The field of addiction research is becoming more and more complex, and many hypotheses have been proposed to account for the transition from recreational use to impulsive consumption and to the last stage of this chronic, relapsing disease: compulsive use and addiction. Redish et al. have reviewed some of these theories and have proposed a classification under eight categories in such a way that some researchers will be surprised to find themselves listed under this or that category. The theories cited each stress different aspects, functional or neuropsychological, and different phase of the process, or consider either physiological mechanisms or structural neurobiology. In the target article, Redish et al. propose one more theory, which is more specific and cognitively oriented: the process of decision (decision-making) is hypothesized to be a “unified framework for addiction” and to be operational to provide a classification of potential vulnerabilities. From a Herculean analysis of the literature, but from this restricted point of view, they have identified ten potential different constitutive vulnerabilities.

Scientific analysts alert us about the breaking down and fragmentation of knowledge, a crisis due in part to the reductionism inherent in modern scientific progress. What is needed is to turn toward a more difficult task: to try and propose holistic theories and to conform to the principle of parsimony. Entities should not be multiplied without necessities, according to the principle of Occam's razor. Moreover, most authors now agree about the reality of a common clinical syndrome for all the drug and non-drug addictions (Goodman Reference Goodman1990; Reference Goodman2008) and, underlying it, a common set of neuronal systems, whose dysregulations is supposed to be responsible for the set of symptoms (see Koob & Le Moal Reference Koob and Le Moal1997; 2001; 2006). The question is to know at which stage the process is examined. It seems that Redish et al. are considering the stage of addiction when maladaptive choices are made in spite of their deleterious consequences, whereas vulnerability is generally studied as an intrinsic factor operating at the beginning of the process, accounting for the huge individual differences in the propensity to move toward impulsive drug-taking or gambling (Anthony et al. Reference Anthony, Warner and Kessler1994; Piazza & Le Moal Reference Piazza and Le Moal1996; Piazza et al. Reference Piazza, Deminière, Le Moal and Simon1989; Substance Abuse and Mental Health Services Administration 2003).

At one moment of the process, there is a passage from impulsive control disorder to compulsive disorder – from a stage where increasing tension and arousal occur before the impulsive act, with pleasure, gratification, or relief during the act, followed by regret or guilt, to a stage of recurrent and persistent thoughts (obsessions) that cause marked anxiety and stress followed by repetitive behaviors (compulsions) that are aimed at preventing or reducing distress. The first stage is most closely associated with positive reinforcement (pleasure, gratification); the compulsive stage is most closely associated with negative reinforcement and relief of anxiety and/or stress (Koob & Le Moal Reference Koob and Le Moal1997). Addiction involves persistent plasticity in the activity of neuronal circuits mediating a decreased function of the brain reward system and a recruitment of anti-reward systems, now well identified, driving aversive states (Koob & Le Moal Reference Koob and Le Moal2005; Reference Koob and Le Moal2008). For the purpose of this commentary, the withdrawal/negative affect stage can be defined as the presence of motivational signs of withdrawal in humans, that is, chronic instability, emotional pain, malaise, dysphoria, and loss of motivation for natural rewards. As dependence and withdrawal develop, brain anti-reward systems are recruited (Koob & Le Moal Reference Koob and Le Moal2008). Another critical problem is chronic relapse in which addicts return to compulsive drug-taking after acute withdrawal; relapse corresponds to the preoccupation/anticipation stage of the addiction cycle just outlined. A unified framework for addiction cannot avoid the fact, well documented from clinical observations, that affects and emotions are important, if not central, neuropsychological dimensions in this human condition. Needless to say, these dimensions interact with the process of decision-making.

All the neurobiological theories of addiction (we refer to the last stage of the process) agree (see Koob & Le Moal Reference Koob and Le Moal2006) about an interconnected network of regions and neuronal systems including subcortical and cortical structures (Everitt & Wolf Reference Everitt and Wolf2002; Goldstein & Volkow Reference Goldstein and Volkow2002; Hyman & Malenka Reference Hyman and Malenka2001; Kalivas & McFarland Reference Kalivas and McFarland2003; Koob & Le Moal Reference Koob and Le Moal2001; See et al. Reference See, Fuchs, Ledford, McLaughlin, Shinnick-Gallagher, Pitkanen, Shekhar and Cahill2003). All agree also about the fact that the prefrontal region (orbitofrontal, medial-prefrontal, prelimbic/cingulate), associated with the basolateral amygdala, is more and more dysregulated as the addictive process progresses. These two regions are key elements for some of the symptoms and for the dysregulation of the motivational systems described earlier, and finally for the affective negative state from which addicts suffer so much. A pure cognitive approach of the syndrome does not account for what is seen in the real world. Decision-making, impulse control, and loss of willpower to resist drugs (Bechara Reference Bechara2005) are of course very important, and a correlative deactivation of the prefrontal cortex will produce enormous consequences. Drugs, games, and diets are mental objects; they are not addictive by themselves. They will be addictive if they are met by a person vulnerable to their intrinsic properties. If we consider a “healthy mind” making a decision, two groups of specialists in the field have developed the same idea: emotion participates to reasoning-reflective activity and at least in the first stages of decision-making. Kahneman and colleagues (see Kahneman Reference Kahneman2003) have described a perception-intuition system – stimulus bound, fast, automatic, effortless, slow-learning and emotional; and a reasoning system – slow, controlled, effortful, not flexible, less emotional with different contents such as conceptual representations and temporal references. The Iowa group (Bechara et al. Reference Bechara, Damasio, Tranel and Anderson1998; Reference Bechara, Damasio and Damasio2000, Damasio et al. Reference Damasio, Grabowski, Bechara, Damasio, Ponto, Parvizi and Hichwa2000) refers also to a first emotional stage of the decision, a sort of impulsive system, followed by a reflective one; a peripheral emotional reaction occurs before and orients decision-making. It is possible that these two groups refer to two different parts of the frontal cortex (dorsolateral versus ventromedial).

The subject who enters the impulsive-compulsive spiral is, even before his or her first consumption, not immune to psychopathological disorders. He or she does not have a “healthy mind.” Different parts of the brain present some sort of dysfunction. It is well documented that the main source of vulnerability (not to be confounded with polydrug use) is the existence of psychopathologies and behavioral disorders (depression, anxiety disorders, impulsivity, stress, self-dysregulations, etc.), and that holds for more than 80%, if not all, of the subjects who will succumb (Goodman Reference Goodman2008; Le Moal & Koob Reference Koob and Le Moal2007; Shaffer et al. Reference Shaffer, LaPlante, LaBrie, Kidman, Donato and Stanton2004). These psychopathologies are related to neurobiological dysfunctions in both affective and cognitive systems. It has been proposed with robust arguments (Baumeister et al. Reference Baumeister, Heatherton and Tice1994; Baumeister & Vohs Reference Baumeister and Vohs2004) that these vulnerable individuals are unable to self-regulate their emotions, desires, motivations, and pleasures. Self-regulation failure is a complex construct; it may lead to impulse control disorders. Here again, the causal mechanisms are not clear and may involve cortical dysfunctions due to subcortical mesolimbic dysfunctions (Piazza & Le Moal Reference Piazza and Le Moal1996), in a negative feedback manner.

In conclusion, there is, I feel, something missing in Redish et al.'s scholarly review: the agency of reward-emotional systems (Koob & Le Moal Reference Koob and Le Moal2008) and of the stress systems (al'Absi Reference al'Absi2007).