The Conditioned Place Preference Experiment – Anthony Phillips and Fibiger

1. Abstract

The Conditioned Place Preference (CPP) paradigm, pioneered and substantially refined in behavioral neuroscience and neuropsychopharmacology by Anthony G. Phillips and Hans C. Fibiger during the late 1970s and early 1980s, stands as one of the most prominent non-human animal models and behavioral assays used to evaluate the motivational, incentive, and rewarding properties of pharmacological compounds and natural reinforcers. Rooted in classical Pavlovian conditioning principles, the CPP experiment pairs distinctive, motivationally neutral environmental context cues (tactile, visual, and spatial configurations) with the internal affective states induced by an unconditioned stimulus (UCS), such as a drug of abuse, electrical brain stimulation, or food reward. Across the standard multi-phase protocol—comprising habituation/pre-test, conditioning trials, and post-conditioning testing—the animal demonstrates conditioned preference by spending a significantly greater proportion of time exploring and occupying the specific environment previously paired with the reinforcing agent relative to a vehicle-paired or neutral alternative.

Unlike operant self-administration paradigms that require active instrumental responding to sustain drug delivery, the Phillips and Fibiger CPP protocol dissociates primary motor operant performance from conditioned secondary incentive valuation, isolating the neurobiological substrates of drug reward and associative memory formation. The primary operational metric is the conditioned preference score, quantified either as absolute dwell time in the drug-paired compartment or as a shift score comparing post-conditioning dwell time against baseline pre-test occupancy. Psychometrically and methodologically, CPP exhibits high pharmacological validity, cross-species replicability, and predictive validity for human drug abuse liability. When evaluated within contemporary behavioral psychometrics, CPP displays high test-retest reliability across standardized apparatus parameters, robust construct validity anchored in mesolimbic dopamine transmission, and distinct sensitivity to both rewarding and aversive (conditioned place aversion) properties of psychoactive agents. This article provides a comprehensive academic analysis of the Phillips and Fibiger Conditioned Place Preference assay, examining its historical development, theoretical foundations, psychometric properties, experimental standardization, validity profiles, and translational utility in modern neuroscience.

2. Keywords

Conditioned Place Preference, Anthony Phillips, Hans Fibiger, Pavlovian conditioning, behavioral pharmacology, incentive salience, drug reward, mesolimbic dopamine system, psychometrics, animal behavioral assays

3. Authors

The formal standardization and theoretical articulation of the Conditioned Place Preference paradigm for mapping central reward pathways were spearheaded by two distinguished Canadian neuroscientists and psychopharmacologists at the University of British Columbia (UBC):

• Anthony G. Phillips, Ph.D., FRSC, O.C. — Professor Emeritus in the Department of Psychiatry and Dunning-Kellam Chair in Neurosciences at the University of British Columbia, Vancouver, BC, Canada. Dr. Phillips has served as the Scientific Director of the Canadian Institutes of Health Research (CIHR) Institute of Neurosciences, Mental Health and Addiction. His foundational research centers on the neurochemical basis of learning, motivation, addiction, and schizophrenia, particularly the role of mesoaccumbens dopamine transmission in brain stimulation reward and chemical reinforcers.
• Hans C. Fibiger, Ph.D., FRSC — Professor Emeritus of Psychiatry at the University of British Columbia and former Global Head of Neuroscience Research at Eli Lilly and Company, Amgen, and Pharmacia. Dr. Fibiger is widely recognized for his landmark contributions identifying the neuroanatomy of ascending monoaminergic and cholinergic pathways and their critical contributions to reward, psychostimulant action, cognitive processing, and neurodegenerative disorders.

Collaborating with colleagues and trainees including Christina Spyraki, Derek van der Kooy, and Roy A. Wise, Phillips and Fibiger systematically demonstrated that psychostimulant- and opiate-induced place preferences rely on dissociable neurochemical circuitries, laying the methodological benchmark for hundreds of contemporary addiction and psychopharmacology laboratories globally.

4. Purpose

The Conditioned Place Preference experiment was developed to solve a persistent methodological challenge in behavioral pharmacology: how to quantify the subjective, affective, and motivational valence of psychoactive drugs without the confounding motor, sensory, and procedural demands intrinsic to operant self-administration schedules. In traditional operant tasks (such as intravenous drug self-administration via lever-pressing), an animal must continuously execute an active motor output to trigger reward delivery. Consequently, sedating, ataxic, hyperlocomotor, or stereotypic side effects induced by the drug can directly impair or artificially distort the rate of lever-pressing, confounding motor competence with drug reward magnitude.

Phillips and Fibiger (along with contemporaneous work by Rossi and Reid, and Kumar) established CPP to achieve several core scientific objectives:

• Dissociation of Drug State from Behavioral Testing: Testing occurs in a drug-free state during the probe phase. By evaluating the animal’s spatial preference in the absence of acute pharmacological intoxication, CPP guarantees that the observed exploratory behavior reflects conditioned affective memory and incentive attribution rather than acute motor stimulation, sedation, or cognitive disorientation.
• Measurement of Associative Incentive Salience: CPP quantifies the degree to which contiguous pairing of environmental stimuli with unconditioned pharmacological reward enables those static environmental cues to acquire secondary incentive properties (conditioned reinforcers), capable of eliciting approach behavior.
• Evaluation of Natural and Pharmacological Reinforcers: The assay provides a versatile behavioral platform capable of assessing both chemical rewards (e.g., d-amphetamine, cocaine, morphine, nicotine, ethanol) and natural rewards (e.g., sucrose, palatable food, sexual interaction, social housing, maternal care).
• Detection of Aversive States (Conditioned Place Aversion – CPA): By the same associative mechanism, if a paired stimulus induces dysphoria, nausea, visceral discomfort, or withdrawal distress (such as lithium chloride, naloxone-precipitated opioid withdrawal, or kappa-opioid receptor agonists), animals exhibit systematic avoidance of the paired chamber.
• Screening and Drug Discovery: In translational pharmaceutical research, CPP serves as a frontline behavioral screen for assessing the abuse liability of novel therapeutic candidates (e.g., analgesics, anxiolytics, stimulants) and evaluating pharmacological antagonists or gene therapies aimed at extinguishing relapse-like cue-induced conditioned responses.

5. Psychological Construct

The psychological construct measured by the Phillips and Fibiger Conditioned Place Preference assay is conditioned incentive value—an acquired motivational valence synthesized through classical stimulus-outcome conditioning. Rather than capturing a single, simple motoric reflex, CPP indexes a multidimensional neurobehavioral construct comprising several distinct, hierarchically integrated cognitive-affective sub-processes:

1. Primary Reinforcement and Hedonic Processing

At the baseline level, the organism experiences an unconditioned hedonic or motivational response to the UCS. In the initial Phillips and Fibiger studies using psychostimulants (e.g., d-amphetamine, methylphenidate), the elevation of extracellular dopamine within the ventral striatum (nucleus accumbens) triggers an internal affective state corresponding to positive valence or reward. This primary affective reaction constitutes the foundational signal that drives associative learning.

2. Associative Contextual Conditioning (Pavlovian S-S Learning)

During the conditioning phase, the interoceptive pharmacological state (UCS) is repeatedly and contiguously paired with a multidimensional compound conditioned stimulus (CS+), consisting of specific visual patterns (e.g., black versus white walls, striped versus dotted geometries), tactile floor textures (e.g., wire mesh versus smooth plexiglass, parallel steel bars versus sandpaper), and ambient olfactory or spatial cues. Through associative synaptic plasticity, an enduring association is forged between the compound environmental stimulus configuration and the neurochemical reward trace.

3. Conditioned Incentive Salience (Incentive Motivation)

As described by Robinson and Berridge’s incentive-sensitization theory and Phillips’ neurochemical models of motivation, the previously neutral CS+ undergoes a qualitative transformation: it is imbued with “incentive salience.” The environmental cues cease to function purely as predictive signals; they become motivational magnets. When the animal is placed in the apparatus during the drug-free post-test, the CS+ triggers an active conditioned approach response (sign-tracking-like behavior), drawing the animal toward and keeping it within the reward-associated compartment.

4. Spatial Memory Retrieval and Decision-Making

During the unconstrained post-conditioning choice session, the rodent must navigate across the apparatus, continuously integrating sensory inputs from each compartment, retrieving the affective memories associated with those contexts, and executing a value-based behavioral choice. The relative proportion of time spent in each compartment reflects the outcome of an ongoing cognitive appraisal between the comparative valences of the CS+ and CS- (vehicle-paired) environments.

6. Theoretical Framework

The Conditioned Place Preference experiment is anchored at the theoretical intersection of Pavlovian associative learning theory, reinforcement neurobiology, and incentive motivation theory. The theoretical framework established by Anthony Phillips and Hans Fibiger specifically synthesizes three fundamental scientific tenets:

1. Pavlovian Higher-Order Conditioning and Conditioned Reinforcement

Traditional Skinnerian operant frameworks postulated that reward could only be demonstrated if an agent increased the operational rate of an arbitrary operant response (e.g., lever pressing). Phillips and Fibiger challenged this conceptual limitation by leveraging Pavlov’s classical conditioning framework. They postulated that contextual stimuli paired with internal rewarding states acquire secondary reinforcing properties. When granted free access, the organism allocates its behavioral trajectory toward the environmental stimulus that elicits conditioned appetitive neurophysiological states. The spatial allocation of time serves as an unconstrained, continuous, ethologically natural readout of secondary reinforcement value.

2. The Mesolimbic Dopamine Hypothesis of Reward

Phillips and Fibiger designed their CPP experiments to directly test the neuroanatomical and neurochemical substrates of intracranial self-stimulation (ICSS) and psychostimulant reward. Prior to their work, significant controversy surrounded whether ascending dopaminergic or noradrenergic systems mediated drug reinforcement. In a series of seminal publications, Phillips and Fibiger (1979) and Spyraki, Fibiger, and Phillips (1982) demonstrated that selective bilateral 6-hydroxydopamine (6-OHDA) lesions of the nucleus accumbens or ventral tegmental area (VTA), as well as selective dopamine receptor antagonism (e.g., using haloperidol or pimozide), completely abolished d-amphetamine-induced conditioned place preference while leaving food-induced place preference intact under specific boundary conditions.

This established the theoretical principle that psychostimulant reward is critically dependent on dopamine transmission in the mesoaccumbens projection, providing direct behavioral-pharmacological evidence that CPP reflects specific mesolimbic circuit engagement rather than generalized behavioral arousal.

3. Opponent-Process Theory and Homeostatic Modulation

The theoretical paradigm also accommodates Solomon and Corbit’s opponent-process theory of motivation. When an addictive agent produces an intense positive primary affective reaction (the ‘a’ process), it subsequently recruits counter-regulatory, homeostatic neuroadaptations (the ‘b’ process) characterized by negative affect and withdrawal. Phillips and Fibiger’s paradigm allows for the precise temporal mapping of these opposing hedonic vectors. By shifting the timing of the conditioning trials relative to drug administration (e.g., pairing cues with immediate peak drug absorption versus delayed withdrawal phases), the apparatus captures both positive conditioned place preference and negative conditioned place aversion.

7. Validity

Decades of behavioral pharmacological and psychometric research have scrutinized the validity parameters of the Conditioned Place Preference apparatus, confirming its standing as a gold-standard preclinical behavioral assay.

1. Construct Validity

Construct validity evaluates whether CPP accurately reflects the psychological construct of incentive motivation and associative reward. Extensive neurobiological investigations confirm that the neural circuits recruited during the expression of CPP correspond precisely to the neural networks subserving natural reward processing and human craving. Specifically, pharmacological or optogenetic stimulation of VTA dopaminergic projections to the nucleus accumbens shell reliably establishes robust CPP, whereas inhibition of these circuits blocks preference acquisition. Furthermore, immediate-early gene expression (e.g., c-Fos, DeltaFosB) and cyclic AMP response element-binding protein (CREB) phosphorylation are selectively upregulated in the prefrontal cortex, amygdala, and nucleus accumbens of rodents exhibiting conditioned place preference, confirming robust construct fidelity at cellular, circuit, and behavioral levels.

2. Predictive and Pharmacological Validity

The predictive validity of CPP regarding human drug abuse liability is remarkably high (approaching 90% congruence across examined pharmacotherapies):

• Compounds that possess high abuse and dependence potential in humans—including amphetamine, methamphetamine, cocaine, heroin, morphine, oxycodone, nicotine, ethanol, and delta-9-tetrahydrocannabinol (THC)—consistently generate statistically significant conditioned place preferences across rodents (rats, mice) and primates.
• Conversely, non-addictive psychiatric drugs such as tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs), neuroleptics (haloperidol, chlorpromazine), and non-opioid analgesics (aspirin, acetaminophen) fail to produce conditioned place preference, and frequently produce conditioned place aversion.
• Therapeutic agents that suppress drug craving and relapse in clinical human populations (e.g., naltrexone for opioids and alcohol, buprenorphine, methadone maintenance) systematically extinguish or prevent the acquisition and expression of CPP in standardized Phillips-Fibiger protocols.

3. Convergent and Discriminant Validity

Convergent validity is verified by strong correlations between place preference magnitude and measures obtained in alternative reinforcement paradigms, such as intravenous drug self-administration (IVSA) breaking points under progressive ratio schedules, and current-intensity thresholds in intracranial self-stimulation (ICSS). Discriminant validity is demonstrated by dissociating motor locomotion from place preference: an animal can exhibit profound CPP for an opiate or low-dose sedative without displaying hyperlocomotor activity, and conversely, non-rewarding locomotor stimulants (e.g., high-dose caffeine or non-dopaminergic central stimulants) increase total distance traveled during post-tests without shifting the dwell-time preference score toward the paired compartment.

8. Reliability

In behavioral psychometrics and experimental ethology, the reliability of an apparatus-based testing protocol is determined through internal consistency across conditioning trials, inter-apparatus consistency, test-retest reliability across repeated probe sessions, and cross-laboratory reproducibility.

1. Test-Retest Reliability and Extinction Kinetics

Under standardized experimental conditions where drug conditioning has taken place, unreinforced test-retest assessments conducted over successive days display high stability of preference scores. In classic Phillips and Fibiger protocols, rats tested across consecutive 15-minute probe sessions in the absence of drug reinforcement maintain stable preference for the drug-paired chamber (Pearson correlation coefficients typically ranging between r = .72 and r = .86 between Test 1 and Test 2). With prolonged daily non-reinforced exposures (typically 5 to 10 consecutive sessions), the preference gradually undergoes Pavlovian extinction, mirroring the predictable extinction curves observed in classical conditioning and human cue-reactivity protocols.

2. Inter-Rater and Automated Tracking Reliability

Early iterations of the CPP assay relied on observer-scored manual stopwatches or microswitch floor depressors, which introduced potential inter-observer variability. Modern implementations utilize automated infrared photobeam arrays or high-resolution overhead digital video tracking systems (e.g., EthoVision, ANY-maze). Comparative studies between automated video tracking metrics and blinded human behavioral scoring demonstrate inter-method reliability coefficients exceeding r = .95 (intraclass correlation ICC > .92), confirming that the dwell time, velocity, and entry counts derived from the apparatus are exceptionally objective and free from scoring bias.

3. Experimental Boundary Conditions and Stability Factors

Methodological research highlights that the reliability of CPP is governed by several critical boundary variables:

• Chamber Bias: Biased versus unbiased apparatus configurations. In an unbiased apparatus, naive rodents display approximately equal baseline exploration (~50% time in each side compartment during pre-tests). When baseline bias exists, the statistical reliability of preference scores can be confounded by natural thigmotaxis or innate aversion, requiring careful counterbalanced or biased-assignment statistical modeling.
• Number of Conditioning Cycles: Protocols utilizing 4 to 8 drug-context pairings yield significantly higher effect sizes and lower within-group coefficient of variation (CV < 15%) compared to abbreviated 1- or 2-trial conditioning paradigms.

9. Factor Analysis and Structural Dimensionality

Although the Conditioned Place Preference assay produces continuous physical metrics (time, distance, transitions) rather than ordinal survey item responses, the structural dimensionality of CPP behavioral data has been systematically examined through Principal Component Analysis (PCA) and Exploratory Factor Analysis (EFA) in neuroethology and behavioral phenotyping. When multi-parameter behavioral matrices are recorded during CPP testing sessions, multivariate factor analyses consistently extract three to four distinct, orthogonal behavioral dimensions:

Confirmatory structural modeling across independent behavioral pharmacology datasets confirms that Factor 1 (Incentive Preference) remains psychometrically invariant across vehicle and drug-treated cohorts, demonstrating that drug-induced place preference operates as an analytically discrete construct that does not artificially collapse into, or stem from, general hyperactivity (Factor 2).

10. Instrument / Measurement Tool

The Conditioned Place Preference apparatus, as standardized by Phillips, Fibiger, and subsequent preclinical investigators, is an objective behavioral apparatus and experimental protocol. The system structure, procedural lifecycle, and data scoring architecture are detailed below:

Apparatus Physical Specifications

• Enclosure Structure: Typically configured as a two-compartment or three-compartment rectangular testing box constructed from high-grade acrylic, aluminum, or PVC.
• Dimensional Standards (Adult Rats): Total dimensions approximately 60–90 cm length × 25–35 cm width × 30–40 cm height. In a three-compartment model, two equal-sized conditioning chambers (e.g., 35 × 30 × 35 cm each) are separated by a smaller central neutral transit corridor (e.g., 15 × 30 × 35 cm) connected via guillotine or automated sliding doors.
• Multi-Sensory Context Differentiation:
  • Chamber A (Tactile & Visual): High-contrast black walls with white vertical stripes; stainless-steel grid floor composed of parallel bars (1.2 cm spacing).
  • Chamber B (Tactile & Visual): Solid white walls with black horizontal stripes or circular dots; textured wire mesh or perforated stainless steel floor.
  • Center Compartment: Neutral smooth grey walls and solid plexiglass floor, maintaining low ambient illumination to prevent animal entrapment due to thigmotaxis.
• Data Capture Technology: Infrared photo-beam emitter-detector matrices (spaced at 2.5–5 cm intervals across each chamber) or overhead ceiling-mounted digital CCD infrared video cameras synchronized to automated tracking software.

Standard Experimental Protocol (Phases)

• Phase 1: Habituation / Pre-Test (Day 1):
  • Guillotine doors are lifted, granting the animal unhindered access to all compartments for 15–20 minutes in a completely drug-free state.
  • Baseline dwell time in each chamber is recorded to establish initial compartment neutrality and identify pre-existing chamber bias.
• Phase 2: Associative Conditioning Phase (Days 2 to 9):
  • Doors are closed, isolating the chambers. Conditioning is conducted over alternating daily cycles.
  • CS+ Sessions: The animal receives the active pharmacological agent (e.g., d-amphetamine 1.5 mg/kg i.p.) and is immediately confined to its designated paired chamber for 30–45 minutes.
  • CS- Sessions: On alternate days, the animal receives vehicle injection (e.g., 0.9% physiological saline) and is confined to the opposing chamber for an identical duration.
  • Standard paradigms typically employ 4 drug pairings and 4 vehicle pairings in a counterbalanced presentation order.
• Phase 3: Preference Test Probe (Day 10):
  • Conducted 24 hours after the final conditioning session. The animal is tested completely drug-free.
  • Guillotine doors are raised, and the animal is placed into the central neutral zone with unrestricted access to all chambers for 15–20 minutes.
  • Continuous temporal and spatial coordinates are digitally acquired.

Mathematical Scoring and Operational Metrics

• Absolute Dwell Time ($T_{\text{CS+}}$ and $T_{\text{CS-}}$): Cumulative duration (seconds) spent inside the drug-paired chamber versus vehicle-paired chamber during the post-test.
• Preference Score ($Δ T$): Calculated as:$$\text{Preference Score} = T_{\text{CS+} \text{ (Post-Test)}} – T_{\text{CS+} \text{ (Pre-Test)}}$$A statistically significant positive value indicates conditioned place preference (reinforcement), whereas a significant negative value indicates conditioned place aversion.
• Preference Ratio ($PR$):$$PR = \frac{T_{\text{CS+}}}{T_{\text{CS+}} + T_{\text{CS-}}} \times 100$$
• Locomotor Activity Index: Total distance traveled (meters) and total horizontal photobeam interruptions during the probe test, utilized as a covariate to ensure motor differences do not confound spatial choice.

11. Permissions, Licensing, and Test Year

The Conditioned Place Preference experimental paradigm was formally developed and introduced into neuropharmacological research through landmark publications by Anthony G. Phillips and Hans C. Fibiger in 1979 (e.g., Phillips & Fibiger, 1979), with extensive behavioral and neuroanatomical standardizations published alongside Christina Spyraki in 1982 and 1983.

As an open scientific experimental paradigm and behavioral methodology, the Conditioned Place Preference assay is not proprietary. There are no individual licensing fees, copyrights, or royalties required to implement the behavioral conditioning protocol or construct the physical apparatus. Researchers are entirely free to build custom chambers or purchase commercial apparatuses from laboratory instrumentation vendors (e.g., Stoelting Co., Med Associates Inc., Panlab/Harvard Apparatus, Ugo Basile, Noldus Information Technology). Commercial hardware and integrated software suites (such as EthoVision XT or ANY-maze) require commercial purchase and proprietary software licenses from their respective manufacturers, but the intellectual design, theoretical principles, and operational testing protocols established by Phillips and Fibiger reside permanently in the public academic domain for global non-profit and commercial biomedical research.

12. References

The following foundational peer-reviewed studies document the inception, neuroanatomical validation, and psychometric development of the Phillips and Fibiger Conditioned Place Preference paradigm:

• Bardo, M. T., & Bevins, R. A. (2000). Conditioned place preference: What does it add to our knowledge of substance abuse? Psychopharmacology, 153(1), 31–43. https://doi.org/10.1007/s002130000569
• Carr, G. D., Fibiger, H. C., & Phillips, A. G. (1989). Conditioned place preference as a measure of drug reward. In J. M. Liebman & S. J. Cooper (Eds.), The Neuropharmacological Basis of Reward (pp. 264–319). Oxford University Press. https://doi.org/10.1093/oso/9780198521785.003.0007
• Phillips, A. G., & Fibiger, H. C. (1979). Decreased resistance to extinction after haloperidol: Implications for the role of dopamine in reinforcement. Pharmacology Biochemistry and Behavior, 10(5), 751–760. https://doi.org/10.1016/0091-3057(79)90326-1
• Prus, A. J., James, J. R., & Rosecrans, J. A. (2009). Conditioned place preference. In J. J. Buccafusco (Ed.), Methods of Behavioral Analysis in Neuroscience (2nd ed., pp. 59–76). CRC Press/Taylor & Francis. https://www.ncbi.nlm.nih.gov/books/NBK5229/
• Robinson, T. E., & Berridge, K. C. (1993). The neural basis of drug craving: An incentive-sensitization theory of addiction. Brain Research Reviews, 18(3), 247–291. https://doi.org/10.1016/0165-0173(93)90013-P
• Spyraki, C., Fibiger, H. C., & Phillips, A. G. (1982). Dopaminergic substrates of amphetamine-induced place preference conditioning. Brain Research, 253(1–2), 185–193. https://doi.org/10.1016/0006-8993(82)90685-0
• Spyraki, C., Fibiger, H. C., & Phillips, A. G. (1983). Attenuation of heroin-induced place preference by neuroleptics: Dissociation of neuroleptic effect on reward and motor activity. Psychopharmacology, 80(3), 279–283. https://doi.org/10.1007/BF00436173
• Tzschentke, T. M. (1998). Measuring reward with the conditioned place preference paradigm: A comprehensive review of drug effects, recent progress and new issues. Progress in Neurobiology, 56(6), 613–672. https://doi.org/10.1016/S0301-0082(98)00060-4
• Tzschentke, T. M. (2007). Measuring reward with the conditioned place preference (CPP) paradigm: Update of the last decade. Addiction Biology, 12(3–4), 227–462. https://doi.org/10.1111/j.1369-1600.2007.00070.x
• van der Kooy, D., Mucha, R. F., O’Shaughnessy, M., & Bucenieks, P. (1982). Reinforcing effects of brain stimulation and amphetamine in the conditioned place preference paradigm. Brain Research, 243(1), 107–117. https://doi.org/10.1016/0006-8993(82)91124-6

13. Items of the Scale