Elsevier

Behavioural Processes

Volume 64, Issue 3, 31 October 2003, Pages 345-354
Behavioural Processes

Pathological gambling severity is associated with impulsivity in a delay discounting procedure

https://doi.org/10.1016/S0376-6357(03)00150-5Get rights and content

Abstract

Research and clinical expertise indicates that impulsivity is an underlying feature of pathological gambling. This study examined the extent to which impulsive behavior, defined by the rate of discounting delayed monetary rewards, varies with pathological gambling severity, assessed by the South Oaks Gambling Screen (SOGS). Sixty-two pathological gamblers completed a delay discounting task, the SOGS, the Eysenck impulsivity scale, the Addiction Severity Index (ASI), and questions about gambling and substance use at intake to outpatient treatment for pathological gambling. In the delay discounting task, participants chose between a large delayed reward (US $1000) and smaller more immediate rewards (US $1–$999) across a range of delays (6 h to 25 years). The rate at which the delayed reward was discounted (k value) was derived for each participant and linear regression was used to identify the variables that predicted k values. Age, gender, years of education, substance abuse treatment history, and cigarette smoking history failed to significantly predict k values. Scores on the Eysenck impulsivity scale and the SOGS both accounted for a significant proportion of the variance in k values. The predictive value of the SOGS was 1.4 times that of the Eysenck scale. These results indicate that of the measures tested, gambling severity was the best single predictor of impulsive behavior in a delay discounting task in this sample of pathological gamblers.

Introduction

Pathological gambling is a maladaptive pattern of gambling behavior that persists despite substantial adverse consequences. Pathological gamblers expend a remarkable amount of money, time, and emotional resources on gambling. Typically, they incur substantial debt and experience family and social relationship problems because of gambling. Some pathological gamblers even loose their jobs or engage in illegal activities to support their gambling (American Psychiatric Association (APA), 1994; Petry and Armentano, 1999). Pathological gambling affects an estimated 1.6% of the adult population in the United States and Canada (Shaffer et al., 1999).

Research and clinical expertise suggest that the construct of impulsivity is integral to understanding pathological gambling behavior (DSM-IV; APA, 1994). Impulsivity as measured on personality inventories (e.g. the Eysenck impulsivity scale; Eysenck and Eysenck, 1978) or other questionnaires is generally elevated in gamblers (Carlton and Manowitz, 1994, Castellani and Rugle, 1995, Blanco et al., 1996, Blaszczynski et al., 1997, Vitaro et al., 1997, Steel and Blaszczynski, 1998, Barnes et al., 1999, Vitaro et al., 1998, Blaszczynski, 1999, Petry, 2001a, Vitaro et al., 2001, Petry, 2002). For example, pathological gamblers have higher Eysenck impulsivity scale scores compared to a control group (Petry, 2001a) and normative data (Blaszczynski et al., 1997, Steel and Blaszczynski, 1998).

There is general consensus that personality assessments indicate elevated levels of impulsivity in pathological gamblers relative to controls. However, research has not fully elucidated the nature of the relationship between impulsivity and severity of gambling problems. Blaszczynski et al. (1997) reported that pathological gamblers who engaged in gambling-related illegal activities had greater scores on the Eysenck impulsivity scale (Eysenck et al., 1985) than gamblers who did not engage in gambling-related illegal activities. Steel and Blaszczynski (1998) operationally defined severity of pathological gambling by scores on the South Oaks Gambling Screen (SOGS; Lesieur and Blume, 1987) and found that SOGS scores were positively correlated with the impulsivity and non-planning subscales of the Eysenck Personality Inventory (Eysenck and Eysenck, 1978) in pathological gamblers.

Impulsivity is a multidimensional construct (e.g. Eysenck and Eysenck, 1978). Therefore, a body of research that incorporates a variety of impulsivity assessment tools is necessary to provide convergent and divergent evidence for the nature of the relationship between ‘impulsivity’ and pathological gambling severity. Research on the relationship between impulsivity and pathological gambling to date primarily has used personality assessments to define impulsivity. The tendency to act impulsively in a variety of general contexts is typically assessed on these questionnaires (e.g. “Do you often do things spur of the moment?”). Impulsive behavior specifically involving monetary choices may, however, be particularly relevant to pathological gambling given the nature of the disorder.

One potentially useful metric of impulsive decisions regarding money is the behavioral economic process of temporal discounting of delayed rewards. Delay discounting refers to the observation that when given the choice between a relatively small magnitude reward delivered immediately and a relatively large magnitude reward delivered after some delay, the length of the delay and the magnitude of the reward govern preference. Preference for the small immediate reward over the large delayed reward indicates an impulsive choice, and the converse indicates self-control (Green et al., 1994a, Green et al., 1994b).

In a typical delay discounting task, participants are asked to make a series of choices between large rewards (e.g. $1000) delayed by various delay intervals (e.g. 6 h to 25 years) and smaller immediate rewards (e.g. $1–$999). At each delay, the magnitude of the small immediate reward is adjusted until the small immediate and large delayed rewards are subjectively equivalent in value, referred to as the indifference point. The rate at which the value of a reward is discounted as a function of the delay to its delivery is best described by a hyperbolic equation developed by Mazur (1987):vd=V(1+kd),where vd is the present subjective value of a delayed commodity, V is the value of the commodity, d is the delay to delivery of the commodity, and k is an empirically derived constant proportional to the degree of delay discounting. Higher k values indicate steeper discounting functions. The hyperbolic discounting function indicates that the proportional decrease in subjective reward value following a delay is larger at proximal delay intervals than more remote delay intervals (as opposed to the constant proportional decrease in reward value across proximal and distal delays indicated by the exponential discounting function of conventional economic theory).

Two studies indicate that pathological gamblers more steeply discount monetary rewards than controls (Petry and Casarella, 1999, Petry, 2001a). The primary purpose of the present study was to extend those findings by examining the relationship between severity of pathological gambling and impulsivity using a delay discounting of monetary rewards procedure. Severity of pathological gambling was defined using the SOGS (Lesieur and Blume, 1987). The SOGS is the most widely used instrument for assessing problem gambling (e.g. Shaffer et al., 1999, National Research Council, 1999). Scores range from 0 to 20, with scores ≥5 indicative of probable pathological gambling (Lesieur and Blume, 1987). Typically, a moderate to high correlation exists between SOGS scores, DSM criteria, and other indices of gambling frequency and severity (Stinchfield, 2002, Petry, 2003). Greater SOGS scores indicate greater gambling problem severity, making the SOGS a useful tool with which to examine measures expected to vary as a function of gambling problem severity.

Delay discounting rates can differ depending on gender and age (Green et al., 1994b, Kirby and Marakovic, 1996), years of education (Petry et al., 2002), substance abuse history (Madden et al., 1997, Vuchinich and Simpson, 1998, Kirby et al., 1999, Madden et al., 1999, Petry, 2001b, Johnson and Bickel, 2002, Petry, 2002), and cigarette smoking status (Bickel et al., 1999, Mitchell, 1999). Furthermore, delay discounting rates have been found to correlate with personality assessments of impulsivity (Madden et al., 1997, Kirby et al., 1999, Richards et al., 1999, Petry, 2002). The relationship between these variables and delay discounting rates was therefore examined. We hypothesized a positive association between Eysenck impulsivity scale scores and delay discounting rates. We also expected that gambling problem severity would be associated with discounting rates, and we determined whether gambling problem severity predicted discounting rates after controlling for the other variables tested.

Section snippets

Participants

Participants were pathological gamblers (n=62) enrolled in a treatment trial evaluating cognitive-behavioral therapy for pathological gambling (see Petry and Roll, 2001 for a further description). Individuals older than 18 years with a DSM-IV (American Psychiatric Association, 1994) diagnosis of pathological gambling and one or more gambling days in the past month were eligible for participation. Individuals who were non-English speaking or had uncontrolled major psychosis or active suicidality

Delay discounting rates

Median indifference points (and interquartile range) from the delay discounting task were $990 (900–999), $960 (850–999), $900 (700–999), $700 (500–800), $500 (225–650), $250 (100–475), and $100 (20–225) at delays of 6 h, 1 day, 1 week, 2 months, 1 year, 5 years, and 25 years, respectively. Thus, $1000 was discounted to 50% of its value when delivery was delayed by 1 year. The median k value that resulted from fitting the indifference points to Eq. (1) was 0.079 (0.028–0.425) with an associated

Discussion

The purpose of the present study was to elucidate the nature of the relationship between severity of pathological gambling and impulsivity. The delay discounting task was used to assess a specific behavioral definition of impulsivity, in relation to choices about money. We found that impulsive choices were predicted by severity of gambling problems. Further, severity of gambling problems predicted degree of impulsivity on this task above and beyond the variance accounted for by impulsivity as

Acknowledgements

Yola Ammerman, JoAnne Boccuzzi, Heather Gay, Jaime Kelley, and Cheryl Molina assisted in data collection and management, and Susan Garthwait with manuscript preparation. The staffs at the Compulsive Gambling Treatment Program, its Bettor Choice Programs, and the Connecticut Council on Problem Gambling are thanked for their participation in this project. This research was supported in part by NIH grants R01-MH60417, R01-MH60417-Supp, R01-DA13444, R29-DA12056, P50-AA03510, P50-DA09241, and the

References (50)

  • C. Blanco et al.

    Pathological gambling and platelet MAO activity: a psychobiological study

    Am. J. Psychiatry

    (1996)
  • A. Blaszczynski

    Pathological gambling and obsessive-compulsive spectrum disorders

    Psychol. Rep.

    (1999)
  • A. Blaszczynski et al.

    Impulsivity in pathological gambling: the antisocial impulsivist

    Addiction

    (1997)
  • G.B. Bovasso et al.

    Predictive validity of the Addiction Severity Index’s composite scores in the assessment of 2-year outcomes in a methadone maintenance population

    Psychol. Addict. Behav.

    (2001)
  • P.L. Carlton et al.

    Factors determining the severity of pathological gambling in males

    J. Gambling Stud.

    (1994)
  • B. Castellani et al.

    A comparison of pathological gamblers to alcoholics and cocaine misusers on impulsivity, sensation seeking, and craving

    Int. J. Addict.

    (1995)
  • S.B.G. Eysenck et al.

    Impulsiveness and venturesomeness: their position in a dimensional system of personality description

    Psychol. Rep.

    (1978)
  • W. Fals-Stewart et al.

    The timeline followback reports of psychoactive substance use by drug-abusing patients: psychometric properties

    J. Consult. Clin. Psychol.

    (2000)
  • L. Green et al.

    Temporal discounting and preference reversals in choice between delayed outcomes

    Psychon. Bull. Rev.

    (1994)
  • L. Green et al.

    Discounting of delayed rewards: a life span comparison

    Psychol. Sci.

    (1994)
  • J.M. Hinson et al.

    Impulsive decision making and working memory

    J. Exp. Psychol. Learn. Mem. Cogn.

    (2003)
  • M.W. Johnson et al.

    Within-subject comparison of real and hypothetical money rewards in delay discounting

    J. Exp. Anal. Behav.

    (2002)
  • K.N. Kirby et al.

    Delay-discounting probabilistic rewards: rates decrease as amounts increase

    Psychon. Bull. Rev.

    (1996)
  • K.N. Kirby et al.

    Heroin addicts have higher discount rates for delayed rewards than non-drug using controls

    J. Exp. Psychol. Gen.

    (1999)
  • T.R. Kosten et al.

    Concurrent validity of the Addiction Severity Index

    J. Nerv. Ment. Dis.

    (1983)
  • Cited by (0)

    View full text