Behavioural Models and Theories of Drug Addiction – Lecture Review

Reward Circuitry

• Mesolimbic dopamine (DA) pathway is the central “reward circuit”
  • Ventral tegmental area (VTA) → nucleus accumbens (NAc); receives glutamatergic (excitatory) and GABAergic (inhibitory) inputs
  • Dopamine release from VTA terminals in NAc is increased by every known drug of abuse, amplifying the salience of drug-related cues
• Natural rewards (food, water) elevate NAc DA to roughly 140\text{–}150\% of baseline when organisms are deprived
• Drugs of abuse can produce up to

Why Drug Abuse Matters

• Public-health impact (Australia)
  • 20\text{–}30\% of all cancers attributable to smoking
  • Alcohol & illicit drugs ≈ 1 in 20 deaths
  • Opioids account for 41\% of drug-related mortality; amphetamine deaths are rising
• Treatment limitations
  • Alcohol: poor compliance
  • Nicotine: significant side-effects
  • Opioids: substitution therapies (e.g., methadone)
  • Stimulants: no approved pharmacotherapy
• Functional consequence: chronic drug use remodels decision-making circuitry; an addict’s brain does not make choices the way a non-addict’s brain does

Research Approaches & Animal Models

• Rationale
  • Rodents, primates and humans prefer similar psychoactive drugs
  • Allows investigation of reinforcement, motivation, relapse and neurotoxicity
• Conditioned Place Preference (CPP)
  • Pavlovian pairing of drug with one compartment; saline with the other
  • Preference index reflects appetitive/aversive properties; dose-dependent
  • Limits: confounds (locomotion, attention, memory), passive exposure, Pavlovian ≠ instrumental validity
• Intravenous Self-Administration (IVSA)
  • Jugular catheter → rapid drug bolus mimicking inhalation/smoking kinetics
  • Rats control intake; gold-standard for modeling reinforcement & motivation
• Intracranial Microdialysis
  • Semi-permeable probe samples extracellular neurochemicals
  • First convergent evidence: all abused drugs elevate DA in NAc of freely moving rats
• Fast-Scan Cyclic Voltammetry (FSCV)
  • Carbon-fiber electrode; millisecond resolution for DA, 5\text{-HT}, norepinephrine
  • No viral transfection required; superior temporal precision versus microdialysis
• d-Light Photometry
  • Viral delivery of genetically encoded DA sensor + 485\,\text{nm} excitation
  • High temporal and spatial resolution; identifies cell-type specific (e.g., D1-MSNs) encoding of drug-associated cues
• Screening emerging drugs (e.g., mephedrone) reveals DA/5-HT release similar to MDMA & amphetamine, predicting strong reinforcing effects

Addictive Potential (Henningfield Rankings)

• Separate dimensions: Reinforcement (acute euphoria) vs Dependence (compulsive craving)
• Nicotine ranked hardest to quit; \approx90\% of smokers are daily users
• Alcohol: only \approx15\% of users develop dependence
• Cocaine: \approx8\% transition to dependency
• Individual & situational factors modulate risk (e.g., purpose of use, life stressors)

Method of Administration Is Critical

• “Master–yoked” IVSA paradigm demonstrates:
  • Only the rat with voluntary control over infusions shows robust NAc DA elevations & later addiction-like behavior
  • Identical pharmacology in the yoked partner lacks the same motivational impact

Historical & Diagnostic Perspectives

• Shift from viewing drug taking as crime → disease → mental health disorder emphasizing positive reinforcement via DA
• DSM-5
  • Substance-related disorders & non-substance (gambling); emerging: caffeine, Internet gaming, possible mobile-phone overuse
  • Example: Stimulant Use Disorder criteria include larger-than-intended use, craving, tolerance, withdrawal, etc.

Development of Addiction (General Framework)

1. Recreational / sporadic use
2. Intensified / sustained / escalated use
3. Loss of control → addiction (compulsive seeking despite harm)
   • Interaction of individual vulnerability × drug exposure

Aberrant Learning Theories

• Drugs induce persistent changes in normal learning/memory circuits

Incentive Sensitisation (Robinson & Berridge)

• Distinction between “liking” (hedonic impact) and “wanting” (incentive salience)
  • Sensitisation selectively amplifies “wanting” without necessarily increasing “liking”
• Key components
  1. Addictive drugs enhance mesocorticolimbic DA
  2. DA assigns salience to cues/contexts predictive of reward
  3. Repeated drug exposure → long-lasting hypersensitivity of this system
• Empirical support
  • Amphetamine-sensitised rats press more for a conditioned stimulus (CS^+) paired with sucrose
  • Nicotine-paired cues maintain responding in absence of drug
  • Human smokers show heightened cue-induced craving for cigarettes & food (Mahler & de Wit 2010)

Sign-Tracking vs Goal-Tracking (Individual Differences)

• Sign-trackers (ST) approach the cue itself; goal-trackers (GT) approach the location of reward delivery
  • ST rats: larger phasic DA to CS; more cue-induced relapse after extinction
  • Flagel et al. 2011: High responders to novelty (bHR) → ST phenotype; low responders (bLR) → GT phenotype

Habit Formation

• Repetition shifts control from Action–Outcome (A–O) to Stimulus–Response (S–R) associations
• Corbit et al. 2014: Cocaine sensitisation accelerates habit emergence; behaviour becomes insensitive to outcome devaluation
• Clemens et al. 2014: After extended nicotine training, proximal cues automatically trigger seeking even when nicotine is devalued (paired with nausea)
• Distinction Habit vs Compulsion:
  • Habit: can revert to goal-directed under certain conditions
  • Compulsion: persistent, stereotyped behaviour despite negative consequences

Compulsive Drug-Seeking Models

• Pelloux et al.: 50 % of trials deliver foot-shock instead of cocaine; with extended training rats continue seeking despite shock
• Singer & Robinson 2018: Creativity in obtaining drugs suggests drug-seeking is not purely habitual

Piazza “3-Criterion” Rat Model (DSM-IV translation)

1. Persistence in drug seeking during non-drug periods
2. Resistance to punishment (shock) while self-administering
3. High motivation (break-point on progressive ratio)

• 17\% of rats met all three → mirrors human prevalence of severe addiction
• Addiction-prone rats show enduring loss of NAc long-term depression (LTD)

Frontostriatal Dysfunction & Impulsivity

• Winstanley et al. 2010: High impulsivity → vulnerability; linked to molecular changes in prefrontal cortex (PFC)
• Volkow et al. 2008: Cocaine addicts show decreased striatal D_2 receptors & hypometabolism in OFC, anterior cingulate, dorsolateral PFC → reduced inhibitory control
• Behavioral phenotypes
  • High-impulsive (HI) rats: escalate to compulsive cocaine seeking (3 criteria)
  • Novelty-responsive (HR) predicts acquisition, but impulsivity predicts compulsion
  • Implications for ADHD remain unresolved

Opponent Process / Hedonic Allostasis (Koob)

• Drug use recruits anti-reward (stress) systems over time → dysphoria
• Solomon’s A-process (positive affect) vs B-process (negative affect) model
  • Initial binge/intoxication dominated by positive reinforcement
  • Repeated use → withdrawal/negative affect & preoccupation stage dominated by negative reinforcement
• Evidence
  • Ettenberg et al. 2004: CPP immediately post-cocaine injection; conditioned place aversion 15\,\text{min} later → opponent process within one administration
• Critique: robust mainly under extended-access paradigms; limited human translation

Addiction as a Choice? (Hart, Heyman)

• Epidemiology: 80\text{–}90\% of crack/meth users do not develop addiction
• Clinical observations: Inpatient addicts often choose \$20 over a dose of crack → rational decision-making when alternatives exist
• Natural history
  • Highest remission rate of any psychiatric disorder
  • Median “half-life” to remission:
  • Cocaine \approx4\,\text{yr}
  • Marijuana \approx6\,\text{yr}
  • Alcohol \approx16\,\text{yr} (legal/social factors)
  • By age 30 most who will remit have done so without treatment
• Rat choice experiments (Ahmed et al. 2013)
  • Two levers: cocaine vs sucrose
  • Majority choose sucrose even after weeks of cocaine exposure; 10\text{–}20\% persist with cocaine → parallels human minority
  • Similar findings for heroin & nicotine (rats prefer even low \text{saccharin} concentrations over nicotine)
  • OFC neuronal coding correlates with individual drug vs saccharin preference
  • Social reward paradigm: rats often prefer access to conspecific over heroin

Practical & Ethical Implications

• Addiction is heterogeneous; single-mechanism models (e.g., compulsion) risk oversimplification and may impede tailored interventions
• Over-reliance on “drug use despite punishment” as sole measure of compulsion questioned (five critiques listed in transcript)
• Policy shift from punitive to treatment-oriented approaches justified by neuroscientific evidence of impaired choice circuitry

Numerical & Statistical References (Consolidated)

• Cancer from smoking: 20\text{–}30\%
• Drug-related deaths: \tfrac1{20} of all deaths
• Opioid share of deaths: 41\%
• Daily smokers: 90\%; occasional smokers 10\%
• Alcohol dependence prevalence: 15\%
• Cocaine dependence prevalence: 8\%
• Natural-reward DA increase: 140\text{–}150\% baseline
• Drug-induced DA increase: <10\times baseline
• Rats meeting all 3 Piazza criteria: 17\% of sample
• Choice studies: 10\text{–}20\% of rats persist with drug over sucrose

Key Take-Home Points

• All abused drugs converge on DA elevation in NAc, but pharmacodynamics & subjective effects differ
• Animal models (CPP, IVSA, microdialysis, FSCV, photometry) reliably predict human abuse liability & reveal circuit mechanisms
• Multiple, sometimes competing theories explain transition to addiction: incentive sensitisation, habit formation, hedonic allostasis, frontostriatal dysregulation, and rational choice models
• Individual differences (genetic, personality, environment) critically modulate progression; minority become compulsively addicted
• Treatment & policy must acknowledge heterogeneity and avoid one-size-fits-all assumptions