INTEGRATING BEHAVIORAL ECONOMICS AND BEHAVIORAL GENETICS: DELAYEDREWARD DISCOUNTING AS AN ENDOPHENOTYPE FOR ADDICTIVE DISORDERS

JAMES MACKILLOP, PhD1,2

1DEPARTMENT OF PSYCHOLOGY, UNIVERSITY OF GEORGIA2DEPARTMENT OF BEHAVIORAL AND SOCIAL SCIENCES, BROWN UNIVERSITY

Delayed reward discounting is a behavioral economic index of impulsivity, referring to how much anindividual devalues a reward based on its delay in time. As a behavioral process that varies considerablyacross individuals, delay discounting has been studied extensively as a model for self-control, both in thegeneral population and in clinical samples. There is growing interest in genetic influences on discountingand, in particular, the prospect of discounting as an endophenotype for addictive disorders (i.e., a heritablemechanism partially responsible for conferring genetic risk). This review assembles and critiques theevidence supporting this hypothesis. Via numerous cross-sectional studies and a small number oflongitudinal studies, there is considerable evidence that impulsive discounting is associated with addictivebehavior and appears to play an etiological role. Moreover, there is increasing evidence from diversemethodologies that impulsive delay discounting is temporally stable, heritable, and that elevated levels arepresent in nonaffected family members. These findings suggest that impulsive discounting meets thecriteria for being considered an endophenotype. In addition, recent findings suggest that genetic varia-tion related to dopamine neurotransmission is significantly associated with variability in discountingpreferences. A significant caveat, however, is that the literature is modest in some domains and, in others,not all the findings have been supportive or consistent. In addition, important methodological con-siderations are necessary in future studies. Taken together, although not definitive, there is accumulatingsupport for the hypothesis of impulsive discounting as an endophenotype for addictive behavior and a needfor further systematic investigation.Key words: delay discounting, impulsivity, behavioral economics, genetics, addiction, substance

dependence, review

Situated at the intersection of psychology andeconomics, behavioral economics is a hybriddiscipline that examines the transactions peoplemake with the world around them. The ap-proach has made major contributions to anumber of fields, most obviously to its parentdisciplines, but also to psychiatry, public health,and cognitive neuroscience. One area that hasbeen particularly fertile has been the study ofdelayed reward discounting, a behavioral eco-

nomic index of impulsivity that reflects howmuch a reward loses values based on its distancein the future (i.e., how much a reward isdiscounted based on its delay in time). Emergingfrom the study of reinforcement learning(Ainslie, 1975; Rachlin & Green, 1972) anddevelopmental perspectives in social psychology(Mischel, 1958, 1961), this form of impulsivityhas become a robust model for understandingself-control, and the lack thereof, both in thestudy of general psychological functioning andas a putative factor in a number of psychiatricconditions. This is especially the case for ad-dictive behavior (for a review, see MacKillop,Amlung, Few, Ray, Sweet, & Munafò, 2011). Inaddition, delay discounting has emerged as a keyparadigm in the incipient field of neuroeco-nomics (e.g., Bickel, Pitcock, Yi, & Angtuaco,2009; Kable & Glimcher, 2007; McClure,Laibson, Loewenstein, & Cohen, 2004), whichfurther integrates behavioral economics andcognitive neuroscience.There is increasing evidence that delay

discounting may also be relevant to psychiatricgenetics and, in particular, the genetic basis foraddictive behavior. The goal of the current

The review was partially supported by NIH grants (K23AA016936, P30 DA027827, R01 DA032015) and is based onan invited lecture given at the Association for BehaviorAnalysis International (ABAI) conference, Behavioral Eco-nomics: From Demand Curves to Public Policy (Chicago, IL,March 2011). I have no conflicts of interest to declare withregard to this review.Many thanks to Dr. Tim Hackenberg for organizing the

conference, to Maria Malott and her staff for its coordina-tion, and to the ABAI Science Board for its sponsorship. I amalso grateful for Dr. Tim Hackenberg and Dr. Greg Maddenfor editorial feedback on earlier versions of this manuscript.Contact information: James MacKillop, PhD, Department

of Psychology, 100 Hooper St. University of Georgia, Athens,GA 30605 (e-mail: jmackill@uga.edu).doi: 10.1002/jeab.4

JOURNAL OF THE EXPERIMENTAL ANALYSIS OF BEHAVIOR 2012, 9999, 1–18 NUMBER 9999 (OCTOBER)

1

review is to bring together the diverse sources ofevidence supporting this hypothesis in order toclarify the current knowledge and identifyfuture directions. The first section will reviewthe nature of delay discounting as a phenotype(i.e., any observable characteristic of an organ-ism), both in general and in terms of itsassociation with addictive behavior. The secondsection will review the endophenotype conceptin psychiatric genetics, both at a conceptuallevel and in terms of the specific criteria that areused to evaluate whether a given phenotypeshould be designated as an endophenotype.The third section is the intersection of the firstand second, the empirical evidence that sug-gests delay discounting constitutes an endophe-notype for addictive disorders. Finally, as this is arelatively new area of interest, the fourth sectionhighlights a number of important methodologi-cal considerations for future studies.

Delayed Reward Discounting as a Phenotypeand its Association with Addictive Behavior

Considering delay discounting as a phenotypesimply refers to its phenomenology and mea-surement. As a behavioral characteristic, delaydiscounting refers to an individual’s profile ofintertemporal reward preferences, or theirpreferences with regard to the tension betweensmaller rewards available in the short-term,typically the present, versus larger rewardsavailable at a future time point. This profilecan be operationalized in a number of ways andfor an array of commodities. In children,preferences for smaller immediate rewards atthe cost of larger delayed rewards can beassessed using the famous ‘marshmallow test’(e.g., Mischel, Ebbesen, & Zeiss, 1972), in whichparticipants are offered one marshmallow (oranother candy) immediately or two when theexperimenter comes back at some later point. Inadolescents and adults, discounting is typicallyassessed using tasks comprising dichotomousitems that pit smaller immediate rewards againstlarger delayed rewards, both in the commoncurrency of money. These tasks use permuta-tions of rewards and delays that systematicallyvary the options and quantify the effects of delayon the value of the larger reward. By plotting thesubjective value of the delayed reward acrossvarious delay periods, a discounting curve thatreflects the magnitude of devaluation can begenerated; two examples are provided in

Figure 1. The more precipitously the rewardloses value based on delay, the more impulsivethe individual is considered. More importantly,the expressed preferences across the choicescan be aggregated into a quantitative profile ofhow rapidly a delayed reward loses value to theindividual, the discounting phenotype.

Importantly, however, individual discountingdecision-making profiles can be characterizedusing a number of different methods: modelingthe temporal discounting function via nonlinearregression, area-under-the-curve analysis, orsimply the proportion of impulsive choices(Green &Myerson, 2004; Mazur, 1987; Mitchell,Fields, D’Esposito, & Boettiger, 2005; Myerson,Green, & Warusawitharana, 2001). In terms ofmathematical models, the most widely usedapproach is a hyperbolic single parametermodel (Mazur, 1987), but hyperbola-like (hy-perboloid) two-parameter models can also beapplied (Laibson, 1997; Myerson & Green,1995) and an additive-utility model has recentlybeen proposed (Killeen, 2009). These modelsseek to quantitatively capture the typical form ofthe discounting curve, namely, that the value ofa delayed reward rapidly loses its value duringinitial delays and then decays at a moremoderate rate across subsequent delays. Forexample, in the hyperbolic discounting curvesin Figure 1, in comparing the slope during thefirst month of delay and the slope from 11 to12 months of delay, it is clear that althoughthe absolute delay durations are identical, thedevaluation applied in the short run is consid-erably steeper. Abbreviated choice tasks alsohave been developed to more efficiently char-acterize discounting (e.g., Kirby, Petry, &Bickel, 1999; Madden, Petry, & Johnson, 2009)and are based on the hyperbolic model but canbe also be analyzed using impulsive choice ratios(e.g., Murphy &MacKillop, 2012). Indeed, evenvery small numbers of items or a single item canbe used to infer discounting rates (Anokhin,Golosheykin, Grant, & Heath, 2011; Bradford,2010; Reimers, Maylor, Stewart, & Chater,2009; Wulfert, Block, Santa Ana, Rodriguez, &Colsman, 2002). At this point, while there is nosingle methodology that is uniformly acceptedas themost valid index of discounting, it appearsthat measures with greater precision are moresensitive to addiction-related group differences(MacKillop et al., 2011).

In the context of addictive behavior, delaydiscounting is highly relevant for a number of

2 JAMES MACKILLOP

reasons. At a clinical level, impulsive discountingis a prototypic pattern in substance use disordersand pathological gambling. In other words,alcoholism, heroin addiction, and other formsof addictive behavior all commonly reflectbehavioral patterns of dramatically and persis-tently overvaluing immediate drug rewards atthe very high cost of future outcomes for theindividual. The second connection betweendelay discounting and addictive behavior is ata mathematical level, where impulsive discount-ing may explain self-control failures in individu-als with addictive disorders. In behavioraleconomics, self-control failures are referred toas preference reversals (i.e., a person changingtheir mind about their preference for oneoption over another), which are very commonin the context of addictive disorders. Forexample, large proportions of individuals withsubstance use disorders report being motivatedto change their behavior and do indeed seektreatment (e.g., Etter, Perneger, & Ronchi,1997; Hogue, Dauber, & Morgenstern, 2010),but later voluntarily drop out of treatment toresume drug use or relapse despite successfullycompleting treatment (e.g., McKay, Franklin,Patapis, & Lynch, 2006). This may be explained

by delay discounting because, as noted above,hyperbolic/hyperboloid discounting meansthat changes in value are not constant overtime (Ainslie, 2001; Frederick, Loewenstein, &O’Donoghue, 2003). More specifically, rewardsappear to both disproportionately lose and gainvalue based on temporal proximity (either as thetime to receipt approaches or as initial delaysare implemented), which quantitatively predictsa change in overall preference from a largerdelayed reward to a smaller immediate reward asit becomes temporally near at hand (i.e., apreference reversal). Moreover, the more steep-ly hyperbolic the individual’s discountingcurve, the more likely this is to happen. Thus,hyperbolic discounting provides amathematicalmodel for preference reversals from outcomesthat have larger long-term benefits to those thatare smaller, but immediately available (for morecomprehensive reviews, see Ainslie, 2001 andOdum, 2011a, b).There is also a very robust empirical connec-

tion between discounting and addictive behav-ior. Many cross-sectional categorical studieshave been completed, comparing delay dis-counting in a criterion group of individualsexhibiting addictive behavior to a control group.

Fig. 1. Prototypic hyperbolic delayed reward discounting curves reflecting the discounted subjective value of $100delayed from 1 day to 1 year (curves reflect the immediate discounted value). At 100 days, $100 has lost�50% of its nominalvalue for the low impulsivity profile and �90% of its nominal value for the high impulsivity profile.

BEHAVIORAL ECONOMICS AND BEHAVIORAL GENETICS 3

More impulsive discounting has been found incriterion groups based on levels of alcoholmisuse and dependence (e.g., Petry, 2001;Vuchinich & Simpson, 1998), nicotine depen-dence (e.g., Bickel, Odum, & Madden, 1999),stimulant dependence (e.g., Coffey, Gudleski,Saladin, & Brady, 2003), opiate dependence(e.g., Madden, Petry, Badger, & Bickel, 1997),and pathological gambling (e.g., MacKillop,Anderson, Castelda, Mattson, & Donovick,2006). Moreover, a recent meta-analysis of 46cross-sectional studies found consistent andsubstantial evidence of differences betweengroups in these categorical studies (MacKillopet al., 2011). Specifically, as illustrated inFigure 2, themeta-analysis revealed a statisticallysignificant medium effect size difference be-tween groups across studies, and also revealedstatistically significant and medium effect sizedifferences within addictive behavior categories,suggesting this was common across addictivedrugs (albeit with the exception of marijuanause disorders). In addition, the meta-analysisrevealed a significantly larger effect size instudies using clinical samples, suggesting impul-sive discounting is more robustly associated withclinically-relevant addictive behavior, not simplyparticipation in these behaviors. Beyond cate-

gorical studies, continuous analyses have alsoreported significant positive associations be-tween level of impulsive discounting and thelevel of addictive behavior (e.g., Alessi & Petry,2003; Bobova, Finn, Rickert, & Lucas, 2009;MacKillop et al., 2010; Sweitzer, Donny, Dierker,Flory, & Manuck, 2008). Thus, there is veryrobust evidence of the association betweenimpulsive delay discounting and addictivebehavior in cross-sectional studies.

Of course, cross-sectional associations cannotalone demonstrate causality, but a number oflines of evidence suggest that impulsive dis-counting at least partially predates addictivebehavior and plays an etiological role. Forexample, in a large sample of adolescents,Audrain-McGovern et al. (2009) examineddiscounting and smoking behavior over a 5-year period and found that discounting pre-dicted smoking initiation, not the other wayaround. These findings are similar to a previousfinding that discounting assessed in pre-schoolers was associated with adult drug use20 years later (Ayduk et al., 2000). In addition,two retrospective studies have found highdiscounting in adolescence to be associatedwith an earlier onset of symptoms of alcohol usedisorders (Dom, D’Haene, Hulstijn, & Sabbe,

Fig. 2. Aggregated differences between criterion groups composed of individuals exhibiting various forms of addictivebehavior and control participants in units of effect size (Cohen’s d); data from MacKillop et al. (2011).

4 JAMES MACKILLOP

2006; Kollins, 2003). Finally, studies usinganimal models of discounting have found thatimpulsive discounting predicts acquisition andescalation of drug self-administration in drug-naïve animals (Anker, Perry, Gliddon, &Carroll,2009; Marusich & Bardo, 2009; Perry, Larson,German, Madden, & Carroll, 2005; Perry,Nelson, Anderson, Morgan, & Carroll, 2007),convincingly suggesting that impulsive delaydiscounting serves as a predisposition to addic-tive disorders.In addition to contributing to the develop-

ment of addictive disorders, several prospectivestudies have found that delay discountingpredicts negative outcomes in smoking cessa-tion treatment (Krishnan-Sarin et al., 2007;MacKillop & Kahler, 2009; Sheffer et al., 2012;Yoon et al., 2007). Similarly, impulsive discount-ing has been found to be significantly associatedwith poorer opiate treatment response (Passetti,Clark, Davis, Mehta, White, Checinski, King, &Abou-Saleh, 2011). Most recently, Stanger et al.(2012) found that discounting predicted treat-ment outcome in adolescents with marijuanause disorders. Thus, these studies suggest that,in addition to predating addictive behavior,impulsive discounting also plays an importantrole in maintaining addictive disorders, servingas a risk factor for treatment failure.Taken together, delay discounting as a pheno-

type reflects a decision-making orientationtoward smaller immediate rewards comparedto larger delayed rewards. This orientation canbe thought of as an aspect of personality (e.g.,Odum, 2011b) or simply as a decision-makingprofile. In either case, however, this orientationtoward intertemporal rewards can be objectivelyand precisely characterized using decision-making tasks. Across methodologies and inboth cross-sectional and longitudinal designs,there is converging and convincing evidencethat impulsive discounting contributes to thedevelopment and maintenance of addictivedisorders.

The Endophenotype Concept inPsychiatric Genetics

The substantial role of genetics in addictivedisorders is now well established (for a review,see Goldman, Oroszi, & Ducci, 2005). Addi-tive genetic factors are estimated to contribute50-60% (Goldman et al., 2005), meaning thatapproximately half of the statistical probability

of developing an addictive disorder is attribut-able to genetic influences. Moreover, the recentrevolutionary advances in human moleculargenetics and high-throughput bioinformaticshave offered an unprecedented opportunity topotentially identify the specific genetic poly-morphisms responsible for conferring geneticrisk. Paradoxically, however, from early candi-date gene association studies examining onevariant in a single gene to more recent genome-wide association studies using 1M or morepolymorphisms across genome1, the resultshave been disappointing because the geneticassociations observed in early studies have oftennot been replicated and, where present, theeffect sizes have been very small (Goldmanet al., 2005; Treutlein & Rietschel, 2011).In other words, in spite of the establishedhigh levels of heritability, the specific geneticvariants that confer risk for (or protectionagainst) addictive disorders remain largelyobscure. This disconnect between moderateheritability and very small individual geneticassociations applies to most psychiatric dis-orders (Turkheimer, 2011).One reason why this may be the case is

increasing consensus that diagnosis of a clinicalsyndrome constitutes a low quality phenotype.This is because a diagnostic phenotype by itsnature is inherently heterogeneous. Two indi-viduals could receive an identical diagnosis withmany different permutations of symptoms thatentirely do not overlap. Severity introducesfurther variability, with some individuals barelymeeting criteria and others exhibiting the mostsevere manifestation of the disorder. Moreover,subtypes within disorders have been empiricallyidentified (e.g., Hesselbrock & Hesselbrock,2006; Johnson, van den Bree, & Pickens, 1996;Moss, Chen, & Yi, 2007), reflecting clusters of

1Candidate gene studies involve one or more specificgenes that have been identified as promising targets.Typically, different frequencies of one or more variantswithin the candidate gene are examined between groups ofindividuals who do and do not have the disorder of interest(cases vs. controls), with the disproportionate frequency ofan allele in either the affected group or control groupreflecting risk or protection. These are also referred to asassociation studies because they examine the association ofcertain genotypes with the condition. Genome-wide associa-tion studies are similar to candidate gene association studies,but are atheoretical high-bandwidth investigations thatexamine frequency variation in many thousands of allelesacross the genome, providing a more comprehensiveperspective.

BEHAVIORAL ECONOMICS AND BEHAVIORAL GENETICS 5

characteristics that meaningfully aggregate andmay be diversely genetically influenced. Fur-thermore, the course of substance use disordersis highly variable; large portions of individualsrecover with or without formal treatment, whileothers exhibit stable patterns over long periodsof time and still others exhibit a progressivecourse (e.g., Guttmannova et al., 2011). Thismeans that individuals may be classed togetheras reflecting the phenotype of interest at onepoint but actually reflect very different profileswhen considered longitudinally. Taken togeth-er, the common problem is that dichotomousclinical diagnosis has manifold instances inwhich ‘apples and oranges’ may be combinedand this creates a very heterogeneous andimprecise phenotype.

Toward clarifying the genetic basis forpsychiatric disorders, an endophenotype ap-proach focuses on characterizing narrowermechanistic processes that are more proxi-mally related to genetic variation and in turnconfer risk for a given clinical condition(Gottesman & Gould, 2003; Gottesman &Shields, 1973). Endophenotypes are definedas genetically-influenced characteristics thatputatively connect the dots from specificsources of genetic variation to increases or de-creases in the probability of having a disorder,as illustrated in Figure 3. Genetic variation istheorized to directly result in transcriptionaldifferences, which then result in systems-leveldifferences, which, in turn, result in the ob-served endophenotypic variation, and thereby

Fig. 3. The endophenotype approach in psychiatric genetics. Panel A presents the conceptual premise that for heritablepolygenic conditions such as addictive disorders, alternative mechanistic phenotypes are necessary to map the relationshipfrom genetic factors to risk for the condition. These phenotypes are defined as stable and heritable features of the individualthat predate the onset of the condition, are independent of the condition itself, but are probabilistically related to thedevelopment of the condition. Panel B provides the elaborated endophenotype model that includes the additional relevantlevels of analysis, including the direct effects of genetic variation on protein transcription, those effects on larger biologicalsystems, the proximal endophenotype, and, ultimately, disorder liability. For both panels, the number of spaces in the arrowsreflects the putative strength of the relationship, with fewer spaces reflecting stronger relationship and vice versa.

6 JAMES MACKILLOP

proximally influences risk for the disorder.More simply, endophenotypes are the inter-vening psychobiological mediators betweenspecific forms of genetic variation and dis-order liability. Thus, a key element for theendophenotype approach is the importanceof refining the complex pathways by which riskis communicated from the genome to anindividual’s clinical status. By being moreclosely associated with genetic variation, en-dophenotypes are theorized to be morereliable and reflect larger magnitude relation-ships, making them easier to detect andreplicate in smaller samples. Furthermore,by clarifying the pathophysiological andmotivational processes that contribute to theclinical condition, endophenotypes have thepotential to provide the foundation for im-plementing genetically-informed treatment orprevention.These manifold benefits, however, are

predicated on the successful identification ofendophenotypes and a number of criteriahave been proposed to clarify whether acharacteristic is appropriately thought of asan endophenotype. Since the initial introduc-tion of the concept almost forty years ago(Gottesman & Shields, 1973), the term wasintended to refer to a characteristic or processthat was not readily observable, but wassignificantly associated with the disorder ofinterest, could be objectively and reliablyassessed over time, and was heritable. Inaddition, further criteria for an endopheno-type are evidence that it is state-independent(i.e., present even when the disorder is not)and present in nonaffected family membersat a higher prevalence than in the popula-tion in general, reflecting transmission pat-terns of familial association and cosegregation(Flint & Munafò, 2007; Gottesman & Gould,2003). Additional criteria that have beenproposed are that an endophenotype becontinuous in nature, part of a biologicallyplausible etiological process, and closer tothe genetic causative factors than to thedisorder (Flint & Munafò, 2007). Importantly,it has been noted that an endophenotypemay be developmentally-defined (e.g., evidentonly after a certain age or only pertaining toa specific developmental period) and mayonly be present in response to a laboratorychallenge (Hasler, Drevets, Gould, Gottes-man, & Manji, 2006).

Evidence Supporting Delayed RewardDiscounting as an Endophenotype

Although the notion of delay discounting asan addiction endophenotype is relatively recent,there is nonetheless accumulating support froma number of domains. As discussed previously,there is converging evidence that discounting isrobustly associated with addictive behavior. Inaddition, there is evidence that it meets the coreendophenotype criteria. To start, a number ofstudies have found discounting preferences tobe relatively stable over time. Robust reliabilityhas been demonstrated in adolescents andadults over numerous time intervals, includingone week (Baker, Johnson, & Bickel, 2003;Clare, Helps, & Sonuga-Barke, 2010; Simpson &Vuchinich, 2000), six weeks (Beck & Triplett,2009), two months (Takahashi, Furukawa,Miyakawa, Maesato, & Higuchi, 2007), threemonths (Takahashi et al.; Weatherly, Derenne,& Terrell, 2011), one year (Kirby, 2009), andeven multiple years (Anokhin et al., 2011;Audrain-McGovern et al., 2009).More importantly, there is also evidence from

studies with both animals and human twins thatdelay discounting is heritable. In the formercase, a number of studies have investigateddifferences in discounting between inbredrodent strains. As these strains are isogeneticwithin each strain (Beck et al., 2000) and receiveidentical rearing environments, systematic dif-ferences in discounting across strains can beattributed to genetic differences. The first studyrevealing strain differences did so in the contextof investigating clomipramine as a pharmaco-therapy for impulsivity in Lewis and Fischer 344rats (Anderson &Woolverton, 2005), finding noeffect of themedication, but significantly higherdelay discounting in the Lewis rats. This findinghas subsequently been replicated twice using asteady-state paradigm to minimize within-ses-sion carryover (Madden, Smith, Brewer, Pink-ston, & Johnson, 2008; Stein, Pinkston, Brewer,Francisco, & Madden, 2012). Using six strains,Wilhelm andMitchell (2009) found significantlygreater discounting in Fischer rats compared toCopenhagen and Noble rats, but not Lewis rats,although a recent study suggests that this failureto replicate may be a function of procedurerather than strain (Stein et al., 2012). Studiesusing mice have revealed similar results. In aninitial investigation of four strains, significantbetween-strain differences were present as was

BEHAVIORAL ECONOMICS AND BEHAVIORAL GENETICS 7

covariation between discounting behavior andlocomotor activity (Isles, Humby, Walters, &Wilkinson, 2004). More recently, no differencesin discounting were found between dopamineD4 receptor knock-out mice and wild-typecontrols (Helms, Gubner, Wilhelm, Mitchell,& Grandy, 2008), suggesting that the dopamineD4 receptor is not playing a primary role.

Of particular interest, a number of studieshave further linked strain-based differences indiscounting with propensity for addictive behav-ior. For example, rats with a high preference forsaccharin have been shown to have a propensityfor cocaine consumption using a number ofindices (e.g., Carroll, Anderson, & Morgan,2007; Carroll, Morgan, Lynch, Campbell, &Dess, 2002; Perry, Morgan, Anker, Dess, &Carroll, 2006), and also exhibit significantlymore impulsive discounting than low-preferringrats (Perry et al., 2007). Similarly, rats and micefrom alcohol-preferring strains exhibit signifi-cantly greater delay discounting compared tolow alcohol preference strains (Oberlin &Grahame, 2009; Wilhelm & Mitchell, 2008).However, this is not the case for all alcohol-preferring strains, as mice in another alcohol-preferring strain, STDRHI2, have not beenfound to differ compared to STDRL02 mice, alow alcohol preference strain (Wilhelm, Reeves,Phillips, & Mitchell, 2007). In addition, surpris-ingly, greater impulsivity has been found inDBA/2J mice compared to C57BL/6J mice(Helms, Reeves, & Mitchell, 2006), which is theopposite of what would be predicted by strain-based drug preferences. Although not conclu-sive at this point, these studies nonethelessreveal the interplay of genetic factors, impulsivediscounting, and drug motivation without thetypical confound of drug exposure in humanstudies.

These studies using animal models revealboth aggregate genetic influences on impulsivechoice and, in several cases, the connectionbetween these influences and drug motivation.However, there are limits to generalizability,such as whether the same genetic variants areassociated with greater discounting acrossstrains or whether variation in discounting inhumans is similarly genetically influenced and,if so, attributable to homologous genes. Towardaddressing the question of heritability inhumans, the gold standard is the twin method-ology, in which a given phenotype is examinedin both monozygotic and dizygotic twins and

statistical modeling is used to infer the amountof variation attributable to genetic factorsbased on known differences in the amount ofshared genetic variation (i.e., effectively 100% inmonozygotic twins versus 50% on average indizygotic twins). More specifically, this approachpermits parsing the relative overlap in thecharacteristic or condition to additive geneticfactors (the sum of individual genetic effects);nonadditive genetic factors, such as gene x geneinteractions; shared environmental experiences(events or experience that pertain to bothtwins), and nonshared environmental factors(events or experiences that are specific to anindividual). These designs generate heritabilityestimates of the proportion of variation attrib-utable to these different domains, such as theestimate of 50-60% heritability of addictivedisorders (Goldman et al., 2005).

While many twin studies have examineddisorder status as a phenotype, only one twinstudy has assessed discounting. Anokhin et al.(2011) examined the heritability of delaydiscounting in early adolescent twins who wereassessed using a single item discounting mea-sure at ages 12 and 14 and found evidencefor additive genetic influences at both ages(age 12, a2 ¼ .30; age 14, a2 ¼ .51). Interestingly,however, the best fitting model indicatedsignificant genetic influences at age 12 predict-ing discounting at both age 12 and 14, butalso with significant nonshared environmentalinfluences at both ages 12 and 14. In otherwords, these results suggest that discounting isindeed a heritable phenotype but both geneticand environmental influences appear to beoperative.

Another common methodology for investi-gating endophenotypes is examining whetherthe phenotype of interest aggregates in individ-uals who do not themselves have the disorder,but have a family history (FH) of the condition.This permits assessing the presence of theendophenotype in unaffected but nonethelessat-risk relatives. For heritable disorders, likeaddictive disorders, the presence of a FHincreases risk for the condition and, if FH statusis associated with a specific characteristic, itimplicates the characteristic as a potentialmechanism. To date, only a small number ofstudies have been conducted using this ap-proach with discounting and the findings havebeen mixed. Petry, Kirby, and Kranzler (2002)examined paternal alcohol dependence family

8 JAMES MACKILLOP

history status in 122 adults and found signifi-cantly more impulsive discounting in FHþwomen, but not men. Crean, Richards, and deWit (2002) also examined paternal alcoholismin the context of a tryptophan depletionlaboratory challenge, but found no differencesbetween the two groups and no effect of themanipulation on discounting. In a relativelysmall sample (N ¼ 33), Herting, Schwartz,Mitchell, and Nagel (2010) detected greaterdiscounting in FHþ adolescents compared toFH- at the level of a statistical trend, although asignificant difference was present in terms ofchoice reaction times.The findings from these studies modestly

implicate discounting as a mechanism confer-ring the risk from family history, but significantmethodological issues make this conclusion farfrom definitive. For example, two of the threepreceding studies had very small samples thatwere underpowered for effects below relativelylarge magnitude group differences. Moreover,these studies focused specifically on familyhistory of alcohol misuse, but did not necessarilyrule out the presence of other drug misuse orother forms of addictive behavior in the FH-samples. This is a problem because, as notedabove, impulsive discounting is associated withan array of addictive disorders (MacKillopet al., 2011) and the ostensible ‘control group’may have been contaminated with individualswho also had a family history of addictivebehavior, substantially undermining the experi-mental comparison. Most recently, in the largeststudy to date (N ¼ 298), some of these consid-erations were addressed via careful characteri-zation of family history status for alcohol andother drugs, and also quantification of thedensity of the family history (Acheson, Vincent,Sorocco, & Lovallo, 2011). In this study,FHþ status was dichotomously associated withmore impulsive discounting and significantlypositively related to density of family history ofaddictive disorders. Thus, although the data aresomewhat mixed with regard to family historystatus and discounting, it appears that whensample sizes are sufficiently large and the statusof both FHþ and FH- are thoroughly character-ized, there is support for the aggregation ofimpulsive delay discounting with family historystatus.The preceding studies provide oblique infer-

ences about genetic influences on delay dis-counting, but a small number of molecular

genetic association studies have directly investi-gated the role of specific polymorphisms ondiscounting2. For example, in a moderately-sized nonclinical sample of young adults(N ¼ 166), discounting was examined in rela-tion to two genetic variants, the DRD2/ANKK13-Taq IA single nucleotide polymorphism (SNP;rs1800497) and the variable number of tandemrepeats polymorphism in exon 3 of the dopa-mine D4 receptor gene (DRD4 VNTR) (Eisen-berg et al., 2007). In this case, possession of atleast one DRD2/ANKK1 A1 allele was associatedwith significantly greater discounting, andpossession of both the A1 allele and the longform of DRD4 VNTR was interactively associatedwith substantially more impulsive discounting.In other words, there was no main effect ofDRD4 VNTR genotype, but the A1 allele carrierswho also had at least one long version of DRD4VNTR exhibited disproportionately high levelsof impulsive discounting compared to the othergenotype combinations. In a separate study ofa locus related to the dopamine D2 gene,nonclinical young adults who were C allelecarriers of the DRD2 C957T (rs6277) SNPexhibited more rapid responding duringdiscounting, although not more impulsive dis-counting (White, Lawford, Morris, & Young,2009).Two studies have examined the relationship

between COMT val158met SNP (rs4680) anddiscounting. In the first, a study of alcoholicsand healthy controls (N ¼ 19), individuals who

2Several definitionsmay be useful for this section. Geneticvariation across individuals can come in diverse forms and apolymorphism refers to any instance in which the sameportion of a gene exists in multiple forms. These differentversions are termed alleles. DNA is composed of nucleotidebase pairs and a single nucleotide polymorphism (SNP)refers to genetic variation reflecting a single difference at agiven base. Alternatively, a single nucleotide may be deletedor inserted into the sequence. In contrast, variable numberof tandem repeats (VNTR) polymorphisms refer to DNAsequences that include the repetition of the same series ofnucleotides multiple times, with individuals varying on thenumber of times the series repeats. Larger scale structuraldifferences are also present between individuals, such ascopy number variations (CNVs) which are large scaleportions (thousands tomillions of base-pairs) of the genomethat are repeated or deleted. Thus, across the �3.2Bnucleotide base-pairs that comprise the genome, humangenetic variation can be present from the level of a singlebase-pair all the way up to segments of DNA thousands andeven millions of base-pairs in length.

3This polymorphismwas initially erroneously identified asbeing within the DRD2 gene but is in fact within the nearbyANKK1 gene (Neville, Johnstone, & Walton, 2004).

BEHAVIORAL ECONOMICS AND BEHAVIORAL GENETICS 9

were homozygous for the val variant, exhibitedsignificantly more impulsive discounting(Boettiger et al., 2007). In the second study,individuals with attention deficit/hyperactivitydisorder (ADHD) were compared to controlparticipants (N ¼ 68) and discounting differedby COMT val158met genotype, but met-methomozygotes exhibited significantly greaterdiscounting independent of ADHD diagnosis(Paloyelis, Asherson, Mehta, Faraone, & Kuntsi,2010), a result that differs from Boettiger et al.(2007), where val-val homozygotes exhibited thegreatest discounting. Paloyelis et al. also exam-ined discounting according to genetic variationin the DAT1 (the dopamine transporter gene)10/6 haplotype4 andDRD4 VNTR. Interestingly,diagnosis interacted with genotype, such thatcontrol participants with two copies of the DAT110/6 haplotype exhibited significantly moreimpulsive discounting than those with fewerthan two copies and exhibited a profile ofpreferences that was equivalent to the ADHDparticipants. This is important because thepresence of the 10/6 haplotype is robustlyassociated with ADHD (Asherson et al., 2007)and it suggests that 10/6 haplotype may beresponsible for differences in discounting, butthat this effect may be obscured in individualswith the clinical condition who exhibit agenerally high level of discounting. In otherwords, because clinical populations are com-posed of individuals who have a convergence ofrisk factors and by their nature exhibit nonnor-mative levels of performance, it may be thatnonclinical populations will be especially impor-tant for successfully detecting genetic effects.Notably, no DRD4 VNTR genotype main effectwas present in the Paloyelis et al. (2010) study,converging with Eisenberg et al. (2007) and alsoa recent study by Garcia et al. (2010). It also isworth noting that these associations werepresent only for a decision-making discountingtask, not an experiential discounting task usingactual delays, a point discussed in greater detailbelow.

Clearly only a small number of moleculargenetic studies have been conducted on hu-mans and those that have include small tomoderate sample sizes, often comprising bothclinical and nonclinical populations. As such, it

is challenging to make definitive conclusions,but one theme that is emerging is the implica-tion of genetic variation that results in changesin dopamine neurotransmission and, in partic-ular, reduced levels of dopamine neurotrans-mission. For example, in the case of theassociation between DRD2/ANKK1 Taq IA andinteraction of DRD2/ANKK1 and DRD4 VNTR(Eisenberg et al., 2007), the alleles associatedwith greater impulsivity have both been associ-ated with dopaminergic hypofunction (Jonssonet al., 1996; Jonsson et al., 1999; McGeary, 2009).Similarly, in the association between impulsivediscounting and COMT val-val genotype, the valallele is associated with greater enzymatic meta-bolism of dopamine (i.e., more rapid degrada-tion of dopamine, terminating the activity faster;Savitz, Solms, & Ramesar, 2006), again implicat-ing dopamine hypofunction. Finally, in the caseof the DRD2 C957T polymorphism, the C alleleis associated with reduced striatal D2 binding(Hirvonen et al., 2004), further supporting arole for reduced dopamine activity as a neuro-biological mechanism for genetic influences ondelay discounting.

This hypothesis is also consistent with studiesusing animal models and fMRI that haveimplicated the ventral striatum in discountingdecision-making, a key region in the cortico-mesolimbic dopamine system (for reviews, seeCardinal, 2006; Carter, Meyer, &Huettel, 2010).For example, lesioning the ventral striatum inan animal model of discounting has beenshown to induce significantly more impulsivedelay and probability discounting (Cardinal &Howes, 2005; Cardinal, Pennicott, Sugathapala,Robbins, & Everitt, 2001). Similarly, comparedto Fischer rats, Lewis rats have fewer dopamineD2 and D3 receptors in the striatum and alsofewer dopamine transporters in these regions(Flores, Wood, Barbeau, Quirion, & Srivastava,1998), paralleling the behavioral findings.Moreover, a recent meta-analysis of brainactivity during fMRI discounting paradigmsrevealed the ventral striatum was consistentlyrecruited (Carter et al., 2010). However, it isimportant to note that dopaminergic hypofunc-tion as an underlying mechanism is not con-sistent with all the published findings (e.g.,Koffarnus, Newman, Grundt, Rice, & Woods,2011; Pine, Shiner, Seymour, & Dolan, 2010).Moreover, other neurotransmitter systemshave also been implicated in delay discount-ing (Bevilacqua et al., 2010; Mobini, Chiang,

4A haplotype refers to a combination of alleles that arephysically close to one another and tend to be inheritedtogether.

10 JAMES MACKILLOP

Al-Ruwaitea et al., 2000; Mobini, Chiang, Ho,Bradshaw, & Szabadi, 2000), as have impor-tant interactions between neurotransmittersystems (Winstanley, Dalley, Theobald, &Robbins, 2003; Winstanley, Theobald, Dalley,& Robbins, 2005). Thus, although diverseneurotransmitters are likely involved, an emerg-ing theme is that genetic variation that contrib-utes to reduced corticomesolimbic dopaminetone is associated with more impulsive discount-ing. This hypothesized model is presented inFigure 4.Taken together, acknowledging the relatively

small number of studies and ambiguities to date,the assembled findings nonetheless ‘tick-the-boxes’ for delay discounting as an endopheno-type. There is evidence that delay discountingdecision-making is a stable and heritablephenotype; that elevations in impulsive dis-counting are present in otherwise unaffectedindividuals who are FHþ for addictive disorders;and that specific forms of genetic variationare associated with significant differences indiscounting preferences. These findings thusprovide the critical second link, that geneticfactors are partially responsible for discountingdecision-making.Moreover, in the case of strain-based differences in discounting, there is alsoevidence that differences aggregate with strainsthat also engage in more addictive behavior.This connection has not been made directly inhumans, however, and while the findings to datesuggest this is a highly promising prospect, it isby no means established or a foregone conclu-sion. Future progress will depend on furtherarticulating the links between genetic factors,

discounting preferences, and risk for develop-ing an addictive disorder.

Important Methodological Considerationsfor Future Studies

The preceding sections provide the theoreti-cal and empirical basis for pursuing delaydiscounting as an endophenotype for addictivedisorders. This is clearly a promising area ofresearch and success could have a major impactin terms of understanding the genetic basis ofaddiction, the nature of impulsivity, and evenprevention and treatment. Importantly, how-ever, there are a number of considerations andchallenges that pertain to delay discounting ingeneral, but especially genetic research. Tostart, there are issues of measurement withregard to discounting, as a wide array ofassessments are used in the field (MacKillopet al., 2011) and the discounting indices vary interms of how well they correlate with each other,ranging from very high magnitude to negligiblecorrelations (e.g., Epstein et al., 2003; Krishnan-Sarin et al., 2007; MacKillop et al., 2010;Paloyelis et al., 2010). This may seem surprisingat first glance, but when the diversity of com-modities, delay durations, and size of commodi-ty are considered, these variable relationshipsmake more sense. As such, it will be critical forinvestigators to use consistent assessments acrossinvestigations and, ideally, to use multiple mea-sures to determine whether different magni-tudes of rewards are particularly sensitive togenetic influences or create latent variables thatreflect discounting across reward magnitudes.

Fig. 4. Delay discounting as an endophenotype for addictive disorders. Genetic variation related to dopamineneurotransmission is hypothesized to affect the neuroanatomical and neurochemical substrates responsible for variation inintertemporal choice. Variants associated with more impulsive discounting as a trait are hypothesized to constitute a riskfactor for addictive disorders. The number of spaces in the arrows reflects the putative strength of the relationship, with fewerspaces reflecting stronger relationship and vice versa.

BEHAVIORAL ECONOMICS AND BEHAVIORAL GENETICS 11

In addition, one critical issue is that ofreliability, as the magnitude of genetic influen-ces will be necessarily bounded by the reliabilityof the discounting assessment over time. Forexample, the experiential discounting task wasintentionally designed to be sensitive to statechanges (Reynolds & Schiffbauer, 2004), favor-ing sensitivity to dynamic changes in prefer-ences as a result of experimental manipulationsover general reliability across time. This makesthe resulting phenotypemore fluctuating and inturn less likely to be sensitive to genetic effects.These important differences may clarify vari-ability in findings within studies (e.g., Paloyeliset al., 2010). In general, trait-focused assess-ments that have established temporal reliabil-ities would be predicted to be most sensitive asendophenotypic measures.

The issue of consistency applies also to theindex of discounting. In the case of curve fitting,nonlinear regression is the most typical proce-dure but multiple models can be applied andthe solutions almost always includes some levelof error (i.e., R2 <1.0). For this reason, metricslike assumption-free area under the curve(Myerson et al., 2001; Odum, 2011b) orimpulsive choice ratio (Mitchell et al., 2005;Odum, 2011b) may be more appropriate toavoid incorporating that error. Alternatively,differences in the sensitivity of different meth-ods of deriving discounting functions to geneticinfluences may be informative with regard to thenature of discounting and the optimal deriva-tion methods. For example, if reliable associa-tions with specific loci are consistently of largermagnitude using one model compared to theothers, it would provide further support for thatanalytic strategy. In any case, it will be importantfor these methodological issues to be thorough-ly considered and addressed as findings in thisarea accumulate.

A second consideration represents diversedomains, but the common theme is the impor-tance of assessing alternative aspects of theindividual. The risk here is essentially of ‘third-variable’ confounds, or potential relationshipswith alternative or unmeasured variables thatlead to spurious conclusions. In other words,genetic variables or discounting variables maybe associated with each other by virtue of acommon association with an alternative charac-teristic, so it is important to thoroughly examineand incorporate variation on related domains torule out false positives (or false negatives). In

particular, this applies to both collateral demo-graphic and cognitive factors. As a decision-making phenotype, the typical delay discount-ing task using money as a commodity isnecessarily contextualized within factors suchas the individual’s socioeconomic status, as wellas cognitive processing factors such as workingmemory and attention. In the first case, incomehas typically not been found to be substantiallyrelated to discounting (e.g., MacKillop & Tidey,2011), but this is not always the case and the roleof income should be considered nonetheless.In contrast, aspects of cognitive function havebeen consistently associated with discounting(Bickel, Yi, Landes,Hill, & Baxter, 2011; Hinson,Jameson, & Whitney, 2003; Jaroni, Wright,Lerman, & Epstein, 2004; Shamosh et al.,2008). This is not surprising, as the process ofdiscounting decision-making involves simulta-neously juxtaposing the appeal of one immedi-ately available reward in the present withanother of larger magnitude in the future,requiring abstraction, working memory, inhibi-tion, and other cognitive faculties. Moreover,genetic factors have also been implicated incognitive functioning (e.g., Dickinson &Elvevag, 2009), making these processes high-risk confounds if not simultaneously considered.Thus, high-quality assessment of the discountingphenotype would optimally also characterizethese related factors to incorporate theirrole and to avoid misattribution of geneticassociations.

Similar to these preceding factors, anothersource of potential confounding is a failure toaddress comorbidities in sampling. The associa-tion between impulsive discounting and addic-tion is arguably the most well established,but there is also evidence that it is associatedwith several related conditions, such as ADHD(e.g., Hurst, Kepley, McCalla, & Livermore,2011; Scheres et al., 2006), antisocial personalitydisorder (e.g., Acheson et al., 2011; Petry, 2002),and bipolar disorder (e.g., Ahn et al., 2011). Assuch, case-control studies using clinical casescompared to a matched control group will needto be sure that both groups are fully character-ized in terms of present and past psychiatricconditions to avoid potential confounding.More generally, collateral assessment of relevantvariables on continua, such as attentionalcontrol and antisociality, would be optimalin order to characterize the relationship tothe discounting phenotype simultaneously in

12 JAMES MACKILLOP

genetic studies. This issue is essentially the sameas the potential demographic and cognitive con-founds; a well-designed study should be able tospecifically attribute genetic influences to thediscounting phenotype and in turn to addictivebehavior, but not other conditions. Of course,it is also possible that genetic influences maybe determined to be more robust with othercomorbid conditions. Alternatively, it is entirelyplausible that impulsive discounting will be bestunderstood as a risk factor for a number ofconditions, not only addictive disorders, butperhaps disorders characterized by poor self-control or perhaps externalizing disorders ingeneral. As a consequence, genetic variationthat contributes to impulsive discounting maybe risk loci for a broader domain of psychopa-thology. In light of these possibilities, it is criticalfor these collateral characteristics and con-ditions to be ascertained to examine theserelationships empirically.A further consideration is that although there

is relatively extensive evidence that impulsivediscounting predates addictive behavior, asreviewed above, there is also evidence thatpersistent drug exposure can also result in long-term increases in impulsive delay discounting(Dallery & Locey, 2005; Roesch, Takahashi,Gugsa, Bissonette, & Schoenbaum, 2007;Simon,Mendez, & Setlow, 2007) and probabilitydiscounting (Nasrallah, Yang, & Bernstein,2009). Not all studies have found this to bethe case and important methodological consid-erations apply (for a review, see Stein &Madden, in press), but, nonetheless, the aggre-gated findings suggest that impulsive delaydiscounting represents a recursive etiologicalprocess. In other words, impulsive discountingmay well serve as both a risk factor and acharacteristic that can be exacerbated by addic-tive behavior. Impulsive discounting from thisperspective operates as a ‘Matthew Effect’ (i.e.,advantage potentiating further advantage, dis-advantage potentiating further disadvantage),with individuals with innate propensities todiscount the future being at greatest risk ofdeveloping addictive disorders and in turnbecoming progressively more impulsive, andvice versa. This recursive perspective furthercomplicates the phenotypic picture, as a reliablyobserved level of discounting would nonethelessreflect influences from both innate factors andthe longstanding effects of chronic addictivebehavior. As such, fully characterizing the

genetic basis of discounting in nonclinicalsamples will be critical foundational research.Prospective studies characterizing both geneticfactors and addictive behavior across the life-span will ultimately be essential for validlyinvestigating discounting as an endophenotypefor addictive disorders.Finally, it is worth noting that the issue at hand

is a hypothesis, not a truism. The prospect ofimpulsive discounting as an endophenotypeshould be systematically investigated and theresults considered critically, with the recogni-tion that it may not be supported. It is notablethat because endophenotypes are putativelymore proximal to genetic variation, the associ-ations between endophenotypes and specificgenotypes are theorized to be larger in magni-tudes and more reliable. In a meta-analysis ofseveral of the most widely studied endopheno-types, however, Flint and Munafò (2007) didnot find evidence of larger effect sizes. Theyrecommended caution against presuming thatendophenotypes necessarily have a simplegenetic architecture, and this may well be thecase for discounting also. Furthermore, inthinking critically about genetic influences ondiscounting, there are potentially influencesbeyond individual genetic factors. For example,impulsive discounting may in fact be a functionof one or more gene x environment interac-tions, meaning that genetic influences may onlybe evident in individuals with certain environ-mental exposures, such as early life stress (e.g.,Enoch, 2012). Alternatively, if level of impulsivediscounting is indeed largely a function of lowcorticomesolimbic dopamine tone, it is possiblethat an array of combinations of genetic variantsmight contribute to this systems-level character-istic, making impulsive discounting nonspecificto individual loci and an emergent property ofthe system. For example, variants responsible fortransporter molecules clearing dopamine fromthe synapse and variants responsible for differ-ences in the density or binding potential ofdopamine receptor subtypes could both leadto the same aggregate profile of dopaminergicactivity, high or low. Based on the model inFigure 4, these systems-level effects would beassociated with commensurate differences indiscounting, but this would happen by way ofentirely different loci. Capturing relationshipslike this may require novel strategies, such asexamining aggregate genetic risk scores (e.g.,Derringer et al., 2010; Karlson et al., 2010) or

BEHAVIORAL ECONOMICS AND BEHAVIORAL GENETICS 13

examining combinatorial genetic profiles thathave known systemic effects (e.g., contrastingindividuals who are positive and negative for apanel of risk or hypofunction variants). Funda-mentally, although the findings to date arepromising, it is important to recognize thatthe endophenotype approach is not the onlypossible framework for understanding geneticinfluences on discounting and the multiplicityof influences and mechanisms are worthy ofconsideration.

Conclusions

The goal of the current review was to lay outthe theoretical and empirical basis for consider-ing impulsive delay discounting as an endophe-notype for addictive disorders. Although it isonly a recent area of focus, there is considerableevidence supporting this perspective. As aphenotype, impulsive discounting is stableover time, heritable, and robustly associatedwith addictive behavior. A small number ofstudies have identified direct associations be-tween genetic variation and variation in dis-counting, and several preclinical studies havetriangulated genetics, discounting, and addic-tion liability, but these links remain considerablyless well established. Moreover, caution is war-ranted in light of a history of highly promisinginitial findings that are ultimately false starts (orat least much more modest starts) in addictiongenetics. This is particularly important becausethere are methodological issues pertaining todiscounting as a phenotype that require carefulconsideration. Nonetheless, the existing litera-ture provides strong basis for systematicallyinvestigating this hypothesis.

References

Acheson, A., Vincent, A. S., Sorocco, K. H., & Lovallo, W. R.(2011). Greater discounting of delayed rewards inyoung adults with family histories of alcohol and druguse disorders: studies from the Oklahoma family healthpatterns project. Alcoholism: Clinical and ExperimentalResearch, 35, 1607–1613.

Ahn, W. Y., Rass, O., Fridberg, D. J., Bishara, A. J., Forsyth, J.K., Breier, A., … O’Donnell, B. F. (2011). Temporaldiscounting of rewards in patients with bipolar disorderand schizophrenia. Journal of Abnormal Psychology, 120(4), 911–921.

Ainslie, G. (1975). Specious reward: a behavioral theory ofimpulsiveness and impulse control. Psychological Bulletin,82(4), 463–496.

Ainslie, G. (2001). Breakdown of Will. Cambridge UniversityPress.

Alessi, S. M., & Petry, N. M. (2003). Pathological gamblingseverity is associated with impulsivity in a delay dis-counting procedure. Behavioural Processes, 64(3), 345–354.

Anderson, K. G., & Woolverton, W. L. (2005). Effects ofclomipramine on self-control choice in Lewis andFischer 344 rats. Pharmacology, Biochemistry, and Behavior,80(3), 387–393.

Anker, J. J., Perry, J. L., Gliddon, L. A., & Carroll, M. E.(2009). Impulsivity predicts the escalation of cocaineself-administration in rats. Pharmacology, Biochemistry,and Behavior, 93(3), 343–348.

Anokhin, A. P., Golosheykin, S., Grant, J. D., & Heath, A. C.(2011). Heritability of delay discounting in adolescence:a longitudinal twin study. Behavioral Genetics, 41(2), 175–183.

Asherson, P., Brookes, K., Franke, B., Chen, W., Gill, M.,Ebstein, R. P., … Faraone, S. V. (2007). Confirmationthat a specific haplotype of the dopamine transportergene is associated with combined-type ADHD. AmericanJournal of Psychiatry, 164, 674–677.

Audrain-McGovern, J., Rodriguez, D., Epstein, L. H., Cuevas,J., Rodgers, K., & Wileyto, E. P. (2009). Does delaydiscounting play an etiological role in smoking or is it aconsequence of smoking? Drug and Alcohol Dependence,103(3), 99–106.

Ayduk, O., Mendoza-Denton, R., Mischel, W., Downey, G.,Peake, P. K., & Rodriguez, M. (2000). Regulating theinterpersonal self: strategic self-regulation for copingwith rejection sensitivity. Journal of Personality and SocialPsychology, 79(5), 776–792.

Baker, F., Johnson, M. W., & Bickel, W. K. (2003). Delaydiscounting in current and never-before cigarettesmokers: similarities and differences across commodity,sign, and magnitude. Journal of Abnormal Psychology, 112(3), 382–392.

Beck, J. A., Lloyd, S., Hafezparast, M., Lennon-Pierce,M., Eppig, J. T., Festing, M. F, et al. (2000). Genealogiesof mouse inbred strains. Nature Genetics, 24(1),23–25.

Beck, R. C., & Triplett, M. F. (2009). Test–retest reliability ofa group-administered paper-pencil measure of delaydiscounting. Experimental and Clinical Psychopharmacolo-gy, 17(5), 345–355.

Bevilacqua, L., Doly, S., Kaprio, J., Yuan, Q., Tikkanen, R.,Paunio, T., … Goldman, D. (2010). A population-specific HTR2B stop codon predisposes to severeimpulsivity. Nature, 468(7327), 1061–1066.

Bickel, W. K., Odum, A. L., & Madden, G. J. (1999).Impulsivity and cigarette smoking: delay discounting incurrent, never, and ex-smokers. Psychopharmacology(Berl), 146(4), 447–454.

Bickel, W. K., Pitcock, J. A., Yi, R., & Angtuaco, E. J. (2009).Congruence of BOLD response across intertemporalchoice conditions: fictive and real money gains andlosses. Journal of Neuroscience, 29(27), 8839–8846.

Bickel, W. K., Yi, R., Landes, R. D., Hill, P. F., & Baxter, C.(2011). Remember the future: working memory train-ing decreases delay discounting among stimulantaddicts. Biological Psychiatry, 69(3), 260–265.

Bobova, L., Finn, P. R., Rickert, M. E., & Lucas, J. (2009).Disinhibitory psychopathology and delay discountingin alcohol dependence: personality and cognitive

14 JAMES MACKILLOP

correlates. Experimental and Clinical Psychopharmacology,17(1), 51–61.

Boettiger, C. A.,Mitchell, J. M., Tavares, V. C., Robertson,M.,Joslyn, G., D’Esposito, M, et al. (2007). Immediatereward bias in humans: fronto-parietal networks and arole for the catechol-O-methyltransferase 158(Val/Val)genotype. Journal of Neuroscience, 27(52), 14383–14391.

Bradford, W. D. (2010). The association between individualtime preferences and health maintenance habits.Medical Decision Making, 30(1), 99–112.

Cardinal, R. N. (2006). Neural systems implicated in delayedand probabilistic reinforcement. Neural Networks, 19(8),1277–1301.

Cardinal, R. N., & Howes, N. J. (2005). Effects of lesions ofthe nucleus accumbens core on choice between smallcertain rewards and large uncertain rewards in rats.BMC Neuroscience, 6, 37.

Cardinal, R. N., Pennicott, D. R., Sugathapala, C. L., Robbins,T. W., & Everitt, B. J. (2001). Impulsive choice inducedin rats by lesions of the nucleus accumbens core. Science,292(5526), 2499–2501.

Carroll, M. E., Anderson, M. M., & Morgan, A. D. (2007).Regulation of intravenous cocaine self-administration inrats selectively bred for high (HiS) and low (LoS)saccharin intake. Psychopharmacology (Berl), 190(3), 331–341.

Carroll, M. E., Morgan, A. D., Lynch, W. J., Campbell, U. C.,& Dess, N. K. (2002). Intravenous cocaine and heroinself-administration in rats selectively bred for differen-tial saccharin intake: phenotype and sex differences.Psychopharmacology (Berl), 161(3), 304–313.

Carter, R. M., Meyer, J. R., & Huettel, S. A. (2010).Functional neuroimaging of intertemporal choicemodels: A review. Journal of Neuroscience, Psychology, andEconomics, 3(1), 27–45.

Clare, S., Helps, S., & Sonuga-Barke, E. J. (2010). The quickdelay questionnaire: a measure of delay aversion anddiscounting in adults. Attention Deficit HyperactivityDisorder, 2(1), 43–48.

Coffey, S. F., Gudleski, G. D., Saladin, M. E., & Brady, K. T.(2003). Impulsivity and rapid discounting of delayedhypothetical rewards in cocaine-dependent individuals.Experimental and Clinical Psychopharmacology, 11(1),18–25.

Crean, J., Richards, J. B., & de Wit, H. (2002). Effect oftryptophan depletion on impulsive behavior in menwith or without a family history of alcoholism. BehavioralBrain Research, 136(2), 349–357.

Dallery, J., & Locey, M. L. (2005). Effects of acute andchronic nicotine on impulsive choice in rats. BehavioralPharmacology, 16(1), 15–23.

Derringer, J., Krueger, R. F., Dick, D.M., Saccone, S., Grucza,R. A., Agrawal, A., … Gene Environment AssociationStudies (GENEVA) Consortium. (2010). Predictingsensation seeking from dopamine genes. A candidate-system approach. Psychological Science, 21, 1282–1290.

Dickinson, D., & Elvevag, B. (2009). Genes, cognition andbrain through a COMT lens. Neuroscience, 164(1),72–87.

Dom, G., D’Haene, P., Hulstijn, W., & Sabbe, B. (2006).Impulsivity in abstinent early- and late-onset alcoholics:differences in self-report measures and a discountingtask. Addiction, 101(1), 50–59.

Eisenberg, D. T., Mackillop, J., Modi, M., Beauchemin, J.,Dang, D., Lisman, S. A., … Wilson, D. S. (2007).

Examining impulsivity as an endophenotype using abehavioral approach: a DRD2 TaqI A and DRD4 48-bpVNTR association study. Behavior and Brain Functions,3, 2.

Enoch, M. A. (2012). The influence of gene–environmentinteractions on the development of alcoholism anddrug dependence.Current Psychiatry Reports, 14, 150–158.

Epstein, L. H., Richards, J. B., Saad, F. G., Paluch, R. A.,Roemmich, J. N., & Lerman, C. (2003). Comparisonbetween two measures of delay discounting in smokers.Experimental and Clinical Psychopharmacology, 11, 131–138.

Etter, J.-F., Perneger, T. V., & Ronchi, A. (1997). Distribu-tions of smokers by stage: International comparison andassociation with smoking prevalence. PreventiveMedicine:An International Journal Devoted to Practice and Theory, 26(4), 580–585.

Flint, J., & Munafò, M. R. (2007). The endophenotypeconcept in psychiatric genetics. Psychological Medicine, 37(2), 163–180.

Flores, G., Wood, G. K., Barbeau, D., Quirion, R., &Srivastava, L. K. (1998). Lewis and Fischer rats: acomparison of dopamine transporter and receptorslevels. Brain Research, 814, 34–40.

Frederick, S., Loewenstein, G., & O’Donoghue, T. (2003).Time discounting and time preference: A critical review.In: G. Loewenstein, D. Read, & R. Baumeister (Eds.),Time and decision: Economic and psychological perspectives onintertemporal choice. (pp. 13–86). New York: Russell SageFoundation.

Garcia, J. R., MacKillop, J., Aller, E. L., Merriwether, A. M.,Wilson, D. S., & Lum, J. K. (2010). Associations betweendopamine D4 receptor gene variation with bothinfidelity and sexual promiscuity. PLoS ONE, 5(11),e14162.

Goldman, D., Oroszi, G., & Ducci, F. (2005). The genetics ofaddictions: uncovering the genes. Nature Reviews:Genetics, 6(7), 521–532.

Gottesman, I. I., & Gould, T. D. (2003). The endophenotypeconcept in psychiatry: etymology and strategic inten-tions. American Journal of Psychiatry, 160(4), 636–645.

Gottesman, I. I., & Shields, J. (1973). Genetic theorizing andschizophrenia. British Journal of Psychiatry, 122(566),15–30.

Green, L., & Myerson, J. (2004). A discounting frameworkfor choice with delayed and probabilistic rewards.Psychological Bulletin, 130(5), 769–792.

Guttmannova, K., Bailey, J. A., Hill, K. G., Lee, J. O., Hawkins,J. D., Woods, M. L, et al. (2011). Sensitive periods foradolescent alcohol use initiation: predicting the lifetimeoccurrence and chronicity of alcohol problems inadulthood. Journal of Studies on Alcohol and Drugs, 72(2), 221–231.

Hasler, G., Drevets, W. C., Gould, T. D., Gottesman, I. I., &Manji, H. K. (2006). Toward constructing an endophe-notype strategy for bipolar disorders. Biological Psychia-try, 60(2), 93–105.

Helms, C.M., Gubner, N. R.,Wilhelm, C. J., Mitchell, S. H., &Grandy, D. K. (2008). D4 receptor deficiency in micehas limited effects on impulsivity and novelty seeking.Pharmacology, Biochemistry, and Behavior, 90(3), 387–393.

Helms, C. M., Reeves, J. M., & Mitchell, S. H. (2006). Impactof strain and D-amphetamine on impulsivity (delaydiscounting) in inbred mice. Psychopharmacology (Berl),188(2), 144–151.

BEHAVIORAL ECONOMICS AND BEHAVIORAL GENETICS 15

Herting, M. M., Schwartz, D., Mitchell, S. H., & Nagel, B. J.(2010). Delay discounting behavior and white mattermicrostructure abnormalities in youth with a familyhistory of alcoholism. Alcoholism: Clinical and Experimen-tal Research, 34(9), 1590–1602.

Hesselbrock, V. M., & Hesselbrock, M. N. (2006). Are thereempirically supported and clinically useful subtypes ofalcohol dependence? Addiction, 101(Suppl 1), 97–103.

Hinson, J. M., Jameson, T. L., & Whitney, P. (2003). Im-pulsive decisionmaking and workingmemory. Journal ofExperimental Psychology: Learning, Memory, and Cognition,29(2), 298–306.

Hirvonen, M., Laakso, A., Någren, K., Rinne, J. O.,Pohjalainen, T., & Hietala, J. (2004). C957T polymor-phism of the dopamine D2 receptor (DRD2) geneaffects striatal DRD2 availability in vivo. MolecularPsychiatry, 9, 1060–1061.

Hogue, A., Dauber, S., & Morgenstern, J. (2010). Validationof a contemplation ladder in an adult substance usedisorder sample. Psychology of Addictive Behaviors, 24(1),137–144.

Hurst, R. M., Kepley, H. O., McCalla, M. K., & Livermore, M.K. (2011). Internal consistency and discriminant validityof a delay-discounting task with an adult self-reportedADHD sample. Journal of Attention Disorders, 15, 412–422.

Isles, A. R., Humby, T., Walters, E., &Wilkinson, L. S. (2004).Common genetic effects on variation in impulsivity andactivity in mice. Journal of Neuroscience, 24(30), 6733–6740.

Jaroni, J. L., Wright, S. M., Lerman, C., & Epstein, L. H.(2004). Relationship between education and delaydiscounting in smokers. Addictive Behaviors, 29(6),1171–1175.

Johnson, E. O., van den Bree, M. B., & Pickens, R. W. (1996).Subtypes of alcohol-dependent men: a typology basedon relative genetic and environmental loading. Alcohol-ism: Clinical and Experimental Research, 20(8), 1472–1480.

Jonsson, E., Sedvall, G., Brene, S., Gustavsson, J. P., Geijer, T.,Terenius, L., …Wiesel, F.-A. (1996). Dopamine-relatedgenes and their relationships to monoamine metabo-lites in CSF. Biological Psychiatry, 40(10), 1032–1043.

Jonsson, E. G., Nothen, M. M., Grunhage, F., Farde, L.,Nakashima, Y., Propping, P, et al. (1999). Polymor-phisms in the dopamine D2 receptor gene and theirrelationships to striatal dopamine receptor density ofhealthy volunteers. Molecular Psychiatry, 4(3), 290–296.

Kable, J. W., &Glimcher, P. W. (2007). The neural correlatesof subjective value during intertemporal choice. NatureNeuroscience, 10(12), 1625–1633.

Karlson, E. W., Chibnik, L. B., Kraft, P., Cui, J., Keenan, B. T.,Ding, B., … Plenge, R. M. (2010). Cumulative associa-tion of 22 genetic variants with seropositive rheumatoidarthritis risk. Annals of the Rheumatic Diseases, 69, 1077–1085. DOI: 10.1136/ard.2009.120170.

Killeen, P. R. (2009). An additive-utility model of delaydiscounting. Psychological Review., 116(3), 602–619.

Kirby, K. N. (2009). One-year temporal stability of delay-discount rates. Psychonomic Bulletin and Review, 16(3),457–462.

Kirby, K. N., Petry, N. M., & Bickel, W. K. (1999). Heroinaddicts have higher discount rates for delayed rewardsthan non-drug-using controls. Journal of ExperimentalPsychology: General, 128(1), 78–87.

Koffarnus, M. N., Newman, A. H., Grundt, P., Rice, K. C., &Woods, J. H. (2011). Effects of selective dopaminergic

compounds on a delay-discounting task. BehavioralPharmacology, 22(4), 300–311.

Kollins, S. H. (2003). Delay discounting is associated withsubstance use in college students. Addictive Behaviors, 28(6), 1167–1173.

Krishnan-Sarin, S., Reynolds, B., Duhig, A.M., Smith, A., Liss,T., McFetridge, A.,… Potenza, M. N. (2007). Behavioralimpulsivity predicts treatment outcome in a smokingcessation program for adolescent smokers. Drug andAlcohol Dependence, 88(1), 79–82.

Laibson, D. (1997). Golden eggs and hyperbolic discount-ing. The Quarterly Journal of Economics, 112(443–477).

MacKillop, J., Amlung, M. T., Few, L. R., Ray, L. A., Sweet, L.H., & Munafò, M. R. (2011). Delayed reward discount-ing and addictive behavior: a meta-analysis. Psychophar-macology (Berl), 216(3), 305–321.

MacKillop, J., Anderson, E. J., Castelda, B. A., Mattson, R. E.,& Donovick, P. J. (2006). Divergent validity of measuresof cognitive distortions, impulsivity, and time perspec-tive in pathological gambling. Journal of Gambling Studies,22(3), 339–354.

MacKillop, J., & Kahler, C. W. (2009). Delayed rewarddiscounting predicts treatment response for heavydrinkers receiving smoking cessation treatment. Drugand Alcohol Dependence, 104(3), 197–203.

MacKillop, J., Miranda, J. R., Monti, P. M., Ray, L. A., Tidey,J. W., Rohsenow, D. J.,…Gwaltney, C. J. (2010). Alcoholdemand, delayed reward discounting, and craving inrelation to drinking and alcohol use disorders. Journal ofAbnormal Psychology, 119, 115–125.

MacKillop, J., & Tidey, J. W. (2011). Cigarette demand anddelayed reward discounting in nicotine-dependentindividuals with schizophrenia and controls: an initialstudy. Psychopharmacology (Berl), 216(1), 91–99.

Madden, G. J., Petry, N. M., Badger, G. J., & Bickel, W. K.(1997). Impulsive and self-control choices in opioid-dependent patients and non-drug-using control partic-ipants: drug and monetary rewards. Experimental andClinical Psychopharmacology, 5(3), 256–262.

Madden, G. J., Petry, N. M., & Johnson, P. S. (2009).Pathological gamblers discount probabilistic rewardsless steeply than matched controls. Experimental andClinical Psychopharmacology, 17(5), 283–290.

Madden, G. J., Smith, N. G., Brewer, A. T., Pinkston, J. W., &Johnson, P. S. (2008). Steady-state assessment ofimpulsive choice in Lewis and Fischer 344 rats:between-condition delay manipulations. Journal of theExperimental Analysis of Behavior, 90(3), 333–344.

Marusich, J. A., & Bardo, M. T. (2009). Differences inimpulsivity on a delay-discounting task predict self-administration of a low unit dose of methylphenidate inrats. Behavioural Pharmacology, 20(5–6), 447–454.

Mazur, J. E. (1987). An adjusting procedure for studyingdelayed reinforcement. In: M. L. Commons, J. E. Mazur,J. A. Nevin, & H. Rachlin (Eds.), The Effect of Delay and ofIntervening Events on Reinforcement Value: QuantitativeAnalyses of Behavior (Vol. 5, pp. 55–73). Hillsdale, NJ:Lawrence Erlbaum Associates Inc.

McClure, S. M., Laibson, D. I., Loewenstein, G., & Cohen, J.D. (2004). Separate neural systems value immediate anddelayed monetary rewards. Science, 306(5695), 503–507.

McGeary, J. (2009). The DRD4 exon 3 VNTR polymorphismand addiction-related phenotypes: A review. Pharmacol-ogy, Biochemistry, and Behavior, 93(3), 222–229.

16 JAMES MACKILLOP

McKay, J. R., Franklin, T. R., Patapis, N., & Lynch, K. G.(2006). Conceptual, methodological, and analyticalissues in the study of relapse. Clinical Psychology Review,26(2), 109–127.

Mischel, W. (1958). Preference for delayed reinforcement:An experimental study of a cultural observation. Journalof Abnormal and Social Psychology, 56(1), 57–61.

Mischel, W. (1961). Father-absence and delay of gratifica-tion: Crosscultural comparisons. Journal of Abnormal andSocial Psychology, 63, 116–124.

Mischel, W., Ebbesen, E. B., & Zeiss, A. R. (1972). Cognitiveand attentional mechanisms in delay of gratification.Journal of Personality and Social Psychology, 21(2), 204–218.

Mitchell, J. M., Fields, H. L., D’Esposito, M., & Boettiger, C.A. (2005). Impulsive responding in alcoholics. Alcohol-ism: Clinical and Experimental Research, 29(12), 2158–2169.

Mobini, S., Chiang, T. J., Al-Ruwaitea, A. S., Ho, M. Y.,Bradshaw, C. M., & Szabadi, E. (2000). Effect of central5-hydroxytryptamine depletion on inter-temporalchoice: a quantitative analysis. Psychopharmacology(Berl), 149(3), 313–318.

Mobini, S., Chiang, T. J., Ho, M. Y., Bradshaw, C. M., &Szabadi, E. (2000). Effects of central 5-hydroxytrypta-mine depletion on sensitivity to delayed and probabilis-tic reinforcement. Psychopharmacology (Berl), 152(4),390–397.

Moss, H. B., Chen, C. M., & Yi, H. Y. (2007). Subtypes ofalcohol dependence in a nationally representativesample. Drug and Alcohol Dependence, 91(2–3), 149–158.

Murphy, C. M., & MacKillop, J. (2012). Living in the hereand now: Interrelationships between impulsivity, mind-fulness, and alcohol misuse. Psychopharmacology (Berl.),219, 527–536.

Myerson, J., & Green, L. (1995). Discounting of delayedrewards: Models of individual choice. Journal of theExperimental Analysis of Behavior, 64(3), 263–276.

Myerson, J., Green, L., & Warusawitharana, M. (2001). Areaunder the curve as a measure of discounting. Journal ofthe Experimental Analysis of Behavior, 76(2), 235–243.

Nasrallah, N. A., Yang, T. W., & Bernstein, I. L. (2009). Long-term risk preference and suboptimal decision makingfollowing adolescent alcohol use. PNAS, 106(41),17600–17604.

Neville, M. J., Johnstone, E. C., & Walton, R. T. (2004).Identification and characterization of ANK K1: a novelkinase gene closely linked to DRD2 on chromosomeband 11q23.1. Human Mutation, 23(6), 540–545.

Oberlin, B. G., & Grahame, N. J. (2009). High-alcoholpreferring mice are more impulsive than low-alcoholpreferring mice as measured in the delay discountingtask. Alcoholism: Clinical and Experimental Research, 33(7),1294–1303.

Odum, A. L. (2011a). Delay discounting: trait variable?Behavioural Processes, 87(1), 1–9.

Odum, A. L. (2011b). Delay discounting: I’m a k, you’re a k.Journal of the Experimental Analysis of Behavior, 96, 427–439.

Paloyelis, Y., Asherson, P., Mehta, M. A., Faraone, S. V.,& Kuntsi, J. (2010). DAT1 and COMT effects ondelay discounting and trait impulsivity in male adoles-cents with attention deficit/hyperactivity disorderand healthy controls. Neuropsychopharmacology, 35(12),2414–2426.

Passetti, F., Clark, L., Davis, P., Mehta, M. A., White, S.,Checinski, K., King, M., & Abou‐Saleh, M. (2011). Riskydecision‐making predicts short‐term outcome of com-munity but not residential treatment for opiate addic-tion. Implications for case management. Drug andAlcohol Dependence,118, 12–18.

Perry, J. L., Larson, E. B., German, J. P., Madden, G. J., &Carroll, M. E. (2005). Impulsivity (delay discounting) asa predictor of acquisition of IV cocaine self-administra-tion in female rats. Psychopharmacology (Berl), 178(2–3),193–201.

Perry, J. L., Morgan, A. D., Anker, J. J., Dess, N. K., & Carroll,M. E. (2006). Escalation of i.v. cocaine self-administra-tion and reinstatement of cocaine-seeking behavior inrats bred for high and low saccharin intake. Psychophar-macology (Berl), 186(2), 235–245.

Perry, J. L., Nelson, S. E., Anderson, M. M., Morgan, A. D., &Carroll, M. E. (2007). Impulsivity (delay discounting)for food and cocaine in male and female rats selectivelybred for high and low saccharin intake. Pharmacology,Biochemistry, and Behavior, 86(4), 822–837.

Petry, N. M. (2001). Delay discounting of money and alcoholin actively using alcoholics, currently abstinent alco-holics, and controls. Psychopharmacology (Berl), 154(3),243–250.

Petry, N. M. (2002). Discounting of delayed rewards insubstance abusers: relationship to antisocial personalitydisorder. Psychopharmacology (Berl), 162(4), 425–432.

Petry, N. M., Kirby, K. N., & Kranzler, H. R. (2002). Effects ofgender and family history of alcohol dependence on abehavioral task of impulsivity in healthy subjects. Journalof Studies on Alcohol, 63(1), 83–90.

Pine, A., Shiner, T., Seymour, B., & Dolan, R. J. (2010).Dopamine, time, and impulsivity in humans. Journal ofNeuroscience, 30(26), 8888–8896.

Rachlin, H., & Green, L. (1972). Commitment, choice andself-control. Journal of the Experimental Analysis of Behavior,17(1), 15–22.

Reimers, S., Maylor, E. A., Stewart, N., & Chater, N. (2009).Associations between a one-shot delay discountingmeasure and age, income, education and real-worldimpulsive behavior. Personality and Individual Differences,47, 973–978.

Reynolds, B., & Schiffbauer, R. (2004). Measuring statechanges in human delay discounting: an experientialdiscounting task. Behavioural Processes, 67(3), 343–356.

Roesch, M. R., Takahashi, Y., Gugsa, N., Bissonette, G. B., &Schoenbaum, G. (2007). Previous cocaine exposuremakes rats hypersensitive to both delay and rewardmagnitude. Journal of Neuroscience, 27(1), 245–250.

Savitz, J., Solms, M., & Ramesar, R. (2006). The moleculargenetics of cognition: dopamine, COMT and BDNF.Genes, Brain, and Behavior, 5, 311–328.

Scheres, A., Dijkstra, M., Ainslie, E., Balkan, J., Reynolds,B., Sonuga-Barke, E, et al. (2006). Temporal andprobabilistic discounting of rewards in children andadolescents: effects of age and ADHD symptoms.Neuropsychologia, 44(11), 2092–2103.

Shamosh, N. A., Deyoung, C. G., Green, A. E., Reis, D. L.,Johnson, M. R., Conway, A. R., … Gray, J. R. (2008).Individual differences in delay discounting: relation tointelligence, working memory, and anterior prefrontalcortex. Psychological Science, 19(9), 904–911.

Sheffer, C., MacKillop, J., McGeary, J., Landes, R., Carter, L.,Yi, R., Jones, B.,… Bickel, W. (2012). Delay discounting,

BEHAVIORAL ECONOMICS AND BEHAVIORAL GENETICS 17

locus of control, and cognitive impulsiveness inde-pendently predict tobacco dependence treatment out-comes in a highly dependent, lower socioeconomicgroup of smokers. American Journal on Addictions, 21,221–232. DOI: 10.1111/j.1521-0391.2012.00224.x.

Simon, N. W., Mendez, I. A., & Setlow, B. (2007). Cocaineexposure causes long-term increases in impulsivechoice. Behavioral Neuroscience, 121(3), 543–549.

Simpson, C. A., & Vuchinich, R. E. (2000). Reliability of ameasure of temporal discounting. The PsychologicalRecord, 50, 3–16.

Stanger, C., Ryan, S. R., Fu, H., Landes, R. D., Jones, B. A.,Bickel, W. K, et al. (2012). Delay discounting predictsadolescent substance abuse treatment outcome. Experi-mental and Clinical Psychopharmacology, 20, 205–212.

Stein, J. S., Madden, G. J. (in press). Delay discounting anddrug abuse: empirical, conceptual, and methodologicalconsiderations. In: J. MacKillop & H. de Wit (Eds.), TheWiley-Blackwell Handbook on Addiction Psychopharmacology.Oxford, UK: Wiley-Blackwell.

Stein, J. S., Pinkston, J. W., Brewer, A. T., Francisco, M. T., &Madden, G. J. (2012). Delay discounting in Lewis andFischer 344 rats: Steady-state and rapid-determinationadjusting-amount procedures. Journal of the ExperimentalAnalysis of Behavior, 97, 305–321.

Sweitzer, M. M., Donny, E. C., Dierker, L. C., Flory, J. D., &Manuck, S. B. (2008). Delay discounting and smoking:Association with the Fagerstrom Test for NicotineDependence but not cigarettes smoked per day.Nicotineand Tobacco Research, 10(10), 1571–1575.

Takahashi, T., Furukawa, A., Miyakawa, T., Maesato, H., &Higuchi, S. (2007). Two-month stability of hyperbolicdiscount rates for delayed monetary gains in abstinentinpatient alcoholics. Neurology and Endocrinology Letters,28(2), 131–136.

Treutlein, J., & Rietschel, M. (2011). Genome-wide associa-tion studies of alcohol dependence and substance usedisorders. Current Psychiatry Reports, 13, 147–155.

Turkheimer, E. (2011). Still missing. Research in HumanDevelopment, 8, 227–241.

Vuchinich, R. E., & Simpson, C. A. (1998). Hyperbolictemporal discounting in social drinkers and problemdrinkers. Experimental and Clinical Psychopharmacology, 6(3), 292–305.

Weatherly, J. N., Derenne, A., & Terrell, H. K. (2011).Testing the reliability of delay discounting of tencommodities using the fill-in-the-blank method. ThePsychological Record, 61(1), 113–126.

White, M. J., Lawford, B. R., Morris, C. P., & Young, R. M.(2009). Interaction between DRD2 C957T polymor-phism and an acute psychosocial stressor on reward-related behavioral impulsivity. Behavioral Genetics, 39(3),285–295.

Wilhelm, C. J., & Mitchell, S. H. (2008). Rats bred for highalcohol drinking are more sensitive to delayed andprobabilistic outcomes. Genes, Brain, and Behavior, 7(7),705–713.

Wilhelm, C. J., & Mitchell, S. H. (2009). Strain differences indelay discounting using inbred rats. Genes, Brain, andBehavior, 8, 426–434.

Wilhelm, C. J., Reeves, J. M., Phillips, T. J., & Mitchell,S. H. (2007). Mouse lines selected for alcohol con-sumption differ on certain measures of impulsivity.Alcoholism: Clinical and Experimental Research, 31(11),1839–1845.

Winstanley, C. A., Dalley, J. W., Theobald, D. E., & Robbins,T. W. (2003). Global 5-HT depletion attenuates theability of amphetamine to decrease impulsive choice ona delay-discounting task in rats. Psychopharmacology (Berl),170(3), 320–331.

Winstanley, C. A., Theobald, D. E., Dalley, J. W., & Robbins,T. W. (2005). Interactions between serotonin anddopamine in the control of impulsive choice in rats:therapeutic implications for impulse control disorders.Neuropsychopharmacology, 30(4), 669–682.

Wulfert, E., Block, J. A., Santa Ana, E., Rodriguez, M. L.,& Colsman, M. (2002). Delay of gratification: impul-sive choices and problem behaviors in early andlate adolescence. Journal of Personality, 70(4), 533–552.

Yoon, J. H., Higgins, S. T., Heil, S. H., Sugarbaker, R. J.,Thomas, C. S., & Badger, G. J. (2007). Delay discountingpredicts postpartum relapse to cigarette smokingamong pregnant women. Experimental and ClinicalPsychopharmacology, 15(2), 176–186.

Received: December 23, 2011Final Acceptance: July 5, 2012

18 JAMES MACKILLOP