related learning and incentive motivation. Inbred C57BL/6 mice, outbred CD-1 mice, and outbred Sprague-Dawley rats underwent Pavlovian conditioning in which a conditioned stimulus (CS) was paired with saccharin. Subsequently, subjects were allowed to respond for that CS in tests of responding for conditioned reinforcement. Experiments measured the effects of methylphenidate (MPH) and amphetamine (AMPH) on lever pressing for conditioned reinforcement in mice and rats. We further examined the stability of responding for conditioned reinforcement in mice after repeated testing and the extinction of this behaviour following omission of the reinforcer. We also determined whether the CS exhibited reinforcing properties if it was not paired with saccharin. C57BL/6 and CD-1 mice learned to respond for a conditioned reinforcer similarly to rats, and the behaviour was stable over time. MPH increased responding in CD-1 mice and rats, but not in C57BL/6 mice. AMPH only increased responding in rats. Responding was reduced following omission of the conditioned reinforcer, and responding was only established when the CS was paired with saccharin. These experiments characterize a conditioned reinforcement test which produces stable responding in two different mouse backgrounds. These findings also show that dopaminergic psychomotor stimulants can differently affect rats and mice in tests of responding for conditioned reinforcement.

The role of melanin‐concentrating hormone in conditioned reward learning

The orexigenic neuropeptide melanin-concentrating hormone (MCH) is well positioned to play a key role in connecting brain reward and homeostatic systems due to its synthesis in hypothalamic circuitry and receptor expression throughout the cortico-striatal reward circuit. Here we examined whether targeted-deletion of the MCH receptor (MCH-1R) in gene-targeted heterozygote and knockout mice (KO), or systemic treatment with pharmacological agents designed to antagonise MCH-1R in C57BL/6J mice would disrupt two putative consequences of reward learning that rely on different neural circuitries: conditioned reinforcement (CRf) and Pavlovian-instrumental transfer (PIT). Mice were trained to discriminate between presentations of a reward-paired cue (CS+) and an unpaired CS-. Following normal acquisition of the Pavlovian discrimination in all mice, we assessed the capacity for the CS+ to act as a reinforcer for new nose-poke learning (CRf). Pharmacological disruption in control mice and genetic deletion in KO mice impaired CRf test performance, suggesting MCH-1R is necessary for initiating and maintaining behaviors that are under the control of conditioned reinforcers. To examine a dissociable form of reward learning (PIT), a naïve group of mice were trained in separate Pavlovian and instrumental lever training sessions followed by the PIT test. For all mice the CS+ was capable of augmenting ongoing lever responding relative to CS- periods. These results suggest a role for MCH in guiding behavior based on the conditioned reinforcing value of a cue, but not on its incentive motivational value.

Effects of serum response factor (SRF) deletion on conditioned reinforcement

Serum response factor (SRF) is a ubiquitously expressed stimulus-dependent transcription factor that regulates gene expression by binding to serum response element in the promoter region of target genes. Recent studies in mice have shown that SRF is important for activity-dependent gene expression and synaptic plasticity in the adult brain but is dispensable for neuronal survival. Given these important functions of SRF in the CNS, it is expected to play a critical role in several aspects of learning and memory. Here we evaluated the role of SRF in conditioned reinforcement using two lines of conditional SRF mutant mice. These SRF mutant mice exhibited different spatial patterns of SRF deletion in the post-natal forebrain and notably within the hippocampus. SRF deletion was more widespread in SRF-CKCre mutants than in SRF-SynCre mutants, particularly in areas of the cortex and striatum. Mutant and wild-type mice were trained to associate one auditory cue (CS+) with reward, whereas a second cue remained relatively neutral (CS−). All mice readily acquired this discrimination, entering the food cup during CS+ but not during CS−. In a subsequent test of conditioned reinforcement, in the absence of food, wild-type control mice and SRF-SynCre mice learned to selectively perform an instrumental response that yielded CS+ presentation rather than another response that produced CS−. SRF-CKCre mutants failed to show this preferential responding for CS+. These results suggest a role for SRF in conditioned reinforcement, a manifestation of incentive learning that has been implicated in many aspects of adaptive and maladaptive behavior, such as substance abuse and eating disorders.

Differential Contributions of Dopamine and Serotonin to Orbitofrontal Cortex Function in the Marmoset

We have shown previously that the inhibitory control functions of the orbitofrontal cortex (OFC) are disrupted by serotonin, but not dopamine depletions. However, both dopamine and serotonin terminals and receptors are present within the OFC and thus the aim of the present study was to determine the differential contributions of these neurotransmitters to orbitofrontal function. OFC and dopamine are involved in the process by which neutral stimuli take on reinforcing properties, by virtue of their prior association with reward, and guide behavior. Thus, we compared the performance of marmosets with dopaminergic or serotoninergic OFC depletions on a test of conditioned reinforcement. To further our understanding of serotonin in behavioral flexibility, the effect of these depletions was also compared on the extinction of a visual discrimination. Monkeys with serotonin depletions of the OFC displayed stimulus-bound responding on both tests of conditioned reinforcement and discrimination extinction suggesting that orbitofrontal serotonin plays a specific role in preventing competing, task irrelevant, salient stimuli from biasing responding. In contrast, monkeys with dopamine depletion were insensitive to conditioned reinforcers and displayed persistent responding in the absence of reward in extinction, a pattern of deficits that may reflect basic deficits in the associative processing of reward.

β2-Subunit-containing nicotinic acetylcholine receptors are involved in nicotine-induced increases in conditioned reinforcement but not progressive ratio responding for food in C57BL/6 mice

Nicotine administration potentiates conditioned reinforcement in rats, an effect that persists for weeks after chronic exposure. Little is known regarding the nicotinic receptor subtypes that may mediate this effect. The purpose of this study was to determine whether beta2-subunit-containing nicotinic acetylcholine receptors (beta2*nAChRs) are necessary for lasting effects of nicotine on conditioned and primary reinforcement in mice. Beta2 knockout (beta2KO) and wild-type (WT) mice received 14 days of nicotine exposure (NIC, 200 microg/ml in 2% saccharin) or saccharin alone (SAC) in their drinking water. Five days later, mice received paired presentations of a conditioned stimulus (CS) with water unconditioned stimulus (US) or explicitly unpaired presentations of the CS and US during Pavlovian discriminative approach training. Training was followed by two conditioned reinforcement tests. Mice were subsequently tested for food-reinforced responding in the absence of explicit cues followed by a progressive ratio test. During conditioned reinforcement testing, only mice in the paired condition showed increased responding in the CS-reinforced aperture over inactive apertures. WT-NIC mice showed enhanced conditioned reinforcement compared to WT-SAC animals. beta2KO-SAC mice showed elevated conditioned reinforcement compared to WT-SAC subjects, but beta2KO-NIC and beta2KO-SAC mice did not differ in responding with conditioned reinforcement. Prior nicotine exposure did not alter food-reinforced responding but resulted in elevated break points for food in both genotypes. These data show that nicotine exposure enhances conditioned reinforcement in mice and indicate that beta2*nAChRs are necessary for nicotine-dependent enhancement of incentive aspects of motivation but not motivation for primary reinforcement measured by progressive ratio responding.

Serotonin-Dopamine Interactions in the Control of Conditioned Reinforcement and Motor Behavior

These studies addressed the question of serotonin (5-HT)-dopamine (DA) interactions with regard to reward-related behavior and motor activity in rats. The first experiment evaluated the effect of chronic treatment with fluoxetine (7 mg/kg/day), a serotonin-selective reuptake inhibitor, and buproprion (15 mg/kg/day), a dopamine reuptake inhibitor, on responding for conditioned reinforcement (CR). Chronic fluoxetine, but not buproprion, enhanced CR responding, and also potentiated cocaine-induced increases in CR responding. In the second experiment, animals received intra-accumbens infusions of either 5-HT (0, 1, 5, and 10 μg) or DA (10, 20 μg) prior to the conditioned reinforcement test. Dopamine, but not 5-HT, selectively facilitated CR responding, although 5-HT non-specifically increased responding as well. In the third and fourth experiments, it was demonstrated that intra-accumbens 5-HT causes increased motor activity, which was partially blocked by DA antagonists. The results suggest that chronically increased levels of 5-HT may facilitate reward-related behavior, but most likely via indirect modulatory mechanisms affecting general arousal and motor tone.

Tests of the conditioned reinforcement value of sequential stimuli in pigeons

Egger and Miller (1962) hypothesized that the conditioned reinforcing value of stimuli depends on their information value. Egger and Miller and others have tested this hypothesis by comparing the conditioned reinforcing value of S1 and S2 following S1-S2-reward training. However, none of these experiments have controlled for differential generalization of conditioned reinforcement value from training to comparison tests. That is, the S1 cue pattern during the conditioned reinforcement tests has been very similar to the S1 cue pattern of training, while the training and test S2 cue patterns have been quite dissimilar. In Experiment 1, pigeons in a procedure unconfounded by differential generalization produced S2 reliably more frequently than S1, and pigeons in a confounded procedure produced S1 somewhat more frequently than S2. A significant groups × stimuli interaction was attributed to differential stimulus generalization from training to test for S1 and S2 in the confounded condition. In Experiment 2, pigeons in an unconfounded procedure again produced S2 reliably more frequently under a different testing procedure. The results are interpreted as demonstrating that, following S1-S2-food training trials, S2 is the more effective conditioned reinforcer in unconfounded conditions. A reconceptualization of the information hypothesis is shown to be consistent with these results.

Reinforcement with intragastric infusions of ethanol: Blocking effect of FLA 57

Suppression of oral intake of ethanol by FLA 57 has been reported for rats and was attributed to an inhibition of dopamine β-hydroxylase. We have demonstrated the ability of FLA 57 (50 mg/kg, IP) to suppress bar-pressing for intragastric (IG) delivery of doses of ethanol (25 mg/kg). This indicates that the effect on oral intake of ethanol may not be attributed to a taste factor, e.g., a decreased palatability of the ethanol solution. The same dose of FLA 57 did not suppress responding for IG doses of sweet milk. Thus, there was not an impairment of appetitive behavior in general through some nonspecific depressant or toxic action. Furthermore, the primary reinforcing action of ethanol, when used to establish a buzzer as a conditioned reinforcer through repeated pairings, was blocked if FLA 57 was given before pairings. This was evidenced by a failure of such rats to bar-press above the baseline level in a later test of conditioned reinforcement, which contrasted with the increased responding seen for rats receiving saline instead of FLA 57 before ethanol. These data support the previous findings on oral ethanol and confirm that FLA 57 can impair the mechanism by which ethanol produces positive reinforcement in rats.

The overshadowing of instrumental conditioning by a stimulus that predicts reinforcement better than the response

In an experiment designed to demonstrate the overshadowing of appetitive instrumental conditioning, three groups of rats were given 10 sessions of RI (random interval) training in which reinforcement was delivered 450 msec following the response. The correlated group experienced a stimulus during the response-reinforcer delay interval, while an uncorrelated control group experienced a similar brief stimulus occasionally following responses, but these were not responses that typically produced reinforcement. A no-stimulus control group did not experience the brief response-produced stimulus. The experiment was run in two replications. The first employed a light as the critical stimulus, the second a tone. Over the 10 RI sessions, subjects in the two control conditions increased their rate of responding significantly faster than subjects in the correlated condition in both replications. This finding was interpreted as an instance of the overshadowing of the acquisition of signal value by a response because of the presence of the stimulus, which, in the correlated condition, was a more reliable predictor of reinforcement than the response. A subsequent conditioned reinforcement test confirmed that, in the correlated condition, the stimulus had, indeed, become a signal for reinforcement as a function of RI training.