Defending and defining compulsive behaviour in addiction

摘要

Heinz and colleagues provide interesting insights into the clinical presentation of drug-taking habits and the associated difficulties of breaking them. Although they do not question the existence of habits in addiction, they raise concerns about the psychological construct of habit and its role in the development of compulsivity in addiction. The authors doubt that compulsive behaviour in addiction arises from a predominance of habits over goal-directed behaviour, as suggested by the habit theory.1 Their arguments are based on descriptions of differences in clinical phenotypes of compulsivity in addicted patients and patients with obsessive-compulsive disorder (OCD), without mentioning the many commonalities. For example, they explain that avoidance behaviour in OCD patients is negatively reinforced through the relief of anxiety but do not say that hoarding behaviour in OCD patients is positively reinforced.2 Likewise, they emphasise that the use of alcohol is positively reinforced by its pleasurable effects but do not mention the fact that negative reinforcement underlies chronic opioid use.3 Similar commonalities are also evident in the brain, as reflected by an overlapping neuropathology underlying self-reported compulsivity in both OCD and addiction.4 Focussing solely on different manifestations of orbitofrontal dysfunction (i.e., cases of overactivity or underactivity) distracts from the fact that the same system is impaired in both disorders but expressed in different ways. I wonder whether the authors' questioning of the role of compulsivity in addiction derives from an understanding that equates the psychological concept of compulsivity (i.e., the maladaptive continuation/perseveration of behaviour) with clinical symptoms of compulsions. Compulsive symptoms can of course be expressed in many different ways, as exemplified by the authors' clinical case of an OCD patient with comorbid compulsive alcohol use and gambling behaviour. Whilst there are variations of compulsive symptoms across different disorders, the psychological concept underpinning these behavioural manifestations is the same, namely, a reflection of ongoing actions that have become inappropriate to the immediate context. Moreover, it is worth clarifying that the habit theory does not contradict their observations. Habits (including habitual drug use) do not necessarily develop into compulsions because most people are able to break their habits. If habits are, however, learned under the influence of drugs or stress, the formation of habits is facilitated. In people with impaired prefrontal inhibitory control (such as patients with OCD or addiction), habits run the risk of persisting even if they no longer produce the desirable effects or lead to adverse consequences. This only affects a minority of drug users, as just 15%–20% are thought to develop addiction.5 Importantly, those who make this transition to addiction do not turn into ‘autonomous robots’ but are prone to continue using drugs in response to drug-associated cues despite good intentions to stop using and full awareness of the severe adverse consequences, perpetuated by persistent drug use. Heinz and colleagues rightly point out that negative consequences in humans are not imminent (in contrast to preclinical studies), but immediacy is not necessary. Knowledge about the negative outcome is sufficient. Indeed, most addicted patients are undeterred by knowing the harm, perpetuated by continued drug use. Possibly, the authors' criticism of compulsivity derives from their understanding of the habit construct, which they feel needs to be described in ‘dimensional’ terms. Unfortunately, they do not explain what they mean by ‘dimensional habits’. It may be that they refer to computational models, which explain behaviour through mathematical algorithms, thus providing a continuous measure of habit strength rather than defining it by the absence of goal-directed actions. Yet many computational models st