Review
Impulsivity as a vulnerability marker for substance-use disorders: Review of findings from high-risk research, problem gamblers and genetic association studies

https://doi.org/10.1016/j.neubiorev.2007.11.003Get rights and content

Abstract

There is a longstanding association between substance-use disorders (SUDs) and the psychological construct of impulsivity. In the first section of this review, personality and neurocognitive data pertaining to impulsivity will be summarised in regular users of four classes of substance: stimulants, opiates, alcohol and 3,4-methylenedioxymethamphetamine (MDMA). Impulsivity in these groups may arise via two alternative mechanisms, which are not mutually exclusive. By one account, impulsivity may occur as a consequence of chronic exposure to substances causing harmful effects on the brain. By the alternative account, impulsivity pre-dates SUDs and is associated with the vulnerability to addiction. We will review the evidence that impulsivity is associated with addiction vulnerability by considering three lines of evidence: (i) studies of groups at high-risk for development of SUDs; (ii) studies of pathological gamblers, where the harmful consequences of the addiction on brain structure are minimised, and (iii) genetic association studies linking impulsivity to genetic risk factors for addiction. Within each of these three lines of enquiry, there is accumulating evidence that impulsivity is a pre-existing vulnerability marker for SUDs.

Introduction

The term impulsivity is used widely within psychology to refer to “behaviour that is performed with little or inadequate forethought” (Evenden, 1999b). Whilst some functional, adaptive aspects of impulsivity have been noted (Dickman, 1990), it is generally regarded to be a dysfunctional trait, associated with actions that may be criminal and/or violent, physically harmful to the self (such as suicide), or inappropriate given accepted social standards. The term has a long history in the study of individual differences, as a trait variable of human personality that is stable within an individual and varies normatively across the healthy population (Barratt, 1959; Patton et al., 1995). Within neuropsychology and cognitive neuroscience, impulsivity is often equated with the term ‘disinhibition’, referring to the idea that top-down control mechanisms ordinarily suppress automatic or reward-driven responses that are not appropriate to the current demands (Aron, 2007). These inhibitory control mechanisms may be disrupted following brain injury, or in forms of mental illness, resulting in a predisposition towards impulsive acts.

Defined in this way, impulsivity has clear relevance to substance-use disorders (SUD). Throughout the present article, we will use the term SUD to refer to the abuse of, or dependence upon, illicit substances including stimulants and opiates, as well as alcohol. The early stages of recreational drug taking may be mediated by personality characteristics that influence whether or not the individual will try a substance that is available, and how much of the substance they will consume. Once dependent, drug users may persist in drug-taking despite awareness that their habit is directly harmful to their health, their finances and their interpersonal relationships. Substance users may repeatedly attempt (but fail) to quit drug-taking or reduce drug intake. Each of these phenomena could plausibly be explained by deficient inhibitory control over a response that provides immediate reinforcement. Further understanding of neurobiological and psychological underpinnings of inhibitory control offers obvious promise for pharmacological treatments and behavioural treatment programs for SUD.

A wide array of measures exists for measuring impulsive behaviour in human subjects. Within the field of individual differences, well-validated self-report questionnaires exist to quantify the impulsive personality, including the Barratt Impulsivity Scale (BIS; Patton et al., 1995), the Impulsivity-Venturesomeness-Empathy Scale (IVE; Eysenck et al., 1985), or the UPPS Impulsive Behaviour Scale (Whiteside and Lynam, 2001, Whiteside and Lynam, 2003). Related constructs of Novelty Seeking and Sensation Seeking can be measured with the Tridimensional Personality Questionnaire (TPQ; Cloninger et al., 1991), the Temperament and Character Inventory (TCI; Cloninger et al., 1994) or the Sensation Seeking Scale (SSS) of the Zuckerman–Kuhlman Personality Questionnaire (Zuckerman et al., 1993). As reviewed in Section 2, there is considerable evidence that self-report ratings of impulsivity, novelty seeking and sensation seeking are increased in SUD populations relative to non-drug using controls. Self-report questionnaires assess general dispositional characteristics of the individual: how the individual would typically behave in a given situation, or to what extent the subject agrees or disagrees with particular statements. This introduces a number of caveats in the context of SUD populations. Primarily, most questionnaires do not explicitly distinguish between those current characteristics of the individual that have become instantiated since the onset of drug-taking from those pre-morbid characteristics that preceded the drug use. Whilst it is easy to assume that an inflated questionnaire score reflects an enduring characteristic of the individual, the methodology is unable to demonstrate this effect empirically. In addition, self-report questionnaires are susceptible to demand characteristics and biases in social desirability that may naturally differ between SUD volunteers and control subjects. Moreover, impulsivity may directly interfere with the completion of the questionnaires themselves, such that the impulsive subject may give less consideration to responses than the non-impulsive subject. Finally, introspective ratings assume that individuals have sufficient insight to rate their personality accurately.

These caveats with self-report measures have led to increased interest in direct measurement of inhibitory control processes using laboratory tasks (Dougherty et al., 2003, Dougherty et al., 2005; Evenden, 1999b; Moeller et al., 2001a; Reynolds, 2006). Cognitive and behavioural models of impulsivity have enabled the development of objective tests that measure performance in terms of accuracy and reaction time data. For the purposes of the present review, we will focus on three broad classes of neurocognitive test used to measure impulsivity (see also Fig. 1): (i) measures of response inhibition based on the suppression of an automatic (prepotent) response, namely the Go–No Go test, the Stop Signal test, the Stroop test, and measures of commission errors on Continuous Performance Tests (CPTs) (Logan et al., 1997); (ii) measures of delay-discounting, which define impulsivity in terms of choice preference for a small reward available immediately (or after a short delay) over a larger reward available at some point in the future (Bickel and Marsch, 2001; Reynolds, 2006); and (iii) measures of cognitive impulsivity, a broad term that refers to impulsive behaviour in the arena of decision-making. One element of cognitive impulsivity is ‘reflection impulsivity’, which refers to the tendency to gather and evaluate information before making complex decisions (Kagan, 1966). Inadequate reflection at the pre-decisional stage will reduce the accuracy of the eventual decision (Evenden, 1999a). Reflection impulsivity, measured with the Matching Familiar Figures Test (MFFT) (Kagan, 1966) or the Information Sampling Test (Clark et al., 2006), may be related to psychometric constructs of ‘non-planning impulsivity’ (Patton et al., 1995) or ‘lack of premeditation’ (Whiteside and Lynam, 2001). Cognitive impulsivity may also contribute to abnormal decision-making on tasks where the subject may select between a conservative option and a more risky option that offers a ‘superficially seductive’ gain (Bechara, 2003; Knoch and Fehr, 2008). These measures include the Iowa Gambling Task (IGT) (Bechara et al., 1994), the Risky Gains procedure (Paulus et al., 2003), and the Cambridge Gamble Task (CGT) and Risky Gains Task (RGT) (Rogers et al., 1999a, Rogers et al., 1999b). Impulsivity can be indexed by selection of the highly rewarding option despite the clear potential for negative outcomes. Whilst performance deficits on these tasks need not necessarily indicate impulsivity (Busemeyer and Stout, 2002), there is substantial overlap between the research literatures on impulsivity and decision-making in SUD, and we believe it is important to consider these tasks under the broad term of ‘cognitive impulsivity’.

Section 2 of the review provides an intentionally brief overview of the research in SUD groups showing robust deficits on various neurocognitive tests of impulsivity and elevated self-report impulsivity on questionnaire measures. Whilst few researchers would deny this basic observation, when we consider the source of this impulsivity, the field becomes markedly polarised. One possibility is that the chronic neurobiological effects of drug self-administration cause a gradual attrition of behavioural self-control, plausibly mediated by structural changes in the prefrontal cortex (e.g. Bechara, 2003; Goldstein and Volkow, 2002; Porrino and Lyons, 2000). This attrition may occur via direct neurotoxicity (cell death) or tissue shrinkage, and structural brain imaging and post-mortem studies in SUD groups have established reductions in regional brain volumes, and grey- and white-matter densities associated with many substances of abuse (Chanraud et al., 2007; Cowan et al., 2003; Lyoo et al., 2006; Matochik et al., 2003; Thompson et al., 2004). Even in the absence of such macro-cellular changes, a range of micro-cellular alterations including persistent changes in gene expression, and effects on neurogenesis and synaptogenesis, may cause a gradual breakdown of inhibitory control. Animal studies have elegantly demonstrated that cognitive deficits on tests of inhibitory control can be induced by relatively short-term courses of drug administration (Jentsch et al., 2002; Ricaurte et al., 2000; Robinson and Kolb, 2004). Research in experimental animals is able to quantify baseline cognitive function prior to drug initiation, and then precisely regulate drug dosage, frequency of administration, and other critical behavioural parameters. Moreover, the issue of poly-substance abuse that plagues the human clinical literature on SUD groups is obviated in animal research.

By an alternative explanation, deficient inhibitory control may have been present prior to drug initiation. Indeed, deficient inhibitory control may represent a vulnerability marker for SUD, predisposing individuals towards early recreational experiences with drugs, or mediating the transition from recreational use to dependence. It is important to note that the impact of personality and/or neurocognitive variables may differ across various stages of addiction, from initiation, to regular use, to dependence and later on at relapse (Kreek et al., 2005). Critically, the vulnerability and attrition accounts are by no means mutually exclusive: substance users may have impulsive personalities premorbidly, and this impulsivity may be further exacerbated via chronic substance administration. Nonetheless, the characterisation of vulnerability markers for addiction is essential for detecting at-risk individuals, and in order to implement early detection and treatment intervention and thereby avert the devastating effects of long-term use. The vulnerability account of impulsivity in SUD is clearly related to the endophenotype concept. Endophenotypes are defined as intermediate variables that lie between the ‘fuzzy’ clinical phenomenology of a disorder and the genetic and neurobiological processes responsible for the manifestation of that disorder (Schumann, 2007). In SUD research, impulsivity variables represent a promising candidate endophenotype to bridge the gap between genetic risk loci and the complex clinical manifestations of SUD. By the criteria proposed by Gottesman and Gould (2003), an endophenotype should: (1) be present in the condition of interest (e.g. in case-control studies), (2) be observable regardless of the state of the illness (i.e. to persist in symptom remission), (3) have evidence of heritability, and (4) be present in individuals at risk of developing the disorder (such as unaffected first-degree relatives) at rates above the general population. One purpose of the article is to consider the current literature on impulsivity in SUD in terms of these criteria. In advance of our conclusions, we will refer to impulsivity with the more general term ‘marker’.

Neuroscientific models of addiction have increasingly recognised the vulnerability pathway. Based on substantial evidence that initiation of drug-taking typically occurs during adolescence, Chambers et al. (2003) proposed that adolescence constitutes a high-risk period for the development of SUD due to the relative maturity of subcortical systems responsible for reward processing and motivation, coupled with relative immaturity of prefrontal cortical systems responsible for inhibitory control over these responses. In a similar vein, the triadic model of motivated action (Ernst et al., 2006b) explains the risk-taking behaviour of adolescents as result of the greater maturation of the ventral striatum (the reward system), as compared to the amygdala (avoidance system) and the prefrontal cortex (the regulatory system). These theoretical models are supported by the findings of differential trajectories of maturation across different brain regions in the adolescent brain, revealed by longitudinal structural MRI scanning (Lenroot and Giedd, 2006; Toga et al., 2006). Subsequent work has confirmed these developmental trends using risk-taking measures combined with fMRI scanning, in groups of young children, adolescents and young adults (Bjork et al., 2004b; Ernst et al., 2006a; Eshel et al., 2007; Galvan et al., 2006; van Leijenhorst et al., 2006).

A further observation in support of the vulnerability pathway has emerged from PET imaging with dopamine receptor ligands such as raclopride, a dopamine D2 receptor antagonist. A series of studies by Volkow and colleagues have shown that addictions to a range of substances are reliably associated with reduced dopamine D2 receptor density in the striatum (Volkow et al., 1993, Volkow et al., 1996, Volkow et al., 2001; Wang et al., 1997). Within healthy, drug-naïve individuals, the D2 binding potential was correlated with subjective responses to the psychostimulant methylphenidate, such that individuals with lower D2 density reported pleasurable, hedonic effects of methylphenidate whereas individuals with higher D2 density experienced an anxiogenic response (Volkow et al., 1999). Thus, the healthy individuals who experienced a hedonic response to drug were more similar to the SUD populations in terms of their dopamine transmission. Positive subjective and physiological reactions to initial drug exposure seem likely to influence the risk of developing SUD subsequently, consistent with a number of more recent studies (Brunelle et al., 2004; Fergusson et al., 2003; Grant et al., 2005; Taylor et al., 1999). Whilst these studies demonstrate the importance of pre-morbid characteristics in the development of later addiction, the findings to date have mainly highlighted the role of reward processes, rather than the premorbid impulsivity and inhibitory control, as critical in the development of SUDs.

The issue of impulsivity as a vulnerability marker for substance abuse has been elegantly addressed in some recent animal studies. These studies have divided groups of rodents into high- and low-impulsive subgroups on the basis of behavioural performance, in terms of either discounting preferences (Perry et al., 2005) or premature responses on an attentional task (Dalley et al., 2007). These individual differences remain stable across repeated testing, and can be used to breed strains of more-impulsive animals. The high impulsive subgroup displayed lower levels of striatal dopamine D2 receptor binding (Dalley et al., 2007), mirroring the effect seen in human SUD patients in the studies by Volkow and colleagues. The high impulsive animals also showed more rapid acquisition of drug self-administration and consume more cocaine than rats classified as non-impulsive (Perry et al., 2005; Piazza et al., 1989; Poulos et al., 1995). Considering adolescence as a period of increased risk for drug use, Stansfield and Kirstein (2005) have demonstrated that high impulsive adolescent rats (compared to high impulsive adult animals) also display an enhanced dopamine response to cocaine administration in the nucleus accumbens.

These observations in experimental animals require confirmation in humans and in SUD populations in order for their translational potential to be realised. In 3 Models of vulnerability I: impulsivity in high-risk populations, 4 Models of vulnerability II: impulsivity in problem gambling, 5 Models of vulnerability III: genetic association studies of impulsivity with risk factors for addiction of this review article, we will review three approaches that have been taken to investigate predisposing markers for addiction in humans. First, it is possible to identify individuals at high-risk of developing SUDs, by virtue of substance dependency in a parent. Second, it has been suggested that pathological gambling (PG) may provide a model of drug-free addiction, sharing genetic vulnerability with SUDs but without the concomitant harmful effects on the brain. Third, it is possible that personality or neurocognitive markers of impulsivity may be associated with particular genetic variants that convey risk for addictive disorders. In the next section, we will briefly review the evidence for increased impulsivity in SUD groups, in reference to four specific groups of substances: opiates, psychostimulants, alcohol and 3,4-methylenedioxymethamphetamine (MDMA). With each substance, we will consider the preliminary indications that this impulsivity may predate substance abuse or arise as an effect of long-term exposure. In the subsequent sections, we will consider the three approaches to vulnerability described above, and highlight the methodological issues that should be considered in future work.

Section snippets

Self-report studies

Impulsive behaviour in SUD groups has received the most attention in relation to the psychostimulant drugs, amphetamine and cocaine. Elevated impulsivity scores on self-report measures (typically the BIS) have been demonstrated in cocaine-dependent outpatients (Coffey et al., 2003; Moeller et al., 2004) and young stimulant users (Leland and Paulus, 2005), even after considering the influence of antisocial personality disorder (Moeller et al., 2002). Elevated BIS scores were correlated

Models of vulnerability I: impulsivity in high-risk populations

This section will review a number of studies that have attempted to address the role of impulsivity in the pre-existing vulnerability to SUDs by virtue of examining populations at known high-risk for drug use. These high-risk populations comprise three basic groups: (i) adolescent populations who are considered vulnerable to drug use by virtue of their age; (ii) clinical groups with externalising behavioural disorders (ADHD and Conduct Disorder) who are known to display high rates of subsequent

Models of vulnerability II: impulsivity in problem gambling

Gambling is a widespread and socially acceptable form of entertainment that is known to become problematic or ‘compulsive’ in a minority (around 1–3%) of the population in the US and UK (Shaffer et al., 1999; Sproston et al., 2000). Prevalence estimates vary according to the threshold used in diagnosis: strict ‘pathological gambling’ (where symptoms resemble those of DSM dependence) was reported to have a prevalence of between 0.5% and 1.5% (Petry et al., 2005; Welte et al., 2002), whereas the

Models of vulnerability III: genetic association studies of impulsivity with risk factors for addiction

Family and twin designs indicate a genetic contribution to SUDs in the range of 30–60% (Kreek et al., 2005), and much of this variance is non-specifically associated with multiple drugs of abuse as well as behavioural addictions including PG (Comings et al., 2001; Kendler et al., 2003; Slutske et al., 2000; Tsuang et al., 1998). In the past 15 years, genetic association studies have begun to identify a number of specific gene variants that are implicated in the risk of developing SUDs. Single

Conclusion

The diverse lines of evidence reviewed in the preceding sections supports our overall thesis that impulsive behaviour exists in SUD populations prior to the onset of drug taking, and is associated with the vulnerability to drug use and dependence. The case-control studies reviewed in Section 2 convincingly demonstrate that increased impulsivity is a robust phenomenon across SUD groups dependent upon a range of different substances including stimulants (cocaine or amphetamines), opiates, alcohol

Acknowledgements

A.V. was supported by a Juan de la Cierva post-doctoral contract from the Spanish Ministry of Science and Technology. A.J.L. was supported by a Medical Research Council (UK) post-graduate studentship. L.C. was supported by a project grant from the Economic and Social Research Council and the Responsibility in Gambling Trust (RES-164-0010). This work was supported by a consortium award from the Medical Research Council (UK) and Wellcome Trust to the Behavioural and Clinical Neuroscience

References (351)

  • G.K. Butler et al.

    Impulsivity, risk taking and recreational ‘ecstasy’ (MDMA) use

    Drug and Alcohol Dependence

    (2004)
  • P. Cavedini et al.

    Frontal lobe dysfunction in pathological gambling patients

    Biological Psychiatry

    (2002)
  • H.D. Chilcoat et al.

    Pathways from ADHD to early drug use

    Journal of the American Academy of Child and Adolescent Psychiatry

    (1999)
  • D.B. Clark et al.

    Childhood risk categories for adolescent substance involvement: a general liability typology

    Drug and Alcohol Dependence

    (2005)
  • L. Clark et al.

    Reflection impulsivity in current and former substance users

    Biological Psychiatry

    (2006)
  • L. Clark et al.

    Association between response inhibition and working memory in adult ADHD: a link to right frontal cortex pathology?

    Biological Psychiatry

    (2007)
  • B.T. Conner et al.

    DRD2 genotypes and substance use in adolescent children of alcoholics

    Drug and Alcohol Dependence

    (2005)
  • R.L. Cowan et al.

    Reduced cortical gray matter density in human MDMA (Ecstasy) users: a voxel-based morphometry study

    Drug and Alcohol Dependence

    (2003)
  • J. Crean et al.

    Effect of tryptophan depletion on impulsive behavior in men with or without a family history of alcoholism

    Behavioural Brain Research

    (2002)
  • M.A. Dawes et al.

    Behavioral self-regulation: correlates and 2 year follow-ups for boys at risk for substance abuse

    Drug and Alcohol Dependence

    (1997)
  • R. de la Torre et al.

    Neurotoxicity of MDMA (ecstasy): the limitations of scaling from animals to humans

    Trends in Pharmacological Sciences

    (2004)
  • H. de Wit et al.

    Acute administration of d-amphetamine decreases impulsivity in healthy volunteers

    Neuropsychopharmacology

    (2002)
  • W.M. Dinn et al.

    Cigarette smoking in a student sample: neurocognitive and clinical correlates

    Addictive Behaviors

    (2004)
  • G. Dom et al.

    Differences in impulsivity and sensation seeking between early- and late-onset alcoholics

    Addictive Behaviors

    (2006)
  • D.M. Dougherty et al.

    Suicidal behaviors and drug abuse: impulsivity and its assessment

    Drug and Alcohol Dependence

    (2004)
  • N. Eshel et al.

    Neural substrates of choice selection in adults and adolescents: development of the ventrolateral prefrontal and anterior cingulate cortices

    Neuropsychologia

    (2007)
  • S.B.G. Eysenck et al.

    Age norms for impulsiveness, venturesomeness and empathy in adults

    Personality and Individual Differences

    (1985)
  • G. Fein et al.

    Smaller feedback ERN amplitudes during the BART are associated with a greater family history density of alcohol problems in treatment-naive alcoholics

    Drug and Alcohol Dependence

    (2008)
  • M.T. Fillmore et al.

    Impaired inhibitory control of behavior in chronic cocaine users

    Drug and Alcohol Dependence

    (2002)
  • C.C. Allcock et al.

    Pathological gamblers are neither impulsive nor sensation-seekers

    Australian and New Zealand Journal of Psychiatry

    (1988)
  • A.R. Aron

    The neural basis of inhibition in cognitive control

    Neuroscientist

    (2007)
  • M. Asberg et al.

    5-HIAA in the cerebrospinal fluid. A biochemical suicide predictor?

    Archives of General Psychiatry

    (1976)
  • J.S. Baer et al.

    Task persistence after alcohol consumption among children of alcoholics

    Alcoholism-Clinical and Experimental Research

    (1995)
  • F. Baker et al.

    Delay discounting in current and never-before cigarette smokers: similarities and differences across commodity, sign, and magnitude

    Journal of Abnormal Psychology

    (2003)
  • R.A. Barkley

    Behavioral inhibition, sustained attention, and executive functions: constructing a unifying theory of ADHD

    Psychological Bulletin

    (1997)
  • E.S. Barratt

    Anxiety and Impulsiveness Related to Psychomotor Efficiency

    Perceptual and Motor Skills

    (1959)
  • A. Bechara

    Risky business: emotion, decision-making, and addiction

    Journal of Gambling Studies

    (2003)
  • A. Bechara et al.

    Characterization of the decision-making deficit of patients with ventromedial prefrontal cortex lesions

    Brain

    (2000)
  • H. Begleiter et al.

    What is inherited in the predisposition toward alcoholism? A proposed model

    Alcoholism-Clinical and Experimental Research

    (1999)
  • J. Benjamin et al.

    Population and familial association between the D4 dopamine receptor gene and measures of Novelty Seeking

    Nature Genetics

    (1996)
  • U. Berggren et al.

    The taqI DRD2 A1 allele is associated with alcohol-dependence although its effect size is small

    Alcohol and Alcoholism

    (2006)
  • C. Bergh et al.

    Altered dopamine function in pathological gambling

    Psychological Medicine

    (1997)
  • S.M. Berman et al.

    Reduced visuospatial performance in children with the D2 dopamine receptor A1 allele

    Behavior Genetics

    (1995)
  • W.K. Bickel et al.

    Toward a behavioral economic understanding of drug dependence: delay discounting processes

    Addiction

    (2001)
  • W.K. Bickel et al.

    Impulsivity and cigarette smoking: delay discounting in current, never, and ex-smokers

    Psychopharmacology

    (1999)
  • J.M. Bjork et al.

    Serotonin 2a receptor T102C polymorphism and impaired impulse control

    American Journal of Medical Genetics-B: Neuropsychiatric Genetics

    (2002)
  • J.M. Bjork et al.

    Incentive-elicited brain activation in adolescents: similarities and differences from young adults

    Journal of Neuroscience

    (2004)
  • A. Blaszczynski et al.

    Impulsivity in pathological gambling: the antisocial impulsivist

    Addiction

    (1997)
  • T. Blekher et al.

    Saccadic eye movements are associated with a family history of alcoholism at baseline and after exposure to alcohol

    Alcoholism-Clinical and Experimental Research

    (2002)
  • K. Blum et al.

    Allelic association of human dopamine D2 receptor gene in alcoholism

    Journal of the American Medical Association

    (1990)
  • Cited by (1098)

    View all citing articles on Scopus
    View full text