Abstract

Addictive behavior is a form of compulsive behavior: the relapsing pursuit of a substance or activity that persists despite mounting harm and a declining ability to stop. The incentive-sensitization theory dissociates wanting (the incentive salience a cue acquires) from liking (the pleasure a reward delivers), locating addiction in a sensitized wanting system that grows even as liking fades. The opponent-process or antireward account models a slide of hedonic set-point below baseline, so that use is driven less by reward than by relief of a worsening withdrawal. The actions-to-habits-to-compulsion account traces control from the ventral to the dorsal striatum, ending in punishment-insensitive compulsion; all three converge on weakened prefrontal control over a sensitized striatal circuit and extend to non-substance behavioral addictions such as gambling. Three interactive demonstrations trace the wanting–liking dissociation, the allostatic addiction cycle, and the growth of punishment resistance.

Keywords: addiction, incentive salience, reward, negative reinforcement, compulsion

A person keeps using a drug that has cost them their job, their health, and the pleasure it once gave. They intend to stop, succeed for a week, and relapse at the sight of a street corner or the sound of a name. What makes this behavior addictive is not the chemistry of the substance alone but the grammar of the action: it is repeated, it is hard to resist, it survives the loss of its reward, and it returns after long abstinence. The puzzle addiction poses to cognitive psychology is why a behavior should grow more compelling as it becomes less rewarding (Robinson & Berridge, 1993).

For most of the twentieth century addiction was framed either as moral failure or as the body's pursuit of a drug to stave off physical withdrawal. Neither survived contact with the evidence: people relapse long after withdrawal has passed, and they become addicted to activities — gambling, gaming — that have no withdrawal syndrome at all. The modern view, built from animal learning theory and human neuroimaging, recasts addiction as a disorder of the brain's reward and control systems, in which the neural machinery of ordinary motivated behavior is hijacked and progressively dysregulated (Volkow et al., 2016; Volkow & Morales, 2015).

Key Takeaways
  • Addictive behavior is the compulsive, relapsing pursuit of a reward despite harm — MeSH files it as a kind of compulsive behavior.
  • The incentive-sensitization theory separates wanting (incentive salience) from liking (hedonic pleasure): addiction is a sensitized wanting system, which is why craving outlives enjoyment.
  • The opponent-process / antireward account explains escalation as an allostatic drop in hedonic set-point, shifting the motive for use from reward toward relief of withdrawal — from positive to negative reinforcement.
  • The actions-to-habits-to-compulsion account traces control from the ventral to the dorsal striatum, ending in drug-seeking that persists despite punishment.
  • All three converge on weakened prefrontal control over a sensitized striatal circuit, and all extend to non-substance behavioral addictions such as gambling.

What Addictive Behavior Is

Addictive behavior is best defined by the same contrast that defines compulsion: a break with the ordinary logic of reward. Normal motivated action is performed because it is expected to deliver a valued outcome and is abandoned when the outcome loses value. Addictive action breaks this contingency in a characteristic temporal pattern. Early on, the behavior is genuinely rewarding and roughly goal-directed. With repetition it becomes habitual, triggered by cues rather than chosen for its consequences. In its established form it is compulsive: repeated rigidly, experienced as difficult to control, and continued despite the accumulation of serious harm (Everitt & Robbins, 2005; Lüscher et al., 2020).

Three features set addiction apart from a merely strong habit. The first is escalation: intake tends to grow over time, in dose or frequency, rather than stabilising. The second is withdrawal and negative affect during abstinence — not merely physical symptoms but a dysphoric, anxious state that itself motivates further use. The third is relapse: the propensity to return to the behavior, often triggered by a cue, stress, or a single re-exposure, persists for months or years after the behavior has stopped. A clinical diagnosis of a substance use disorder is made from this cluster — impaired control, social impairment, risky use, and pharmacological criteria such as tolerance — rather than from the quantity consumed (Volkow & Blanco, 2023).

This definition deliberately spans more than drugs. The same functional signature — escalating, cue-triggered, compulsive pursuit maintained despite harm — appears in pathological gambling, and is debated for gaming, eating, and other activities. That breadth is why addiction is increasingly treated as a disorder of behavior and its underlying motivational circuitry, not of any particular chemical (Grant et al., 2010; Potenza, 2006).

Figure 1

The Three-Stage Cycle of Addiction

The three-stage cycle of addiction A loop of three stages. Binge and intoxication leads to withdrawal and negative affect, which leads to preoccupation and anticipation, which returns to binge and intoxication. Each stage is labelled with its principal brain region. Binge / intoxication basal ganglia Withdrawal / negative affect extended amygdala extended amygdala Preoccupation / anticipation prefrontal cortex
Note. On Koob and Volkow's model, addiction is a cycle through three stages, each dominated by a different circuit: binge/intoxication (basal ganglia), withdrawal/negative affect (extended amygdala), and preoccupation/anticipation (prefrontal cortex). With each turn the positive drive weakens and the negative drive deepens. Adapted from Koob and Volkow (2016). Demo 2 runs the deepening cycle interactively.

Types of Addictive Behavior

In the Medical Subject Headings vocabulary, Addictive Behavior (D016739) is filed as a narrower descriptor of compulsive behavior (D003192), and MeSH records four narrower descriptors beneath it. MeSH is an indexing classification built for retrieving the biomedical literature, not a theory of how addiction is organised; its hierarchy reflects historical indexing practice rather than the mechanistic distinctions the experimental literature draws, and the two ways of carving the domain are orthogonal and should not be conflated.

Narrower MeSH descriptor What it denotes
Compulsive Exercise Physical exercise pursued compulsively and continued despite injury or impairment, often entangled with eating disorders.
Compulsive Sexual Behavior Disorder A persistent failure to control intense, repetitive sexual impulses, resulting in distress or harm; recognised in ICD-11 as an impulse-control disorder.
Food Addiction A proposed addiction-like pattern of compulsive consumption of highly palatable food, showing cravings and loss of control analogous to substance use.
Technology Addiction Compulsive overuse of digital technologies — internet, gaming, smartphones — maintained despite functional impairment.

None of these four descriptors yet has its own article on this site, so the terms are listed here but not linked. Their grouping under Addictive Behavior encodes a substantive claim: that compulsive exercise, sexual behavior, eating, and technology use can share the motivational machinery of drug addiction even though no drug is involved. This is the hypothesis of behavioral addiction — that the defining process is the dysregulation of reward-related behavior, not the pharmacology of a substance — and it remains the most actively debated boundary of the construct (Grant et al., 2010; Potenza, 2006).

Three Accounts of Addiction

Cognitive psychology offers three accounts of how a reward-driven behavior becomes addictive. They are not rivals so much as descriptions of the same disorder from the standpoints of incentive learning, of affective homeostasis, and of the control of action. All three locate the fault in the balance between a sensitized striatal reward system and a weakened prefrontal control system, and all three have been consolidated into the influential brain-disease framing of addiction (Volkow et al., 2016; Goldstein & Volkow, 2011).

Incentive-sensitization: wanting versus liking

The incentive-sensitization theory begins from a dissociation that ordinary language hides: wanting a reward and liking it are produced by different brain systems and can move in opposite directions. Liking is the hedonic impact of a reward — the pleasure actually delivered. Wanting, or incentive salience, is the motivational pull a reward and its cues acquire — the force that makes them attractive and that organises behavior toward them. Normally the two travel together, but they are separable, and addiction is the case where they come apart (Robinson & Berridge, 1993; Berridge & Robinson, 2016).

The theory holds that repeated use of an addictive drug sensitizes the mesolimbic dopamine system that computes wanting, so that drugs and drug cues acquire ever more incentive salience. Liking, meanwhile, does not sensitize and often declines with tolerance. The result is a widening gap: the addicted person wants the drug intensely — craves it, is captured by its cues — while enjoying it less and less. This dissociation explains the otherwise paradoxical core of addiction, that craving can be overwhelming even when the drug no longer gives pleasure, and that cues can trigger relapse after long abstinence because the sensitized wanting is durable. Demo 1 drives the two systems apart across repeated exposures.

Wanting pulls away from liking

Incentive-sensitization holds that repeated use sensitizes the wanting system (incentive salience) while the liking system (hedonic pleasure) habituates with tolerance. Raise the sensitization rate and watch craving climb as enjoyment falls — the widening gap is why an addicted person can want a drug intensely while liking it less and less.

Sensitization rate of wanting (s)0.35
0.000.250.500.751.00strength (0–1)use number (n)wantingliking
With sensitization rate s = 0.35, by use 12 wanting has risen to 1.00 while liking has fallen to 0.19. Wanting overtakes liking at use 2: past this point the drug is craved more than it is enjoyed, the dissociation that defines the theory.

The three-stage cycle and the antireward system

The second account starts from affect and its regulation. On the opponent-process view, every hedonic event triggers an opposing reaction that restores balance; with a drug, the pleasurable a-process is followed by a negative b-process. Repeated use strengthens and prolongs the b-process while the a-process stays fixed or weakens, so the net pleasure of each use shrinks and the baseline between uses sinks below where it started. This downward drift of hedonic set-point is allostasis — stability achieved only at a progressively worse operating point — and it recruits a dedicated antireward system in the extended amygdala, which generates the anxiety and dysphoria of withdrawal (Koob & Le Moal, 2008).

Koob and Volkow assembled this into a three-stage cycle (Figure 1): binge/intoxication, driven by reward in the basal ganglia; withdrawal/negative affect, driven by the antireward system of the extended amygdala; and preoccupation/anticipation, driven by craving and failing executive control in the prefrontal cortex (Koob & Volkow, 2016). The engine of escalation is the shift in what motivates use. Early use is positively reinforced — it is taken for the reward. Established use is increasingly negatively reinforced — it is taken to escape the worsening withdrawal the drug itself has created. The worked example quantifies this crossover, and Demo 2 traces the deepening cycle.

From reward to relief: the allostatic cycle

Each use delivers a fixed acute pleasure A = 100, but triggers an opposing b-process that grows toward a ceiling 60: bn = 60[1 − (1 − g)n]. Net pleasure is Pn = A − bn; the withdrawal nadir between uses is |Wn| = bn. Where they cross, relief of withdrawal overtakes reward as the motive to use.

b-process growth rate (g)0.30
0255075100magnitude (units)use number (n)n = 5.02net pleasurewithdrawal
With g = 0.30, net pleasure runs 82.0 → 69.4 → 60.6 over the first three uses while withdrawal deepens 18.0 → 30.6 → 39.4. Reward and relief cross at use 5.02 — from there on the drug is taken more to escape withdrawal than to feel pleasure, the shift from positive to negative reinforcement. (This is the article's worked example.)

From reward to habit to compulsion

The third account describes addiction as a progressive loss of voluntary control over action, tracked through the striatum. Instrumental behavior is governed by two systems: a goal-directed system, sensitive to the value of the outcome and served by the ventral striatum, and a habit system, triggered by stimuli and served by the dorsal striatum. Everitt and Robbins proposed that addiction is a devolution of control from the first to the second and finally beyond it — from actions, to habits, to compulsion — as drug-seeking becomes stimulus-bound and then insensitive even to punishment (Everitt & Robbins, 2005; Everitt & Robbins, 2016).

The defining experimental signature of the compulsive end state is resistance to punishment: an animal, or a person, continues to seek the drug even when seeking is reliably paired with an aversive consequence. Only a subset of users reach this stage, which is one reason addiction is a disorder of the minority of those who ever use. The transition depends on weakened top-down control: the prefrontal circuitry that would normally override a sensitized striatal urge is itself impaired in addiction, a syndrome of impaired response inhibition and misattributed salience that ties this account to the other two (Goldstein & Volkow, 2011; Lüscher et al., 2020). Because the same actions-to-habits-to-compulsion trajectory underlies behavioral addictions, which have no pharmacology to blame, it is the strongest bridge between substance and non-substance addiction. Demo 3 grows the resistance to punishment as addiction progresses.

Seeking that survives punishment

As addiction progresses, control of drug-seeking devolves from a goal-directed system toward a stimulus-bound compulsive one, and sensitivity to punishment falls with it. Move the behaviour along that transition and apply a punishment: early on it suppresses seeking almost entirely, but at the compulsive end seeking persists regardless — the experimental hallmark of compulsion.

Stage of the transition to compulsionhabitual, early
0.00.51.0seeking rateno punishment1.00under punishment0.17
At the habitual, early stage, punishment leaves seeking at 0.17 of its unpunished rate — a partial suppression — the behaviour is sliding out of punishment's reach as control becomes habitual.

Worked Example

Consider the opponent-process account and the allostatic drop it predicts. Model the acute a-process — the drug's direct pleasurable effect — as a fixed A = 100 units. The opposing b-process grows toward an asymptote with repeated use: on the n-th use

bn = Bmax [1 − (1 − g)n]

with ceiling Bmax = 60 and growth rate g = 0.3. The net pleasure of a use is Pn = A − bn, and the withdrawal nadir between uses — how far affect sinks below baseline — is Wn = −bn. Then:

- Use 1: b1 = 60(1 − 0.7) = 18, so P1 = 82, W1 = −18 - Use 2: b2 = 60(1 − 0.49) = 30.6, so P2 = 69.4, W2 = −30.6 - Use 3: b3 = 60(1 − 0.343) = 39.4, so P3 = 60.6, W3 = −39.4

Pleasure falls and withdrawal deepens with every use. The telling quantity is why the person uses. Early on, the pull of reward Pn exceeds the pull of relief |Wn|; later the order reverses. The crossover is where |Wn| = Pn, i.e. bn = A − bn, so bn = 50:

60[1 − (0.7)n] = 50 ⟹ (0.7)n = 0.1667 ⟹ n = ln(0.1667) / ln(0.7) ≈ 5.02

So by about the fifth use, relief of withdrawal overtakes reward as the dominant motive — the behavioural signature of the shift from positive to negative reinforcement. This is the allostatic engine of escalation: each use buys a smaller pleasure and leaves a deeper deficit, and the deficit itself becomes the reason to use again. Demo 2 traces Pn and Wn across uses and marks exactly this crossover.

Discussion

The three accounts are now read as complementary levels of one story. Incentive-sensitization explains craving and its persistence — why wanting outlives liking and why cues trigger relapse. The opponent-process account explains escalation and withdrawal — why intake grows and why abstinence is aversive. The actions-to-habits-to-compulsion account explains the loss of control — why seeking becomes automatic and finally punishment-insensitive. They share a common anatomy: a sensitized, habit-forming striatal system and a weakened prefrontal control system, the two-part imbalance at the centre of the brain-disease model of addiction (Volkow et al., 2016; Volkow & Morales, 2015).

Framing addiction as a brain disorder has been clinically consequential — it underwrites pharmacological and neuromodulatory treatment and counters the view of addiction as mere weakness of will — but it is not without critics, who note that most people who meet criteria for a substance use disorder recover, often without treatment, and that social and economic context powerfully shapes both onset and recovery (Volkow & Blanco, 2023). The tension between a disease that is mechanistically real and a behavior that remains responsive to incentive and circumstance is the productive fault line of the field, and it is sharpest at the boundary of behavioral addiction, where the absence of any drug forces the question of what, exactly, is being dysregulated (Potenza, 2006).

Current Directions

Several fronts are active. The first is the formal recognition of behavioral addictions: gambling disorder is now classified alongside substance use disorders, gaming disorder has entered ICD-11, and the machinery of incentive-sensitization and compulsion is being tested on these drug-free cases to see how far the analogy holds (Grant et al., 2010). The second is a sustained effort to understand why only a minority of users become addicted, combining the circuit model of the transition to compulsion with individual differences in impulsivity, sensitization, and prefrontal control to predict who is vulnerable (Lüscher et al., 2020). The third is treatment aimed at the control side of the imbalance — neuromodulation of prefrontal circuits, and behavioral training to strengthen goal-directed over habitual control — alongside a continuing reassessment of how the brain-disease model should be integrated with the social determinants of use and recovery (Volkow & Blanco, 2023). Whether a single framework can hold the neural, psychological, and social levels together without collapsing into one of them is the open question the field now faces.

Common Misconceptions

Addiction is just seeking pleasure.
In its established form addiction is driven more by wanting than liking and more by relief of withdrawal than by reward. People often crave a drug intensely while enjoying it little, which positive-pleasure accounts cannot explain (Berridge & Robinson, 2016).
Addiction is the body avoiding physical withdrawal.
Physical withdrawal matters, but relapse occurs long after it has passed, and addictions form to activities with no withdrawal syndrome at all. The durable driver is a sensitized motivational system and a dysphoric antireward state, not acute physical symptoms (Robinson & Berridge, 1993).
Only drugs can be addictive.
Gambling meets the criteria for an addictive disorder with no substance involved, and MeSH files compulsive exercise, sexual behavior, eating, and technology use under addictive behavior. The debated claim is that the shared process is dysregulated reward-related behavior (Potenza, 2006).
Anyone who uses a drug enough becomes addicted.
Only a minority of users progress to the compulsive, punishment-insensitive end state; vulnerability depends on sensitization, impulsivity, and the integrity of prefrontal control. Addiction is a disorder of susceptible individuals, not an inevitable consequence of exposure (Everitt & Robbins, 2016).

Glossary

Allostasis.
The achievement of stability through change: in addiction, the hedonic set-point is defended only at a progressively lower level, so the baseline mood between uses sinks below its original value.
Antireward system.
Circuitry centred on the extended amygdala that generates the negative affect of withdrawal; its recruitment is the neural basis of the opponent b-process and of negative reinforcement in addiction.
Behavioral addiction.
An addiction to an activity rather than a substance — gambling, gaming — proposed to share the motivational machinery of drug addiction; the hypothesis that defines addiction by process rather than pharmacology.
Compulsion.
Repetitive action continued despite harm and experienced as difficult to resist; the end state of addiction, marked experimentally by resistance to punishment.
Craving.
An intense, often cue-triggered desire for a drug or activity; in incentive-sensitization theory it is the subjective face of sensitized wanting, which is why it can intensify across abstinence rather than fade.
Incentive salience.
The motivational property — wanting — that a reward and its cues acquire, making them attractive and capable of organising behavior; sensitized in addiction independently of pleasure.
Negative reinforcement.
Strengthening of a behavior by the removal of an aversive state; established drug use is increasingly negatively reinforced as it comes to relieve the withdrawal the drug itself produced.
Opponent process.
The theory that each hedonic reaction (the a-process) triggers an opposing reaction (the b-process) restoring balance; repeated drug use strengthens the b-process, shrinking pleasure and deepening withdrawal.
Relapse.
The return to addictive behavior after a period of abstinence, often triggered by a cue, stress, or re-exposure; its persistence long after withdrawal reflects durable sensitized wanting.
Resistance to punishment.
Continued pursuit of a reward even when it is reliably paired with an aversive consequence; the experimental hallmark of compulsive, as opposed to merely habitual, drug-seeking.
Sensitization.
The increase in a response with repeated exposure; the mesolimbic wanting system is sensitized by addictive drugs, the opposite of the tolerance shown by the pleasure (liking) system.
Tolerance.
The need for more of a drug to achieve the same effect, as the response to a fixed dose declines with repeated use; a pharmacological criterion of substance use disorder.
Wanting versus liking.
The dissociation at the heart of incentive-sensitization: wanting is the motivation to obtain a reward, liking the pleasure it delivers; addiction amplifies wanting while liking fades.
Withdrawal.
The aversive physical and affective state that follows cessation of use in a dependent person; relief of its negative affect becomes a reinforcer that drives continued use, and that affective component outlasts the physical one.

Key Researchers

Kent C. Berridge

(Department of Psychology, University of Michigan). With Robinson, formulated the incentive-sensitization theory and the dissociation of wanting from liking, reshaping how addiction's craving is understood. ORCID · Wikipedia

Barry J. Everitt

(Department of Psychology, University of Cambridge). With Robbins, framed the actions to habits to compulsion account of addiction and mapped the shift of behavioural control from the ventral to the dorsal striatum. ORCID · Wikipedia

George F. Koob

(National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health). Developed the opponent-process and antireward account of addiction and, with Volkow, the three-stage neurocircuitry model of the addiction cycle. Faculty

Marc N. Potenza

(Department of Psychiatry, Yale School of Medicine). A leading researcher on behavioral addictions, especially gambling, who argued for including non-substance conditions within the addiction framework. Faculty · Google Scholar

Trevor W. Robbins

(Department of Psychology, University of Cambridge). With Everitt, built the influential account of how reward-driven drug-seeking becomes compulsive, and leads the dimensional study of impulsivity and compulsivity in addiction. ORCID · Wikipedia

Nora D. Volkow

(National Institute on Drug Abuse, National Institutes of Health). Through neuroimaging, established the role of dopamine and prefrontal dysfunction in addiction and shaped the modern brain-disease model and its clinical translation. ORCID · Wikipedia

Frequently Asked Questions

What makes a behavior addictive rather than just a strong habit?

A habit is automatic but harmless and can be dropped once it stops paying. A behavior is addictive when it escalates, produces a dysphoric withdrawal during abstinence, is pursued compulsively despite serious harm, and is prone to relapse long after it has stopped. The defining feature is this compulsive persistence against the person's own interest.

What is the difference between wanting and liking in addiction?

Wanting is the motivational pull, the craving, that a reward and its cues acquire, while liking is the pleasure the reward actually delivers. Addictive drugs sensitize the wanting system but not the liking system, so an addicted person can crave a drug intensely while enjoying it less and less. This dissociation explains craving that outlasts pleasure.

Why do people relapse after they have stopped and are no longer in withdrawal?

Because the sensitized wanting system is durable. Cues associated with past use retain and even increase their incentive salience over time, so a street, a smell, or a stressful moment can trigger intense craving and relapse months or years into abstinence, long after any physical withdrawal has resolved.

How does addiction shift from pleasure-seeking to escape?

Repeated use strengthens an opposing, antireward process that deepens withdrawal and lowers baseline mood. Early use is positively reinforced, taken for reward, but as the deficit grows, use becomes negatively reinforced, taken to relieve the withdrawal the drug itself created. The motive shifts from reward to relief.

Can activities like gambling be genuine addictions?

Increasingly they are classified as such. Gambling disorder is now grouped with substance use disorders, and gaming disorder has entered the ICD-11. The argument is that the core of addiction is dysregulated reward-related behavior (sensitized wanting, loss of control, compulsion) that can occur with no drug involved.

Why do only some people who use drugs become addicted?

The compulsive, punishment-insensitive end state is reached by only a minority of users. Vulnerability depends on how strongly the wanting system sensitizes, on trait impulsivity, and on the integrity of the prefrontal control that would otherwise override a drug urge. Addiction is a disorder of susceptible individuals, not an automatic result of exposure.

What does it mean to call addiction a brain disease?

It means addiction involves lasting changes in reward and control circuits (a sensitized striatal system and a weakened prefrontal one) that make drug-seeking compulsive and hard to reverse by will alone. The framing supports medical treatment, though critics stress that most people recover and that social context strongly shapes the course.

Is addiction a kind of compulsive behavior?

Yes. MeSH files addictive behavior as a narrower category of compulsive behavior, and the actions-to-habits-to-compulsion account holds that established addiction is literally compulsive: drug-seeking becomes rigid, stimulus-bound, and continued despite punishment, the same functional signature that defines compulsion generally.

References

Berridge, K. C., & Robinson, T. E. (2016). Liking, wanting, and the incentive-sensitization theory of addiction. American Psychologist, 71(8), 670–679. https://doi.org/10.1037/amp0000059

Everitt, B. J., & Robbins, T. W. (2005). Neural systems of reinforcement for drug addiction: From actions to habits to compulsion. Nature Neuroscience, 8(11), 1481–1489. https://doi.org/10.1038/nn1579

Everitt, B. J., & Robbins, T. W. (2016). Drug addiction: Updating actions to habits to compulsions ten years on. Annual Review of Psychology, 67, 23–50. https://doi.org/10.1146/annurev-psych-122414-033457

Goldstein, R. Z., & Volkow, N. D. (2011). Dysfunction of the prefrontal cortex in addiction: Neuroimaging findings and clinical implications. Nature Reviews Neuroscience, 12(11), 652–669. https://doi.org/10.1038/nrn3119

Grant, J. E., Potenza, M. N., Weinstein, A., & Gorelick, D. A. (2010). Introduction to behavioral addictions. The American Journal of Drug and Alcohol Abuse, 36(5), 233–241. https://doi.org/10.3109/00952990.2010.491884

Koob, G. F., & Le Moal, M. (2008). Addiction and the brain antireward system. Annual Review of Psychology, 59, 29–53. https://doi.org/10.1146/annurev.psych.59.103006.093548

Koob, G. F., & Volkow, N. D. (2016). Neurobiology of addiction: A neurocircuitry analysis. Lancet Psychiatry, 3(8), 760–773. https://doi.org/10.1016/S2215-0366(16)00104-8

Lüscher, C., Robbins, T. W., & Everitt, B. J. (2020). The transition to compulsion in addiction. Nature Reviews Neuroscience, 21(5), 247–263. https://doi.org/10.1038/s41583-020-0289-z

Potenza, M. N. (2006). Should addictive disorders include non-substance-related conditions? Addiction, 101(Suppl 1), 142–151. https://doi.org/10.1111/j.1360-0443.2006.01591.x

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

Volkow, N. D., Koob, G. F., & McLellan, A. T. (2016). Neurobiologic advances from the brain disease model of addiction. New England Journal of Medicine, 374(4), 363–371. https://doi.org/10.1056/NEJMra1511480

Volkow, N. D., & Blanco, C. (2023). Substance use disorders: A comprehensive update of classification, epidemiology, neurobiology, clinical aspects, treatment and prevention. World Psychiatry, 22(2), 203–229. https://doi.org/10.1002/wps.21073

Volkow, N. D., & Morales, M. (2015). The brain on drugs: From reward to addiction. Cell, 162(4), 712–725. https://doi.org/10.1016/j.cell.2015.07.046