- Email: [email protected]
Pathological gambling and the five-factor model of personality
Personality and Individual Differences 43 (2007) 873–880 www.elsevier.com/locate/paid
Pathological gambling and the five-factor model of personality I R. Michael Bagby a,*, David D. Vachon b, Eric L. Bulmash c, Tony Toneatto a, Lena C. Quilty d, Paul T. Costa e a b
Centre for Addiction and Mental Health, Department of Psychiatry, University of Toronto, Canada Centre for Addiction and Mental Health, Department of Criminology, University of Toronto, Canada c Department of Psychology, Queen’s University, Canada d Centre for Addiction and Mental Health, Canada e National Institute of Aging, Canada
Received 12 September 2006; received in revised form 18 January 2007; accepted 2 February 2007 Available online 18 April 2007
Abstract The goal of this investigation was to examine the personality differences between non-treatment seeking pathological gamblers (PGs) and non-pathological gamblers (NPGs) using the domain and facet traits of the five-factor model of personality (FFM), as measured by the NEO PI-R. Compared to NPGs, PGs scored significantly higher on the neuroticism domain and significantly lower on the conscientiousness domain. Significant differences between PGs and NPGs also emerged for three of four FFM facet traits associated with impulsivity, with PGs scoring higher on impulsiveness and lower on self-discipline and deliberation facets. Both PGs and NPGs had equally high scores (relative to the norm) on excitement-seeking, the fourth facet associated with impulsivity, suggesting that excitement-seeking characterizes gambling behavior rather than pathological gambling. The present findings suggest that the overall personality profile of the PG is one that combines high impulsivity with emotional vulnerability. Importantly, the results also suggest that excitement-seeking, a personality construct akin to sensation-seeking, may not be a specific marker of PG but rather a characteristic common to all those who gamble. 2007 Published by Elsevier Ltd. q
This study was supported by a grant obtained from the Ontario Problem Gambling Research Centre. Corresponding author. Address: Centre for Addiction and Mental Health, College Street Site, 250 College Street, Suite 647A, Toronto, Ontario, Canada M5T 1R8. Tel.: +1 416 535 8501x6939; fax: +1 416 260 4125. E-mail address: [email protected] (R.M. Bagby). *
0191-8869/$ - see front matter 2007 Published by Elsevier Ltd. doi:10.1016/j.paid.2007.02.011
874
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
Keywords: Pathological gambling; Five-factor model of personality; Impulsivity
1. Introduction In the past several years, there has been an increase in gambling accessibility in North America, and the proliferation of internet gaming sites, video lottery terminals, and legalized gambling venues has afforded more opportunity to gamble and increased the likelihood for some to develop pathological gambling (PG) (Petry, 2005). PG is categorized as an impulse-control disorder in the 4th edition of Diagnostic and Statistical Manual for Mental Disorders (DSM-IV; APA, 1994) and is defined as a pattern of ‘‘maladaptive gambling behavior that disrupts personal, family, or vocational pursuits’’ (p. 615). Recent estimates indicate that 70–90% of North Americans have engaged in some form of gambling activity (Raylu & Oei, 2002), yet epidemiological studies indicate that only 1–2% of the general population meet criteria for PG (Petry, Stinson, & Grant, 2005), suggesting that the majority of individuals who gamble do not develop this disorder. Individual differences in personality may play an important role in explaining risk for the development of PG. A variety of dimensional personality models have been utilized to study personality in PG. Steel and Blaszczynski (1996) employed Eysenck and Eysenck’s three-factor model of personality (Eysenck & Eysenck, 1975), which incorporates the personality domains of psychoticism, extraversion, and neuroticism (PEN). Combining the PEN model with other psychological inventories, they identified four constructs – psychological distress, sensation-seeking, criminal liveliness, and impulsiveness – that explained 62% of the variance in gambling behavior. Nower, Derevensky, and Gupta (2004) similarly reported PEN impulsivity and intensity-seeking as highly predictive of PG behaviour in youth. An ‘‘alternative’’ five-factor model, developed by Zuckerman, Kuhlman, Thornquist, and Kiers (1991) and composed of five personality factors – neuroticism-anxiety, activity, sociability, impulsive sensation-seeking, aggression-hostility – has been popular in PG research because of its focus on impulsivity and sensation-seeking. Investigations using this model have reported mixed results: Although Breen and Zuckerman (1999) reported that impulsive sensation-seeking predicted gamblers who ‘chased’ their losses, Langewisch and Frisch (1998) found no evidence that this model could predict severity of PG symptomatology. Cloninger, Svrakic, and Przybeck’s (1993) biopsychosocial seven-factor model, which differentiates between character (i.e., cooperativeness, self-directedness, and self-transcendence) and temperament (i.e., harm avoidance, novelty-seeking, reward dependence, and persistence) dimensions, has also been utilized. Gerdner and Svensson (2003), for example, employed this model and found no significant relationship between gambling behaviors and impulsivity, a lower-order facet trait located in the novelty-seeking temperament. More recently, Slutske, Caspi, Moffitt, and Poulton (2005) used a modified version of the multidimensional personality questionnaire (MPQ; Patrick, Curtin, & Tellegen, 2002) in a prospective investigation with a birth cohort of young adults to examine the relationship between personality and PG, and found that PG at 21 years of age was associated with higher levels of negative emotionality and lower levels of constraint measured at 18 years of age. Although each of these models offers a valuable and unique contribution to the framing of personality in relation to pathological gambling, inconsistencies among outcomes that use differently defined traits makes clear the need for a single, overarching model.
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
875
In the current investigation we examine differences in five-factor model (FFM) personality traits between non-treatment seeking PGs and a sample of individuals who report engaging in gambling activities but do not meet criteria for PG, that is, non-pathological gamblers (NPGs). The gambling literature uses various terms (e.g., pathological, problem, disordered, at-risk, and compulsive gambling) and definitional criteria to operationalize problematic gambling behavior. In this investigation, we used the DSM-IV criteria as five or more of the 10 criteria. As PG is conceptualized as an impulse-control disorder, the use of the FFM, as measured by the revised NEO personality inventory (NEO PI-R; Costa & McCrae, 1992), seems particularly relevant for the examination of personality traits associated with PG. Four of the 30 facet traits measured by the NEO PI-R – deliberation, excitement-seeking, impulsiveness, self-discipline – account for two-thirds of the variance in the nine most commonly used measures of impulsivity (Whiteside & Lynam, 2001), making the NEO PI-R an ideal platform to assess the full range of impulsivity traits in PG. Moreover, each of the FFM domain traits demonstrates a unique and substantial heritability coefficient (Jang, McCrae, Angleitner, Riemann, & Livesley, 1998) and the FFM personality structure has been extracted in clinical samples (Bagby et al., 1999) and across a variety of cultures, languages, and countries (McCrae & Costa, 1995). We hypothesized that the PGs would have higher scores on the impulsivity and excitementseeking facet scales and lower scores on the self-discipline and deliberation facet scales compared to the NPGs. As neuroticism is associated with and considered a vulnerability factor for psychopathology in general (Malouff, Thorsteinsson, & Shutte, 2005), we also hypothesized that PGs would have higher scores on the neuroticism domain scale compared to NPGs.
2. Method 2.1. Participants Participants (N = 292) from two separate investigations conducted at the Centre for Addiction and Mental Health were solicited via advertisement in local newspapers for a study on gambling behavior with the general requirement that they had gambled some time in their lives, including the past year.1 PG was diagnosed according to the DSM-IV criteria and interviews were conducted by research assistants trained in structured diagnostic assessments.
3. Measures Personality traits were assessed using the five domain and 30 facet personality traits of the fivefactor model of personality (FFM), as measured by the revised NEO personality inventory (NEO PI-R; Costa & McCrae, 1992). 1
Other data from some of the participants (n = 204) in the current report have been reported in an earlier investigation (Bagby, Vachon, Bulmash, & Quilty, in press). A 2 (investigation) · 2 (PG/NPG) comparison revealed no significant differences for any of the domain or facet traits.
876
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
3.1. Procedure and participant group assignment Participants completed the NEO PI-R and other tests as part of other, larger research protocols. Those included in the PG group (n = 106; 56 men; 50 women) had to have met the DSMIV diagnostic criteria for pathological gambling of five or more symptoms in the last 12 months (current) or in the period prior to the past 12 months (lifetime).2 The NPG group (n = 177; 96 women; 81 men) was composed of participants who did not meet DSM-IV criteria for either current or lifetime PG.
4. Results Means and standard deviations for the NEO PI-R domain and facet trait scores for the PGs and NPGs groups, as well as the results from the statistical tests for between-group differences, are displayed in Table 1. There was no significant difference in the distribution of men and women across the PG and NPG groups; PGs were significantly older than the NPGs, mean age = 43.11 (SD = 12.79) vs. 38.75 (SD = 12.88), respectively, t(291) = 2.83, p < 0.01. Between-group differences were assessed using MANCOVA, with age serving as the covariate. Significant differences were found for three of four impulsivity traits, with PGs scoring higher on impulsiveness and lower on deliberation and self-discipline compared to the NPGs; no significant differences emerged for the excitement-seeking facet. Compared to NPGs, PGs also scored significantly higher on the neuroticism domain and significantly lower on the conscientiousness domain, with moderate effect sizes (i.e., Cohen’s d = 0.50) associated with both of these differences. For the facets within neuroticism excluding impulsivity, PGs scored higher on the facet traits of depression, self-consciousness, and vulnerability. For the facets within the conscientiousness domain excluding selfdiscipline and deliberation, PGs scored higher on the competence and dutifulness facet traits. The facet trait Trust, which is located in the Agreeableness domain, was also significant, although the overall effect for this domain was not significant.
5. Discussion Three of four impulsivity-related facet traits – impulsiveness, self-discipline, and deliberation – distinguished PGs from non-pathological gamblers. The PGs and NPGs were indistinguishable on the facet trait of excitement-seeking, however, and both groups demonstrated elevated scores relative to the normative sample on this facet trait (about 0.5 SD higher). High scorers on excitement-seeking typically seek thrill and are drawn to colorful, noisy environments. Although this description seems intuitively linked with gambling problems, the current findings suggest that excitement-seeking is associated with gambling behavior in general, rather than with pathological gambling specifically. This outcome and interpretation is generally consistent with a past report
2 There were no significant differences among those with lifetime but no current PG (n = 67), current but no lifetime PG (n = 10), and current and lifetime PG (n = 97) for any of the personality domain and facet traits.
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
877
Table 1 Means, standard deviations, test of significance, and effect size differences for the NEO PI-R scores of pathological and non-pathological gamblers Domains and facets
PGs (n = 106)
NPGs (n = 177)
F(1,272)
d
M
SD
M
SD
Neuroticism N1 Anxiety N2 Angry hostility N3 Depression N4 Self-consciousness N5 Impulsiveness N6 Vulnerability
60.71 55.57 56.97 61.21 55.92 58.90 59.37
10.89 10.37 10.72 11.67 10.73 10.05 12.62
54.87 52.80 54.28 55.10 51.81 54.15 53.36
11.61 10.23 11.16 12.10 10.16 10.19 12.72
16.77* 4.95 4.07 14.95* 10.33* 15.11* 14.09*
0.54 0.27 0.25 0.52 0.38 0.47 0.48
Extraversion E1 Warmth E2 Gregariousness E3 Assertiveness E4 Activity E5 Excitement-seeking E6 Positive emotions
49.69 45.58 49.76 50.45 47.46 56.37 47.63
11.79 12.92 11.45 10.34 9.80 9.37 12.12
52.02 48.43 51.87 50.67 49.33 56.47 50.44
10.23 10.88 10.93 10.00 10.29 10.59 10.32
1.52 2.52 1.12 0.00 1.33 0.26 3.34
0.20 0.22 0.18 0.02 0.19 0.01 0.23
Openness O1 Fantasy O2 Aesthetics O3 Feelings O4 Actions O5 Ideas O6 Values
50.02 53.83 48.73 51.34 45.23 50.22 49.87
11.11 10.18 10.75 10.87 11.39 10.07 10.34
51.08 53.01 49.31 51.27 46.48 51.92 51.54
11.49 9.64 11.30 10.45 12.28 11.11 10.52
0.40 0.32 0.13 0.03 0.29 1.38 1.46
0.10 0.08 0.05 0.1 0.11 0.17 0.16
Agreeableness A1 Trust A2 Straightforwardness A3 Altruism A4 Compliance A5 Modesty A6 Tender-mindedness
42.17 41.13 43.06 45.18 41.47 47.91 52.63
12.51 12.62 11.83 13.13 11.68 12.00 11.39
45.43 45.19 45.91 48.83 43.89 48.03 51.33
12.76 11.33 11.69 12.40 12.15 11.70 12.68
4.95 9.05* 3.93 4.62 3.77 0.09 0.48
0.26 0.32 0.24 0.28 0.21 0.01 0.11
Conscientiousness C1 Competence C2 Order C3 Dutifulness C4 Achievement striving C5 Self-discipline C6 Deliberation
38.57 40.85 45.17 40.04 41.80 39.04 42.44
11.15 11.94 11.28 11.54 11.56 11.06 12.03
45.01 46.54 48.28 45.69 45.79 44.14 47.63
11.92 11.69 10.90 11.54 12.16 12.42 11.65
20.89* 15.74* 5.72 19.47* 5.94 12.79* 12.72*
0.58 0.48 0.28 0.49 0.35 0.46 0.43
* Note: After Bonferroni correction: p < 0.01 for domains (0.05/5 domains), p < 0.008 for all facets (0.05/6 facets per domain). All means and standard deviations are expressed as T-scores. PGs = pathological gamblers, NPG = nonpathological gamblers.
indicating that sensation seeking, a variable akin to excitement-seeking, is associated with gambling but not severity of gambling problems (Langewisch & Frisch, 1998).
878
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
Impulsivity and impulse disorders may be partially explained by genetic abnormalities in riskreward pathways. Comings, Rosenthal, Lesieur, and Rugle (1996) demonstrated that PG is associated with a reduction in dopamine D2 receptor density, which has been implicated in other impulse-related disorders (Blum et al., 1996). Twin studies have indicated that approximately 50% of the variance for risk of developing PG can be accounted for by genetic factors (Slutske et al., 2001). Taken together, it is possible that the inability to exercise control over gambling may be linked to neurobiological correlates that lead to an increased focus on reward, a lesser response to aversive consequences, and impairments in altering behavior based on risk-reward learning. In the current investigation PGs also scored significantly higher on the neuroticism domain and three of its facet traits excluding the impulsiveness facet. Neuroticism represents a predisposition to develop psychopathology and to experience a wide range of negative affects (Costa & McCrae, 1992). Similarly, Roy, Custer, and Lorenz (1989) reported a strong association with neuroticism, as measured by the Eysenck personality questionnaire (EPQ; Eysenck & Eysenck, 1975), and Slutske et al. (2005) also reported high levels of negative emotionality, as measured by a modified version of the multidimensional personality questionnaire (MPQ; Patrick et al., 2002), in those with disordered gambling behaviour compared to a sample of non-gambling controls. The conscientiousness domain is the trait with the strongest conceptual links with impulse-control as it captures the capacity to resist impulses, manage desires, and apply guiding principles to actively control behavior (Costa & McCrae, 1992). Thus, it was not surprising that PGs scored lower than the NPGs on four of six facet traits of this domain. This result is also consistent with previous research in which PGs score significantly higher than normal controls on the psychoticism dimension of the EPQ (Roy et al., 1989), a personality factor inversely correlated with the conscientiousness personality domain (Costa & Widiger, 2001). Slutske et al. (2005) also reported problem gambling to be significantly and negatively associated with the constraint personality dimension of the MPQ. In sum, results from the current study suggest the overall personality profile of the pathological gambler is one that combines high impulsivity with emotional vulnerability complemented by a high level of excitement-seeking common to PGs and NPGs. Although speculative, one possible interpretation for the development of pathological gambling is that it results from maladaptive efforts to regulate affect or dampen the effects of high neuroticism. After gambling behavior has been behaviorally conditioned and losses begin to accrue, high impulsivity may render the PG unable to modify, control, or stop their gambling behavior. Alternatively, the personality traits of the pathological gambler may be related solely to dysfunctional risk-reward and biochemical pathways that cause mood disturbance, elevated impulsivity, and an inability to regulate affect and/or behavior. Finally, PGs might not represent one homogenous population, but rather qualitatively distinct subtypes that are influenced by different emotional and biological factors but exhibit similar phenomenological features (Blaszczynski & Nower, 2002). Future investigations are needed to begin to examine and test these competing explanations. We believe this study has several methodological strengths, including the use of non-pathological gamblers (instead of non-gambling controls) as a comparison group and non-treatment seeking PGs (instead of treatment seeking PGs). NPGs resemble pathological gamblers in their attraction to gambling and engagement in gambling behavior, which decreases the probability of conflating gambling behavior with personality and addresses directly potential differences in
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
879
those who gamble with no gambling pathology from those who gamble with maladaptive and debilitative patterns of gambling behavior. Treatment seeking PGs comprise less than 2% of the PGs found in community samples and are therefore not representative of the population of pathological gamblers (Wallisch, 1996); the use of non-treatment seeking PGs in this study enhances the generalization of our results. There were also limitations associated with the current investigation. Due to the cross-sectional nature of this study, the temporal relationships between personality and PG could not be assessed. Whereas the use of non-treatment seeking PGs increases the overall generalizability of the results, the personality traits that characterize the PGs in this study may not be applicable to treatment seeking PGs. Several potential confounding factors related to social desirability, demographic differences (e.g., socio-economic status, education, ethnicity), memory errors, and context effects were not controlled for. Finally, we did not calculate inter-rater reliability estimates across the interviewers who conferred the diagnosis of PG, as different raters were used in the two studies, the participants from which were combined to maximize statistical power. We believe it unlikely that systematic interviewer error influenced diagnosis, however, as the proportion of PGs to NPGs across the two study samples did not differ and the interviewers were well trained and bi-weekly consultation sessions were held in which diagnostic issues were discussed and cases reviewed. References American Psychiatric Association (APA) (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Press. Bagby, R. M., Costa, P. T., McCrae, R. R., Livesley, W. J., Kennedy, S. H., Levitan, R. D., et al. (1999). Replicating the five factor model of personality in a psychiatric sample. Personality and Individual Differences, 27(6), 1135–1139. Bagby, R. M., Vachon, D. D., Bulmash, E., & Quilty, L. C. (in press). Pathological gambling and personality disorders: a review and re-examination of prevalence rates. Journal of Personality Disorders. Blaszczynski, A., & Nower, L. (2002). A pathways model of problem and pathological gambling. Addiction, 97, 487–499. Blum, K., Dherid, P. J., Wood, R. C., Braverman, E. R., Chen, T. J., Cull, J. G., et al. (1996). The D2 dopamine receptor gene as a determinant of reward deficiency syndrome. Journal of the Royal Society of Medicine, 89, 396–400. Breen, R. B., & Zuckerman, M. (1999). Chasing in gambling behaviour: personality and cognitive determinants. Personality and Individual Differences, 27, 1097–1111. Cloninger, C. R., Svrakic, D. M., & Przybeck, T. R. (1993). A psychobiological model of temperament and character. Archives of General Psychiatry, 50, 975–990. Comings, D. E., Rosenthal, R. J., Lesieur, H. R., & Rugle, L. (1996). A study of the dopamine D2 receptor gene in pathological gambling. Pharmacogenetics, 6, 223–234. Costa, P. T., & McCrae, R. R. (1992). Revised NEO personality inventory (NEO-PI-R) and NEO five-factor inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources. Costa, P. T., & Widiger, T. A. (2001). Personality disorders and the five factor model of personality. Washington, DC: American Psychological Association. Eysenck, H. J., & Eysenck, S. B. G. (1975). Manual of the EPQ (Eysenck personality questionnaire). San Diego, CA: Educational and Industrial Testing Service. Gerdner, R., & Svensson, K. (2003). Predictors of gambling problems among male adolescents. International Journal of Social Welfare, 12, 182–192. Jang, K. L., McCrae, R. R., Angleitner, A., Riemann, R., & Livesley, J. W. (1998). Heritability of facet-level traits in a cross-cultural twin sample: support for a hierarchical model of personality. Journal of Personality and Social Psychology, 74, 1556–1565.
880
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
Langewisch, M. W. J., & Frisch, G. R. (1998). Gambling behaviour and pathology in relation to impulsivity, sensation seeking, and risky behaviour in male college students. Journal of Gambling Studies, 14, 245–262. Malouff, J. M., Thorsteinsson, E. B., & Shutte, N. S. (2005). The relationship between the five-factor model of personality and symptoms of clinical disorders: a meta-analysis. Journal of Psychopathology and Behavioral Assessment, 27, 101–114. McCrae, R. R., & Costa, P. T. Jr., (1995). Personality trait structure as a human universal. American Psychologist, 52, 509–516. Nower, L., Derevensky, J. L., & Gupta, R. (2004). The relationship of impulsivity, sensation seeking, coping, and substance use in youth gamblers. Psychology of Addictive Behaviours, 18, 49–55. Patrick, C. J., Curtin, J. J., & Tellegen, A. (2002). Development and validation of a brief form of the multidimensional personality questionnaire. Psychological Assessment, 14, 150–163. Petry, N. M. (2005). Pathological gambling: etiology, comorbidity, and treatment. Washington, DC: American Psychological Association. Petry, N. M., Stinson, F. S., & Grant, B. F. (2005). Comorbidity of DSM-IV pathological gambling and other psychiatric disorders: results from the national epidemiologic survey on alcohol and related conditions. Journal of Clinical Psychiatry, 66, 564–574. Raylu, N., & Oei, T. P. S. (2002). Pathological gambling: a comprehensive review. Clinical Psychological Review, 22, 1009–1061. Roy, A., Custer, R., & Lorenz, V. (1989). Personality factors in pathological gambling. Acta Psychiatrica Scandinavica, 80, 37–39. Slutske, W. S., Caspi, A., Moffitt, T. E., & Poulton, R. (2005). Personality and problem gambling: a prospective study of a birth cohort of young adults. Archives of General Psychiatry, 62, 769–775. Slutske, W. S., True, W. R., Goldberg, J., Eisen, S., Xian, H., Lyons, M. J., et al. (2001). A twin study of the association between pathological gambling and antisocial personality disorder. Journal of Abnormal Psychology, 110, 297–308. Steel, Z., & Blaszczynski, A. (1996). The factorial structure of pathological gambling. Journal of Gambling Studies, 12, 3–20. Wallisch, L. S. (1996). Gambling in Texas: 1995 surveys of adult and adolescent gambling behavior. Austin, TX: Texas Commission on Alcohol and Drug Abuse. Whiteside, S. P., & Lynam, D. R. (2001). The five factor model and impulsivity: using a structural model of personality to understand impulsivity. Personality and Individual Differences, 20, 669–689. Zuckerman, M., Kuhlman, D. M., Thornquist, M., & Kiers, H. (1991). Five (or three) robust questionnaire scale factors of personality without culture. Personality and Individual Differences, 12, 929–941.
Pathological gambling and the five-factor model of personality I R. Michael Bagby a,*, David D. Vachon b, Eric L. Bulmash c, Tony Toneatto a, Lena C. Quilty d, Paul T. Costa e a b
Centre for Addiction and Mental Health, Department of Psychiatry, University of Toronto, Canada Centre for Addiction and Mental Health, Department of Criminology, University of Toronto, Canada c Department of Psychology, Queen’s University, Canada d Centre for Addiction and Mental Health, Canada e National Institute of Aging, Canada
Received 12 September 2006; received in revised form 18 January 2007; accepted 2 February 2007 Available online 18 April 2007
Abstract The goal of this investigation was to examine the personality differences between non-treatment seeking pathological gamblers (PGs) and non-pathological gamblers (NPGs) using the domain and facet traits of the five-factor model of personality (FFM), as measured by the NEO PI-R. Compared to NPGs, PGs scored significantly higher on the neuroticism domain and significantly lower on the conscientiousness domain. Significant differences between PGs and NPGs also emerged for three of four FFM facet traits associated with impulsivity, with PGs scoring higher on impulsiveness and lower on self-discipline and deliberation facets. Both PGs and NPGs had equally high scores (relative to the norm) on excitement-seeking, the fourth facet associated with impulsivity, suggesting that excitement-seeking characterizes gambling behavior rather than pathological gambling. The present findings suggest that the overall personality profile of the PG is one that combines high impulsivity with emotional vulnerability. Importantly, the results also suggest that excitement-seeking, a personality construct akin to sensation-seeking, may not be a specific marker of PG but rather a characteristic common to all those who gamble. 2007 Published by Elsevier Ltd. q
This study was supported by a grant obtained from the Ontario Problem Gambling Research Centre. Corresponding author. Address: Centre for Addiction and Mental Health, College Street Site, 250 College Street, Suite 647A, Toronto, Ontario, Canada M5T 1R8. Tel.: +1 416 535 8501x6939; fax: +1 416 260 4125. E-mail address: [email protected] (R.M. Bagby). *
0191-8869/$ - see front matter 2007 Published by Elsevier Ltd. doi:10.1016/j.paid.2007.02.011
874
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
Keywords: Pathological gambling; Five-factor model of personality; Impulsivity
1. Introduction In the past several years, there has been an increase in gambling accessibility in North America, and the proliferation of internet gaming sites, video lottery terminals, and legalized gambling venues has afforded more opportunity to gamble and increased the likelihood for some to develop pathological gambling (PG) (Petry, 2005). PG is categorized as an impulse-control disorder in the 4th edition of Diagnostic and Statistical Manual for Mental Disorders (DSM-IV; APA, 1994) and is defined as a pattern of ‘‘maladaptive gambling behavior that disrupts personal, family, or vocational pursuits’’ (p. 615). Recent estimates indicate that 70–90% of North Americans have engaged in some form of gambling activity (Raylu & Oei, 2002), yet epidemiological studies indicate that only 1–2% of the general population meet criteria for PG (Petry, Stinson, & Grant, 2005), suggesting that the majority of individuals who gamble do not develop this disorder. Individual differences in personality may play an important role in explaining risk for the development of PG. A variety of dimensional personality models have been utilized to study personality in PG. Steel and Blaszczynski (1996) employed Eysenck and Eysenck’s three-factor model of personality (Eysenck & Eysenck, 1975), which incorporates the personality domains of psychoticism, extraversion, and neuroticism (PEN). Combining the PEN model with other psychological inventories, they identified four constructs – psychological distress, sensation-seeking, criminal liveliness, and impulsiveness – that explained 62% of the variance in gambling behavior. Nower, Derevensky, and Gupta (2004) similarly reported PEN impulsivity and intensity-seeking as highly predictive of PG behaviour in youth. An ‘‘alternative’’ five-factor model, developed by Zuckerman, Kuhlman, Thornquist, and Kiers (1991) and composed of five personality factors – neuroticism-anxiety, activity, sociability, impulsive sensation-seeking, aggression-hostility – has been popular in PG research because of its focus on impulsivity and sensation-seeking. Investigations using this model have reported mixed results: Although Breen and Zuckerman (1999) reported that impulsive sensation-seeking predicted gamblers who ‘chased’ their losses, Langewisch and Frisch (1998) found no evidence that this model could predict severity of PG symptomatology. Cloninger, Svrakic, and Przybeck’s (1993) biopsychosocial seven-factor model, which differentiates between character (i.e., cooperativeness, self-directedness, and self-transcendence) and temperament (i.e., harm avoidance, novelty-seeking, reward dependence, and persistence) dimensions, has also been utilized. Gerdner and Svensson (2003), for example, employed this model and found no significant relationship between gambling behaviors and impulsivity, a lower-order facet trait located in the novelty-seeking temperament. More recently, Slutske, Caspi, Moffitt, and Poulton (2005) used a modified version of the multidimensional personality questionnaire (MPQ; Patrick, Curtin, & Tellegen, 2002) in a prospective investigation with a birth cohort of young adults to examine the relationship between personality and PG, and found that PG at 21 years of age was associated with higher levels of negative emotionality and lower levels of constraint measured at 18 years of age. Although each of these models offers a valuable and unique contribution to the framing of personality in relation to pathological gambling, inconsistencies among outcomes that use differently defined traits makes clear the need for a single, overarching model.
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
875
In the current investigation we examine differences in five-factor model (FFM) personality traits between non-treatment seeking PGs and a sample of individuals who report engaging in gambling activities but do not meet criteria for PG, that is, non-pathological gamblers (NPGs). The gambling literature uses various terms (e.g., pathological, problem, disordered, at-risk, and compulsive gambling) and definitional criteria to operationalize problematic gambling behavior. In this investigation, we used the DSM-IV criteria as five or more of the 10 criteria. As PG is conceptualized as an impulse-control disorder, the use of the FFM, as measured by the revised NEO personality inventory (NEO PI-R; Costa & McCrae, 1992), seems particularly relevant for the examination of personality traits associated with PG. Four of the 30 facet traits measured by the NEO PI-R – deliberation, excitement-seeking, impulsiveness, self-discipline – account for two-thirds of the variance in the nine most commonly used measures of impulsivity (Whiteside & Lynam, 2001), making the NEO PI-R an ideal platform to assess the full range of impulsivity traits in PG. Moreover, each of the FFM domain traits demonstrates a unique and substantial heritability coefficient (Jang, McCrae, Angleitner, Riemann, & Livesley, 1998) and the FFM personality structure has been extracted in clinical samples (Bagby et al., 1999) and across a variety of cultures, languages, and countries (McCrae & Costa, 1995). We hypothesized that the PGs would have higher scores on the impulsivity and excitementseeking facet scales and lower scores on the self-discipline and deliberation facet scales compared to the NPGs. As neuroticism is associated with and considered a vulnerability factor for psychopathology in general (Malouff, Thorsteinsson, & Shutte, 2005), we also hypothesized that PGs would have higher scores on the neuroticism domain scale compared to NPGs.
2. Method 2.1. Participants Participants (N = 292) from two separate investigations conducted at the Centre for Addiction and Mental Health were solicited via advertisement in local newspapers for a study on gambling behavior with the general requirement that they had gambled some time in their lives, including the past year.1 PG was diagnosed according to the DSM-IV criteria and interviews were conducted by research assistants trained in structured diagnostic assessments.
3. Measures Personality traits were assessed using the five domain and 30 facet personality traits of the fivefactor model of personality (FFM), as measured by the revised NEO personality inventory (NEO PI-R; Costa & McCrae, 1992). 1
Other data from some of the participants (n = 204) in the current report have been reported in an earlier investigation (Bagby, Vachon, Bulmash, & Quilty, in press). A 2 (investigation) · 2 (PG/NPG) comparison revealed no significant differences for any of the domain or facet traits.
876
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
3.1. Procedure and participant group assignment Participants completed the NEO PI-R and other tests as part of other, larger research protocols. Those included in the PG group (n = 106; 56 men; 50 women) had to have met the DSMIV diagnostic criteria for pathological gambling of five or more symptoms in the last 12 months (current) or in the period prior to the past 12 months (lifetime).2 The NPG group (n = 177; 96 women; 81 men) was composed of participants who did not meet DSM-IV criteria for either current or lifetime PG.
4. Results Means and standard deviations for the NEO PI-R domain and facet trait scores for the PGs and NPGs groups, as well as the results from the statistical tests for between-group differences, are displayed in Table 1. There was no significant difference in the distribution of men and women across the PG and NPG groups; PGs were significantly older than the NPGs, mean age = 43.11 (SD = 12.79) vs. 38.75 (SD = 12.88), respectively, t(291) = 2.83, p < 0.01. Between-group differences were assessed using MANCOVA, with age serving as the covariate. Significant differences were found for three of four impulsivity traits, with PGs scoring higher on impulsiveness and lower on deliberation and self-discipline compared to the NPGs; no significant differences emerged for the excitement-seeking facet. Compared to NPGs, PGs also scored significantly higher on the neuroticism domain and significantly lower on the conscientiousness domain, with moderate effect sizes (i.e., Cohen’s d = 0.50) associated with both of these differences. For the facets within neuroticism excluding impulsivity, PGs scored higher on the facet traits of depression, self-consciousness, and vulnerability. For the facets within the conscientiousness domain excluding selfdiscipline and deliberation, PGs scored higher on the competence and dutifulness facet traits. The facet trait Trust, which is located in the Agreeableness domain, was also significant, although the overall effect for this domain was not significant.
5. Discussion Three of four impulsivity-related facet traits – impulsiveness, self-discipline, and deliberation – distinguished PGs from non-pathological gamblers. The PGs and NPGs were indistinguishable on the facet trait of excitement-seeking, however, and both groups demonstrated elevated scores relative to the normative sample on this facet trait (about 0.5 SD higher). High scorers on excitement-seeking typically seek thrill and are drawn to colorful, noisy environments. Although this description seems intuitively linked with gambling problems, the current findings suggest that excitement-seeking is associated with gambling behavior in general, rather than with pathological gambling specifically. This outcome and interpretation is generally consistent with a past report
2 There were no significant differences among those with lifetime but no current PG (n = 67), current but no lifetime PG (n = 10), and current and lifetime PG (n = 97) for any of the personality domain and facet traits.
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
877
Table 1 Means, standard deviations, test of significance, and effect size differences for the NEO PI-R scores of pathological and non-pathological gamblers Domains and facets
PGs (n = 106)
NPGs (n = 177)
F(1,272)
d
M
SD
M
SD
Neuroticism N1 Anxiety N2 Angry hostility N3 Depression N4 Self-consciousness N5 Impulsiveness N6 Vulnerability
60.71 55.57 56.97 61.21 55.92 58.90 59.37
10.89 10.37 10.72 11.67 10.73 10.05 12.62
54.87 52.80 54.28 55.10 51.81 54.15 53.36
11.61 10.23 11.16 12.10 10.16 10.19 12.72
16.77* 4.95 4.07 14.95* 10.33* 15.11* 14.09*
0.54 0.27 0.25 0.52 0.38 0.47 0.48
Extraversion E1 Warmth E2 Gregariousness E3 Assertiveness E4 Activity E5 Excitement-seeking E6 Positive emotions
49.69 45.58 49.76 50.45 47.46 56.37 47.63
11.79 12.92 11.45 10.34 9.80 9.37 12.12
52.02 48.43 51.87 50.67 49.33 56.47 50.44
10.23 10.88 10.93 10.00 10.29 10.59 10.32
1.52 2.52 1.12 0.00 1.33 0.26 3.34
0.20 0.22 0.18 0.02 0.19 0.01 0.23
Openness O1 Fantasy O2 Aesthetics O3 Feelings O4 Actions O5 Ideas O6 Values
50.02 53.83 48.73 51.34 45.23 50.22 49.87
11.11 10.18 10.75 10.87 11.39 10.07 10.34
51.08 53.01 49.31 51.27 46.48 51.92 51.54
11.49 9.64 11.30 10.45 12.28 11.11 10.52
0.40 0.32 0.13 0.03 0.29 1.38 1.46
0.10 0.08 0.05 0.1 0.11 0.17 0.16
Agreeableness A1 Trust A2 Straightforwardness A3 Altruism A4 Compliance A5 Modesty A6 Tender-mindedness
42.17 41.13 43.06 45.18 41.47 47.91 52.63
12.51 12.62 11.83 13.13 11.68 12.00 11.39
45.43 45.19 45.91 48.83 43.89 48.03 51.33
12.76 11.33 11.69 12.40 12.15 11.70 12.68
4.95 9.05* 3.93 4.62 3.77 0.09 0.48
0.26 0.32 0.24 0.28 0.21 0.01 0.11
Conscientiousness C1 Competence C2 Order C3 Dutifulness C4 Achievement striving C5 Self-discipline C6 Deliberation
38.57 40.85 45.17 40.04 41.80 39.04 42.44
11.15 11.94 11.28 11.54 11.56 11.06 12.03
45.01 46.54 48.28 45.69 45.79 44.14 47.63
11.92 11.69 10.90 11.54 12.16 12.42 11.65
20.89* 15.74* 5.72 19.47* 5.94 12.79* 12.72*
0.58 0.48 0.28 0.49 0.35 0.46 0.43
* Note: After Bonferroni correction: p < 0.01 for domains (0.05/5 domains), p < 0.008 for all facets (0.05/6 facets per domain). All means and standard deviations are expressed as T-scores. PGs = pathological gamblers, NPG = nonpathological gamblers.
indicating that sensation seeking, a variable akin to excitement-seeking, is associated with gambling but not severity of gambling problems (Langewisch & Frisch, 1998).
878
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
Impulsivity and impulse disorders may be partially explained by genetic abnormalities in riskreward pathways. Comings, Rosenthal, Lesieur, and Rugle (1996) demonstrated that PG is associated with a reduction in dopamine D2 receptor density, which has been implicated in other impulse-related disorders (Blum et al., 1996). Twin studies have indicated that approximately 50% of the variance for risk of developing PG can be accounted for by genetic factors (Slutske et al., 2001). Taken together, it is possible that the inability to exercise control over gambling may be linked to neurobiological correlates that lead to an increased focus on reward, a lesser response to aversive consequences, and impairments in altering behavior based on risk-reward learning. In the current investigation PGs also scored significantly higher on the neuroticism domain and three of its facet traits excluding the impulsiveness facet. Neuroticism represents a predisposition to develop psychopathology and to experience a wide range of negative affects (Costa & McCrae, 1992). Similarly, Roy, Custer, and Lorenz (1989) reported a strong association with neuroticism, as measured by the Eysenck personality questionnaire (EPQ; Eysenck & Eysenck, 1975), and Slutske et al. (2005) also reported high levels of negative emotionality, as measured by a modified version of the multidimensional personality questionnaire (MPQ; Patrick et al., 2002), in those with disordered gambling behaviour compared to a sample of non-gambling controls. The conscientiousness domain is the trait with the strongest conceptual links with impulse-control as it captures the capacity to resist impulses, manage desires, and apply guiding principles to actively control behavior (Costa & McCrae, 1992). Thus, it was not surprising that PGs scored lower than the NPGs on four of six facet traits of this domain. This result is also consistent with previous research in which PGs score significantly higher than normal controls on the psychoticism dimension of the EPQ (Roy et al., 1989), a personality factor inversely correlated with the conscientiousness personality domain (Costa & Widiger, 2001). Slutske et al. (2005) also reported problem gambling to be significantly and negatively associated with the constraint personality dimension of the MPQ. In sum, results from the current study suggest the overall personality profile of the pathological gambler is one that combines high impulsivity with emotional vulnerability complemented by a high level of excitement-seeking common to PGs and NPGs. Although speculative, one possible interpretation for the development of pathological gambling is that it results from maladaptive efforts to regulate affect or dampen the effects of high neuroticism. After gambling behavior has been behaviorally conditioned and losses begin to accrue, high impulsivity may render the PG unable to modify, control, or stop their gambling behavior. Alternatively, the personality traits of the pathological gambler may be related solely to dysfunctional risk-reward and biochemical pathways that cause mood disturbance, elevated impulsivity, and an inability to regulate affect and/or behavior. Finally, PGs might not represent one homogenous population, but rather qualitatively distinct subtypes that are influenced by different emotional and biological factors but exhibit similar phenomenological features (Blaszczynski & Nower, 2002). Future investigations are needed to begin to examine and test these competing explanations. We believe this study has several methodological strengths, including the use of non-pathological gamblers (instead of non-gambling controls) as a comparison group and non-treatment seeking PGs (instead of treatment seeking PGs). NPGs resemble pathological gamblers in their attraction to gambling and engagement in gambling behavior, which decreases the probability of conflating gambling behavior with personality and addresses directly potential differences in
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
879
those who gamble with no gambling pathology from those who gamble with maladaptive and debilitative patterns of gambling behavior. Treatment seeking PGs comprise less than 2% of the PGs found in community samples and are therefore not representative of the population of pathological gamblers (Wallisch, 1996); the use of non-treatment seeking PGs in this study enhances the generalization of our results. There were also limitations associated with the current investigation. Due to the cross-sectional nature of this study, the temporal relationships between personality and PG could not be assessed. Whereas the use of non-treatment seeking PGs increases the overall generalizability of the results, the personality traits that characterize the PGs in this study may not be applicable to treatment seeking PGs. Several potential confounding factors related to social desirability, demographic differences (e.g., socio-economic status, education, ethnicity), memory errors, and context effects were not controlled for. Finally, we did not calculate inter-rater reliability estimates across the interviewers who conferred the diagnosis of PG, as different raters were used in the two studies, the participants from which were combined to maximize statistical power. We believe it unlikely that systematic interviewer error influenced diagnosis, however, as the proportion of PGs to NPGs across the two study samples did not differ and the interviewers were well trained and bi-weekly consultation sessions were held in which diagnostic issues were discussed and cases reviewed. References American Psychiatric Association (APA) (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Press. Bagby, R. M., Costa, P. T., McCrae, R. R., Livesley, W. J., Kennedy, S. H., Levitan, R. D., et al. (1999). Replicating the five factor model of personality in a psychiatric sample. Personality and Individual Differences, 27(6), 1135–1139. Bagby, R. M., Vachon, D. D., Bulmash, E., & Quilty, L. C. (in press). Pathological gambling and personality disorders: a review and re-examination of prevalence rates. Journal of Personality Disorders. Blaszczynski, A., & Nower, L. (2002). A pathways model of problem and pathological gambling. Addiction, 97, 487–499. Blum, K., Dherid, P. J., Wood, R. C., Braverman, E. R., Chen, T. J., Cull, J. G., et al. (1996). The D2 dopamine receptor gene as a determinant of reward deficiency syndrome. Journal of the Royal Society of Medicine, 89, 396–400. Breen, R. B., & Zuckerman, M. (1999). Chasing in gambling behaviour: personality and cognitive determinants. Personality and Individual Differences, 27, 1097–1111. Cloninger, C. R., Svrakic, D. M., & Przybeck, T. R. (1993). A psychobiological model of temperament and character. Archives of General Psychiatry, 50, 975–990. Comings, D. E., Rosenthal, R. J., Lesieur, H. R., & Rugle, L. (1996). A study of the dopamine D2 receptor gene in pathological gambling. Pharmacogenetics, 6, 223–234. Costa, P. T., & McCrae, R. R. (1992). Revised NEO personality inventory (NEO-PI-R) and NEO five-factor inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources. Costa, P. T., & Widiger, T. A. (2001). Personality disorders and the five factor model of personality. Washington, DC: American Psychological Association. Eysenck, H. J., & Eysenck, S. B. G. (1975). Manual of the EPQ (Eysenck personality questionnaire). San Diego, CA: Educational and Industrial Testing Service. Gerdner, R., & Svensson, K. (2003). Predictors of gambling problems among male adolescents. International Journal of Social Welfare, 12, 182–192. Jang, K. L., McCrae, R. R., Angleitner, A., Riemann, R., & Livesley, J. W. (1998). Heritability of facet-level traits in a cross-cultural twin sample: support for a hierarchical model of personality. Journal of Personality and Social Psychology, 74, 1556–1565.
880
R.M. Bagby et al. / Personality and Individual Differences 43 (2007) 873–880
Langewisch, M. W. J., & Frisch, G. R. (1998). Gambling behaviour and pathology in relation to impulsivity, sensation seeking, and risky behaviour in male college students. Journal of Gambling Studies, 14, 245–262. Malouff, J. M., Thorsteinsson, E. B., & Shutte, N. S. (2005). The relationship between the five-factor model of personality and symptoms of clinical disorders: a meta-analysis. Journal of Psychopathology and Behavioral Assessment, 27, 101–114. McCrae, R. R., & Costa, P. T. Jr., (1995). Personality trait structure as a human universal. American Psychologist, 52, 509–516. Nower, L., Derevensky, J. L., & Gupta, R. (2004). The relationship of impulsivity, sensation seeking, coping, and substance use in youth gamblers. Psychology of Addictive Behaviours, 18, 49–55. Patrick, C. J., Curtin, J. J., & Tellegen, A. (2002). Development and validation of a brief form of the multidimensional personality questionnaire. Psychological Assessment, 14, 150–163. Petry, N. M. (2005). Pathological gambling: etiology, comorbidity, and treatment. Washington, DC: American Psychological Association. Petry, N. M., Stinson, F. S., & Grant, B. F. (2005). Comorbidity of DSM-IV pathological gambling and other psychiatric disorders: results from the national epidemiologic survey on alcohol and related conditions. Journal of Clinical Psychiatry, 66, 564–574. Raylu, N., & Oei, T. P. S. (2002). Pathological gambling: a comprehensive review. Clinical Psychological Review, 22, 1009–1061. Roy, A., Custer, R., & Lorenz, V. (1989). Personality factors in pathological gambling. Acta Psychiatrica Scandinavica, 80, 37–39. Slutske, W. S., Caspi, A., Moffitt, T. E., & Poulton, R. (2005). Personality and problem gambling: a prospective study of a birth cohort of young adults. Archives of General Psychiatry, 62, 769–775. Slutske, W. S., True, W. R., Goldberg, J., Eisen, S., Xian, H., Lyons, M. J., et al. (2001). A twin study of the association between pathological gambling and antisocial personality disorder. Journal of Abnormal Psychology, 110, 297–308. Steel, Z., & Blaszczynski, A. (1996). The factorial structure of pathological gambling. Journal of Gambling Studies, 12, 3–20. Wallisch, L. S. (1996). Gambling in Texas: 1995 surveys of adult and adolescent gambling behavior. Austin, TX: Texas Commission on Alcohol and Drug Abuse. Whiteside, S. P., & Lynam, D. R. (2001). The five factor model and impulsivity: using a structural model of personality to understand impulsivity. Personality and Individual Differences, 20, 669–689. Zuckerman, M., Kuhlman, D. M., Thornquist, M., & Kiers, H. (1991). Five (or three) robust questionnaire scale factors of personality without culture. Personality and Individual Differences, 12, 929–941.