Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Jul 28.
Published in final edited form as: J Pers. 2010 Apr;78(2):441–470. doi: 10.1111/j.1467-6494.2010.00622.x

A Self-Regulatory Model of Behavioral Disinhibition in Late Adolescence: Integrating Personality Traits, Externalizing Psychopathology, and Cognitive Capacity

Tim Bogg 1, Peter R Finn 1
PMCID: PMC3145208  NIHMSID: NIHMS193557  PMID: 20433626

Abstract

Two samples with heterogeneous prevalence of externalizing psychopathology were used to investigate the structure of self-regulatory models of behavioral disinhibition and cognitive capacity. Consistent with expectations, structural equation modeling in the first sample (N = 541) showed a hierarchical model with three lower-order factors of impulsive sensation-seeking, anti-sociality/unconventionality, and lifetime externalizing problem counts, with a behavioral disinhibition superfactor best accounted for the pattern of covariation among six disinhibited personality trait indicators and four externalizing problem indicators. The structure was replicated in a second sample (N = 463) and showed that the behavioral disinhibition superfactor, and not the lower-order impulsive sensation-seeking, anti-sociality/unconventionality, and externalizing problem factors, was associated with lower IQ, reduced short-term memory capacity, and reduced working memory capacity. The results provide a systemic and meaningful integration of major self-regulatory influences during a developmentally important stage of life.

Behavioral disinhibition refers to a pattern of antisocial, impulsive, norm-violating, sensation-seeking, and externalizing tendencies and problems (e.g., substance use, attention deficit; Iacono et al., 1999), which has been shown to be a primary psychological substrate of several of the leading behavioral contributors to mortality (i.e., alcohol, drug, and firearms-related deaths; Bogg & Roberts, 2004; Mokdad et al., 2004). Researchers index behavioral disinhibition using a variety of constructs, including personality traits related to impulsivity and socialization (Clark & Watson, 1999; Donovan, Jessor, & Costa, 1991), sensation-seeking (Justus, Finn, & Steinmetz, 2000), externalizing psychopathology (alcohol, marijuana, and other drug problems; Krueger & Markon, 2006), and, less frequently, as cognitive ability variables, such as intelligence, short-term memory, and working memory (Finn, Justus, Mazas, & Steinmetz, 1999). Although these and other indicators of behavioral disinhibition have and continue to be targets of inquiry, a more complete model of the self-regulatory influences fundamental to behavioral disinhibition remains somewhat obscured by a lack of integration across these disparate research traditions (cf. Finn, 2002).

To address the need for a clearer articulation of the relations among disinhibited personality traits, externalizing problems, and cognitive capacity, the current study addresses two major goals: 1) model the associations among the related domains of disinhibited personality traits and externalizing psychopathology and; 2) investigate the resultant model’s relations to cognitive capacity outcomes previously found to be associated with various indicators of behavioral disinhibition. Compared to previous research, which has focused on a limited set of personality traits or externalizing problems, and has often excluded important cognitive capacity constructs, the approach of the current study attempts to provide a more comprehensive depiction of a self-regulatory model of behavioral disinhibition in late adolescence.

Disinhibited Personality Traits and Externalizing Psychopathology

Recent research demonstrates the presence of an underlying continuum of psychopathology and personality traits related to impulsivity and socialization that can help explain the multi-morbidity of externalizing problems (e.g., Kendler, Prescott, Myers, & Neale, 2003; Krueger & Markon, 2006). Specifically, research has shown that adolescent antisocial behavior, conduct disorder, alcohol dependence, drug dependence, and the personality trait of constraint share a common etiologic (primarily genetic) origin (Krueger et al., 2002). In addition, statistical modeling has shown this externalizing factor to be dimensional, rather than categorical in nature (Krueger, Markon, Patrick, & Iacono, 2005; Markon & Krueger, 2005). Taken together, the findings of a common genetic influence and a latent trait model for antisocial behavior and substance use problems indicate a coherence that can be conceptualized as a liability or spectrum for externalizing problems.

Although substantial progress has been made in understanding the meaning of the multi-morbidity of various externalizing disorders and their relation to disinhibited personality traits (cf. Krueger et al., 2007), it remains unclear how the structure of a more full account of disinhibited factors—a structure that would include multiple personality trait indicators of impulsivity and socialization, as well as externalizing disorder problems—might be related to important cognitive capacity outcomes. The structure of these multiple self-regulatory influences on behavioral disinhibition is important given that Krueger et al. (2002) have shown these traits (measured more narrowly as constraint in their work), while sharing a common additive genetic etiology with externalizing disorder problems, also have a significant component of additive genetic variance that is independent of that shared with externalizing problems. That is, there is a meaningful portion of the additive genetic variance of traits related to impulsivity and socialization that is unique to a source other than that which is shared with the externalizing disorder problems. In fact, in behavior genetic research, Krueger et al. (2002) found only 22 percent of the variance in constraint to be accounted for by the externalizing factor. This finding suggests that traits related to impulsivity and socialization, while sharing meaningful psychological (and biological) space with externalizing problems, are not wholesale components of an externalizing liability. In part, the current research is designed to further elucidate the relations among the components and subsequently examine how these components of behavioral disinhibition are related to multiple indicators of cognitive capacity.

Cognitive Capacity, Self-Regulation, and Behavioral Disinhibition

Working memory is an important component of a system of inter-related executive cognitive functions (Zelazo & Frye, 1998), including attentional capacity, attentional control, and attention shifting (Cowan, 1999), that plays a critical role in self-regulation and decision-making (Barkley, 1997, 2001, Bechara & Martin, 2004; Finn, 2002; Kimberg & Farah, 1993). As it pertains to self-regulation, increased working memory capacity enables fluid shifting of attention during the decision-making process from more salient proximal (immediate) to less salient distal (long-term) outcomes and allows for appropriate weighting and consideration of long-term consequences of decisions (i.e., less impulsive decisions; Finn, 2002; Finn & Hall, 2004; Oberauer, 2002). By contrast, reduced working memory capacity is related to general behavioral disinhibition (Barkley; 1997; 2001; Finn, 2002; Finn & Hall, 2004; Hinson, Jameson, & Whitney, 2003), which reflects the basic dispositional processes underlying externalizing problems (Finn, 2002; Krueger et al., 2002; Slutske et al., 2002). More generally, working memory is required for activated self-directed speech, self-reflection, and maintaining representations for the purpose of problem solving to guide socially adaptive behavior (Barkley, 2001; Finn, 2002; Oberauer, 2002).

In the present research, multiple measures of working memory capacity are included because working memory is complex, involving several inter-related processes associated with behavioral regulation, including short-term memory capacity, resistance to distraction, mental manipulation, attentional control in divided attention/dual task contexts, and maintenance of memory traces over time (Baddeley & Logie, 1999; Cowan, 1999; Engle et al., 1999; Finn, 2002). Engle et al. (1999) and Finn (2002) proposed multidimensional models of working memory capacity that distinguish a short-term capacity dimension, indicated by performance on simple span tasks, from a working memory capacity dimension, indicated by performance on complex, dual-task, span tasks requiring attention-shifting and maintenance. Engle et al. (1999) showed that these two dimensions of working memory were distinct from measures of intelligence. The current research uses this three-dimensional model of short-term memory, working memory, and intelligence and tests its viability using confirmatory factor analyses.

Recent research suggests that diminished executive cognitive capacities involved in working memory, short-term memory, and intelligence contribute to the development and maintenance of externalizing problems (Aytaclar, Tarter, Kirisci, & Lu 1999; Finn & Hall, 2004; Harden & Pihl, 1995; Pihl, Peterson & Finn, 1990; Poon, Ellis, Fitzgerald, & Zucker, 2000). Diminished capacity in these domains of cognitive ability also is thought to reflect a general predisposition to disinhibited, poorly regulated behavior rather than a vulnerability to a specific disorder (Finn, 2002, Finn & Hall, 2004; Giancola, Zeichner, Yarnell, & Dickenson, 1996; Harden & Pihl, 1995). In turn, this disinhibited predisposition can be made manifest as externalizing behavior, such as alcohol dependence, childhood conduct problems, adult antisocial behavior, or other substance abuse (Barkley, 1997; 2001; Finn, 2002; Finn & Hall, 2004; Krueger et al., 2002). As such, these indicators of cognitive capacity represent integral components of a systematic approach to modeling self-regulatory influences on behavioral disinhibition. It should be noted that incentive (i.e., reward) structures, delay of reward components, or learning components are absent from these cognitive tasks. The tasks used in the current study assess general cognitive capacity in the absence of contextual manipulations.

Similarly, the current study did not incorporate an appetitive or incentive (i.e., reward) structure or component to the cognitive tasks. Unlike the Iowa gambling task (Bechara, Damasio, Damasio, & Anderson, 1994), for example, the short-term memory and working memory tasks are not designed to assess or account for the influence of appetitive influences that call upon self-control for optimal performance. Instead, the tasks are agnostic in regard to such influences and are intended to assess decontextualized cognitive capacity (e.g., performance not influenced by monetary gains or losses).

The Present Study

The primary aim of the present research is to examine the structure of a self-regulatory model of behavioral disinhibition. Although other personality trait domains are relevant to behavioral disinhibition (e.g., agreeableness, hostility, emotional stability), we focus on traits related to impulsivity and socialization, as well as externalizing problems, as important and related components of behavioral disinhibition (e.g., Bogg & Roberts, 2004; Finn, 2002; Iacono et al., 1999; McGue, Iacono, & Krueger, 2006). Two large-scale community samples are used to model the relations among these disinhibited (i.e., related to impulsivity and socialization) personality traits and externalizing problems.

Aside from modeling the relations among disinhibited personality traits and externalizing problems, we examine how the resulting model is related to cognitive capacity (Engle et al., 1999); a set of abilities which research has shown differentiate individuals who meet diagnostic criteria for externalizing disorders from those who do not (Aytaclar et al., 1999; Finn & Hall, 2004; Harden & Pihl, 1995). Specifically, we attempt to clarify these relations by examining three competing models of the structure of disinhibited traits and externalizing problems in late adolescent/young adult samples. Previous research indicates individuals in late adolescence and emerging adulthood are establishing patterns of behavioral disinhibition, while remaining somewhat developmentally and neurologically malleable, and are well suited for addressing self-regulatory questions whose answers can provide needed perspective on early intervention and treatment for at-risk individuals (Lubman, Yücel, & Hall, 2007; Monti et al., 2005; Zucker et al., 2006). Based in part on the behavior genetic research of Krueger et al. (2002), we expect the initial structure to yield separable components of disinhibited personality traits and externalizing problems, rather than a single factor. Three approaches guided the modeling process.

First, in line with the idea of an underlying dimension of behavioral disinhibition (e.g., Iacono et al., 1999), a measurement model consisting of one latent variable (with ten indicators—four comprised of lifetime externalizing problem counts and six of disinhibited personality trait scales) was examined. If it best represented the data, then this one-factor model of disinhibited personality traits and externalizing problems would maximize parsimony while indicating that the covariance shared among the ten indicators did, in fact, represent a single dimension of behavioral disinhibition that accounted for the multi-morbidity of externalizing disorders and explained the interrelations of traits related to impulsivity and socialization and externalizing problems.

The second approach to modeling was guided, in part, by research examining the factor structure of personality scales and inventories related to the Big Five personality trait domain of conscientiousness (Roberts, Chernyshenko, Stark, & Goldberg, 2005). Inherent in the domain of conscientiousness is a consideration of disinhibition, which Clark and Watson (1999) in their “Big Three” model of personality traits describe as disinhibition versus constraint, wherein “disinhibited individuals are impulsive and somewhat reckless and are oriented primarily toward the feelings and sensations of the immediate moment; conversely, constrained individuals plan carefully, avoid risk or danger, and are controlled more strongly by the longer-term implication of their behavior (p. 403).” This definition maps neatly onto a similar framework outlined by Pickering and Gray (1999; see also Zuckerman (2003; 2005)), who label this constellation of facets impulsive sensation-seeking. Consistent with these conceptions, Roberts et al. (2005) factor-analyzed 36 conscientiousness-related scales from seven personality inventories and found a six-factor structure labeled order, self-control, responsibility, industriousness, traditionalism, and virtue. Based on this six-factor structure, the impulsivity and socialization-related scales included in the current study load onto the factors of self-control and responsibility. In line with the definition of Pickering and Gray (1999), we take a more expansive view of self-control, and use the label impulsive sensation-seeking for the personality scales aligning with this factor (i.e., the impulsivity subscale of the Eysenck impulsivity/venturesomeness scale, the control subscale of the Multidimensional Personality Questionnaire, and the disinhibition and boredom susceptibility subscales of the Sensation Seeking Scale). Conversely, although there are aspects of irresponsibility in the measures of socialization used in the current study, this factor is more accurately labeled anti-sociality/unconventionality, reflecting the more heterogeneous content of the scales used to assess this factor (i.e., the socialization subscale of the California Psychological Inventory and the psychopathic deviate scale of the Minnesota Multiphasic Personality Inventory-2). A third factor representing externalizing problems (i.e., lifetime alcohol, marijuana, other drug, and conduct/antisocial personality disorder problem counts) was examined in conjunction with impulsive sensation-seeking and anti-sociality/unconventionality to provide representation of the subclinical and clinical symptomatic expression of behavioral disinhibition that might be distinguished from disinhibited personality trait tendencies. Finally, in keeping with the idea of an overlying tendency, a latent superfactor, behavioral disinhibition, was modeled and indicated by the impulsive sensation-seeking, anti-sociality/unconventionality, and externalizing problems factors. If it best represented the data, then this hierarchical four-factor model would provide an intuitive dimensional system of disinhibited personality traits and externalizing tendencies made manifest by the expression of impulsive, antisocial, and externalizing tendencies and problems, organized under a general disposition for behavioral disinhibition.

The third approach to modeling was derived from previous research examining the structure of traits related to disinhibition (Justus et al., 2000). The findings of Justus et al. (2000) revealed a structure consisting of, 1) impulsivity (comprised of the impulsivity subscale of the Eysenck impulsivity/venturesomeness scale and the control subscale of the Multidimensional Personality Questionnaire; 2) social deviance proneness (comprised of the socialization subscale of the California Psychological Inventory, the psychopathic deviate scale of the Minnesota Multiphasic Personality Inventory-2, and a total count of antisocial problems); and 3) excitement seeking (comprised of the disinhibition and boredom susceptibility subscales of the Sensation Seeking Scale). In the current research, a fourth factor related to substance use problems (indicated by lifetime alcohol, marijuana, and other drug problem counts) is examined in conjunction with the three factors described above to provide an account of the covariance among the alcohol, marijuana, and other drug problems that might be distinguishable from antisocial personality and conduct problems (in contrast to the lower-order externalizing factor described in the second approach). Similar to the second approach outlined above, a fifth overlying factor, behavioral disinhibition, was modeled and indicated by the impulsivity, social deviance proneness, excitement seeking, and substance use problems subfactors. If it best represented the data, then this hierarchical five-factor model would serve to replicate and extend the research of Justus et al. (2000) and would suggest a more thoroughly segmented framework with multiple related dimensions of disinhibited traits and externalizing problems, organized under a general disposition for behavioral disinhibition.

Assuming replication of one of the above structural models of disinhibited personality traits and externalizing problems from the first sample to the second, the final model including correlations to cognitive capacity should represent a more coherent self-regulatory system of behavioral disinhibition; one that accounts for the interrelations among externalizing problems and personality traits related to impulsivity and socialization, while allowing for unique relations from these domains to cognitive capacity outcomes.

Method

Participants

Sample 1

We recruited a community sample (n = 541) of adolescents and young adults with a mean age of 20.7 (SD = 1.87) years. The sample was sex-balanced (48.1 % women) and most participants were European-American/Caucasian (81.9 %), followed by Asian/Asian-American (9.4 %), African-American (3.5 %), Hispanic (2.9 %), and Native American (2.2 %). At the time of assessment, the sample averaged 14.1 years of education (SD = 1.66 years), indicating an over-representation of college students in the sample. Nearly half of the Study 1 sample participants met diagnostic criteria for alcohol dependence (47 %), 36.2 % for marijuana dependence, and 43.1 % for other drug dependence. Additionally, 36.9 % met diagnostic criteria for conduct disorder, and 23.3 % for antisocial personality disorder. Slightly more than one-third of the sample (34.1 %) did not meet diagnostic criteria for any of the above problems.

Sample 2

We recruited a community sample (n = 463) of adolescents and young adults with a mean age of 21.98 (SD = 2.85) years. The sample was sex-balanced (46.4 % women) and most participants were European-American/Caucasian (76.9 %), followed by African-American (12.5 %), Asian/Asian-American (6.5 %), Hispanic (3.5 %), and Other (.6 %). At the time of assessment, the sample averaged 13.81 years of education (SD = 1.99 years), indicating an over-representation of college students in the sample. More than half of the Study 2 sample participants met diagnostic criteria for alcohol dependence (56.4 %), 36.1 % for marijuana dependence, and 22.9 % for other drug dependence Additionally, 49 % met diagnostic criteria for conduct disorder and 16 % for antisocial personality disorder. Slightly less than one-third of the sample (30.2 %) did not meet diagnostic criteria for any of the above problems.

Assessment Materials

Diagnostic interviews

Substance dependence (i.e., alcohol, marijuana, and other drug) diagnoses and problem counts were ascertained from responses on the Semi-structured Assessment for the Genetics of Alcoholism (Bucholz et al., 1994) using criteria from the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV, American Psychiatric Association, 1994). Histories and lifetime problem counts for childhood conduct and antisocial personality disorders also were ascertained from responses to the SSAGA, also using DSM-IV criteria. In subsequent analyses, lifetime problem counts for conduct disorder and ASPD were summed to form a CDASPD variable that reflected a lifetime history of antisocial behavior problems and tendencies. Because of their skewed distributions, all problem counts were Blom transformed for subsequent analyses.1 Descriptive statistics for the raw lifetime problem counts are presented in Table 1.

Table 1.

Descriptive Statistics for Externalizing Problem Counts and Disinhibited Personality Trait Scale Scores

Sample 1 Mean (SD) Sample 2 Mean (SD)
Lifetime alcohol problems 19.67 (16.82) 27.99 (21.70)
Lifetime marijuana problems 3.34 (4.18) 10.04 (11.71)
Lifetime other drug problems 2.66 (6.10) 14.37 (26.37)
Lifetime CDASPD problems 13.13 (8.87) 26.54 (16.33)
CPI Socialization 31.95 (6.80) 28.78 (8.04)
EYS Impulsiveness 9.17 (4.72) 10.28 (4.61)
MMPI-2 Psychopathic Deviate 19.43 (5.49) 21.89 (6.31)
MPQ Control 12.43 (6.36) 11.44 (6.01)
SSS Boredom Susceptibility 3.81 (1.98) 3.71 (2.05)
SSS Disinhibition 4.27 (1.79) 4.42 (1.65)
Open in a new tab

Note. Problems are lifetime history problem counts from the Semi-structured Assessment for the Genetics of Alcoholism (SSAGA). CDASPD = Sum of lifetime history problem counts for conduct disorder and antisocial personality disorder from the SSAGA. CPI = California Psychological Inventory; EYS = Eysenck Impulsiveness-Venturesomeness scales; MMPI = Minnesota Multiphasic Personality Inventory; MPQ = Multidimensional Personality Questionnaire; SSS = Sensation Seeking Scale.

Personality trait indicators

Six well-validated personality scales were used to assess traits related to impulsivity and socialization: The Impulsivity scale from the Eysenck Impulsivity-Venturesomeness test (EYS-IMP; Eysenck & Eysenck, 1978), the Control subscale of the Multidimensional Personality Questionnaire (MPQ-Control; Tellegen, 1982), the Disinhibition (DIS) and Boredom Susceptibility (BS) subscales of the Sensation Seeking Scale (SSS; Zuckerman, 1979), the Psychopathic Deviate (Pd) scale of the Minnesota Multiphasic Personality Inventory-2 (MMPI-2-Pd; Hathaway & McKinley, 1989), and the Socialization (So) scale of the California Psychological Inventory (Gough, 1969).2 The EIV-IMP scale is comprised of 19 items (e.g., “Do you need to use a lot of self-control to keep out of trouble?”) using a dichotomous response scale (i.e., “yes” or “no”; sample 2, α = .83). The MPQ-Control scale is comprised of 24 items (e.g., “When faced with a decision I usually take time to consider and weigh all aspects.”) using a dichotomous response scale (i.e., “true” or “false”; sample 2, α = .87). High EIV-IMP scores and low MPQ-Control scores reflect low self-control, spontaneity, recklessness, and a lack of planning and foresight. The SSS-DIS scale is comprised of 10 forced-choice items (e.g., “I like “wild” uninhibited parties” versus “I prefer quiet parties with good conversation”; sample 2, α = .53) and the SSS-BS is comprised of 10 forced-choice items (e.g., “I enjoy looking at home movies or travel slides” versus “Looking at someone’s home movies or travel slides bores me tremendously”; sample 2, α = .55).3 Three items that directly referred to drinking or drug use were dropped from the DIS scale: “I often like to get high (drinking liquor or smoking marijuana)”; “Keeping drinks full is the key to a good party”; and “I feel best after taking a couple of drinks.” The MMPI-2-Pd scale is comprised of 50 items (e.g., “What others think of me does not bother me,” “Sometimes when I was young I stole things,” “No one seems to understand me”) using a dichotomous response scale (i.e., “true” or “false”: sample 2, α = .76). High scores on the MMPI-2-Pd scale indicate impulsivity, problems with authority, antisocial acts, and alienation (Almagor & Koren, 2001). One item (“I have used alcohol excessively”) was dropped from the Pd scale because of contamination with the lifetime alcohol problem counts. The CPI-So scale is comprised of 54 items (e.g., “I often act on the spur of the moment without stopping to think,” “I keep out of trouble at all costs,” “As a youngster in school, I used to give the teachers lots of trouble”: sample 2, α = .80). Low scores on the CPI-So scale reflect under-socialized and norm-violating tendencies. All the personality scales have shown relations with behaviors and outcomes related to substance use and antisocial behaviors and problems (Bogg & Roberts, 2004). Descriptive statistics for the six personality scales are presented in Table 1.

Intelligence

Intelligence was measured using the Shipley Institute of Living Scale estimates of IQ (Zachary, 1986). The Shipley is a self-administered measure of intelligence that strongly correlates (median correlation = .79) with the WAIS Full Scale IQ (Zachary, 1986). As with other abbreviated measures of general intelligence, the Shipley does not include a component for testing memory.

Short-term memory capacity

Short-term memory capacity was assessed with the digits forward and backward scales of the Digit Span subtest of the Wechsler Adult Intelligence Scale-Revised (WAIS-R; Wechsler, 1981). Digits forward and backward are commonly used measures of STM capacity and general attentional capacity in working memory system studies (Engle et al., 1999; Finn, 2002). Short-term memory capacity also was assessed with the letter-number sequencing task from the WAIS-III (Wechsler, 1997). The letter-number sequencing task involves the presentation of increasingly larger sets of letters and numbers (e.g., G-6-B-2) to the participant who, at the end of each set, is asked to verbally recall the numbers in numerical order (e.g., 2–6) and the letters in alphabetical order (e.g., B-G). Set sizes vary from 2 to 8 items.

Working memory capacity

Working memory functions of dual task ability, divided attention, and maintenance capacity were assessed with the Operation-Word Span test (OPWS; Conway and Engle, 1994) and a modified version of the Auditory Consonant Trigram test (ACT: Brown, 1958). The OPWS involves competition for attentional resources (divided attention) and the maintenance of activation of mental representations in a dual task context. This task requires the participant to solve a simple mathematical operation while remembering a word (6/3 + 2 = 4 DOG). The participant reads the math operation aloud, responds “yes” or “no” to indicate if the answer is correct or not and then says the word. One half of the mathematical operations are correct. After a series of operation-word pairs, the participant is asked to recall the words (series vary from 2 to 6 operation/word pairs). The total number of correctly recalled words is the variable derived from the OPWS.

The ACT was modified to include four and five nonsensical strings of consonants, in addition to the original three-string (trigram) consonant stimuli, to increase the overall load on the working memory system. The ACT requires the experimenter to read aloud a string of consonants at a rate of one letter per second, and immediately following the string, to read aloud a random three-digit number. The participant is then asked to begin counting aloud backwards in increments of three from the random three-digit number for an interval of 18 or 36 seconds, at which time the participant is asked to stop counting and recall the original consonant string. This task taps divided attention and the strength of the maintenance/decay of the contents of working memory over time (Brown, 1958; Stuss et al., 1987). Counting backwards is used to prevent rehearsal of the consonant string. The task included four different three, four, and five consonant strings. For each string length, two were followed by 18-second delay intervals and two were followed by 36-second delay intervals. The dependent variable is the total number of correct consonants recalled across all string lengths and delay intervals.

Single-factor and correlated three-factor models of the cognitive capacity variables were compared to assess the appropriateness of the three-factor model indicated by Engle et al. (1999) and Finn (2002). As suggested by a more negative BIC value and a lower AIC value (the fit indices are described in greater detail below), the three-factor model provided better fit [χ2(7, N = 463) = 13.10, p = .07, RMSEA = 0.043, BIC = −42.14, AIC = −0.90] than the one-factor model [χ2(9, N = 463) = 117.56, p < .05, RMSEA = .162, BIC = 62.32, AIC = 99.56] and was used in subsequent correlation analyses. The correlated three-factor model was comprised of 1) Shipley IQ (IQ); 2) short-term memory (indicated by letter number, digits forward, and digits backward scores); and 3) working memory (indicated by operation word-span and auditory consonant trigram scores).

Procedure

Participants were recruited from the community by screening telephone responses to advertisements placed in local newspapers and around the local community. Advertisements were designed using Widom’s (1977) approach to attract responses from individuals varying in terms of the level of disinhibited traits and tendencies. Highly disinhibited participants were targeted with advertisements asking for responses from “adventurous, carefree individuals who have led exciting and impulsive lives,” “daring, rebellious, defiant individuals,” “individuals on probation or who have been in trouble with the law,” as well as “persons with a drinking problem,” and “social drinkers.” Participants with average or low levels of disinhibited traits were targeted with advertisements asking for responses from “persons interested in psychological research” or “quiet, reflective and introspective persons.” Participants were excluded if they were not between 18 and 25 years of age, were taking any psychotropic or antihistamine medications, had never consumed alcohol, had a history of heart disease or psychosis, were not able to speak or read English, or had less than a grade 6 education level.

Participants were asked to refrain from excessive use of alcohol or drugs for the 24-hour period prior to each session, to refrain from any use for at least 12 hours prior to testing, and to eat a meal within three hours of testing. Prior to testing, participants were administered a breath-alcohol test using an AlcoSensor-III (Intoximeters, Inc.) to ensure that their breath-alcohol level (BAL) was 0.00%.

Participants completed a questionnaire asking about when they had last eaten food, their drug use in the past 24 hours, and their level of fatigue. If a participant had a BAL greater than 0.00%, reported taking any other psychoactive drugs the day the testing, appeared to be high the day of testing, or was extremely fatigued, then the participant was rescheduled. Participants read and signed an informed consent to participate, were free to refuse any procedure, and were paid $7.00 per hour. The diagnostic interview was administered first, followed by an interspersed ordering of the personality trait and cognitive capacity measures. The total time of assessment was approximately 2–3 hours in sample 1 and 3–4 hours in sample 2.

Analyses

In samples 1 and 2, the three measurement models of the covariance among the four problem-count indicators (alcohol, marijuana, other drug, and CDASPD) and the six personality trait indicators were analyzed using structural equation modeling (via AMOS 7). A single latent factor model (i.e., behavioral disinhibition), a hierarchical four-factor model (i.e., impulsive sensation-seeking, anti-sociality/unconventionality, externalizing problems, and behavioral disinhibition superfactor), and a hierarchical five-factor model (i.e., impulsivity, social deviance proneness, excitement-seeking, substance use problems, and behavioral disinhibition superfactor) were analyzed.

In the one-factor and four-factor models, residual terms for the substance use problem variables (i.e., alcohol, marijuana, other drug) were allowed to freely co-vary to take into account the unique variance of these variables that was not shared with the other indicators. It was anticipated that the residual variance of the substance use indicators reflected meaningful components of substance use variance that were not captured by covariation with the personality trait indicators and/or the CDASPD indicator in the one- and four-factor models. The residual terms for each substance use variable were expected to represent unique components of substance use problems that would correlate with the residual terms of the other substance use variables in the context of other non-substance-use indicators. In the five-factor model, the three substance use variables indicated a separate and ‘clean’ substance-use latent variable, thereby negating the need to allow for their residual terms to freely co-vary in that model.

The three models were compared using the Bayesian Information Criterion (BIC) and the Akaike Information Criterion (AIC) as the primary arbiters of appropriateness in the process of selecting one model over the others. Both BIC and AIC aid in selecting models by identifying which model among competing models reproduces the observed variances and covariances with the fewest estimated parameters (i.e., with the most parsimony). Lower (i.e., more negative) BIC values indicate better comparative fit in terms of the odds of one model being superior to the other (Raftery, 1995). Specifically, a difference of 10 points between two given models indicates that the odds are approximately 150:1 that the model with the lower (i.e., more negative) BIC value provides a better fit than the model with the higher (i.e., less negative) BIC value (Raftery, 1995). Lower AIC scores also indicate better comparative fit (Akaike, 1987), but are not interpreted as odds. The Root Mean Square Error of Approximation (RMSEA) also is reported, but is not used for comparative purposes. Rather it is used to quantify the closeness of fit of each model in relation to its degrees of freedom (Browne & Cudeck, 1993), with values approaching zero indicating close fit. Browne and Cudeck (1993) advise that a RMSEA value of approximately .08 indicates a reasonable error of approximation. Similarly, the Comparative Fit Index (CFI; Bentler, 1990) is reported. CFI scores range from 0–1, where a score of .85, for example, means that 85 % of the covariation in the data is reproduced by a tested model. A CFI score above .90 suggests adequate fit.

Using only sample 2, correlations were examined between the latent factors of the selected measurement model found in sample 1 and replicated in sample 2 and the three-factor model of cognitive capacity.

Results

Intercorrelations among trait and problem count indicators

Table 2 displays the intercorrelations among the disinhibited personality trait scales and the lifetime externalizing problem counts for both samples.

Table 2.

Intercorrelations Among Lifetime Externalizing Problem Counts and Disinhibited Personality Traits

ALC Problems MARJ Problems Other Drug Problems CDASPD Problems CPI-Soc EYS-Imp MMPI-PD MPQ-Contr SSS-BS SSS-Dis
ALC Problems -- .55 .55 .56 −.45 .34 .35 −.33 .27 .23
MARJ Problems .64 -- .56 .45 −.44 .24 .36 −.26 .17 .25
Other Drug Problems .53 .54 -- .45 −.39 .21 .32 −.19 .15 .12
CDASPD Problems .69 .60 .52 -- −.74 .48 .57 −.43 .38 .32
CPI-Soc −.56 −.51 −.48 −.66 -- −.53 −.71 .52 −.39 −.32
EYS-Imp .49 .37 .33 .51 −.54 -- .41 −.74 .41 .33
MMPI-PD .53 .40 .39 .62 −.73 .46 -- −.33 .28 .21
MPQ-Contr −.45 −.31 −.26 −.39 .46 −.73 −.38 -- −.38 −.32
SSS-BS .21 .15 .12 .23 −.28 .35 .24 −.40 -- .39
SSS-Dis .27 .16 .09 .24 −.22 .36 .12 −.37 .37 --
Open in a new tab

Note. Sample 1 (N = 541) correlations above diagonal, Sample 2 (N = 463) correlations below diagonal. ALC Problems = Sum of alcohol problem counts from the Semi-structured Assessment for the Genetics of Alcoholism (SSAGA), MARJ Problems = Sum of marijuana problem counts from the SSAGA, Other Drug Problems = Sum of other drug problem counts from the SSAGA, CDASPD = Sum of problem counts for conduct disorder and antisocial personality disorder from the SSAGA, CPI-Soc = Socialization scale from California Psychological Inventory, EYS-Imp = Impulsivity scale from Eysenck Impulsivity-Venturesomeness test , MMPI-PD = Psychopathic Deviate scale form Minnesota Multiphasic Personality Inventory-2, MPQ-Contr = Control subscale from Multidimensional Personality Questionnaire, SSS-BS = Boredom Susceptibility scale from the Sensation Seeking Scale, SSS-Dis = Disinhibition scale from the Sensation Seeking Scale. All correlations are statistically significant at p < .05.

Sample 1 Model Comparisons

Table 3 displays the fit statistics and indices associated with the one-, four-, and five-factor models in sample 1. As is indicated by its more negative BIC value and lower AIC value, the hierarchical four-factor model of impulsive sensation-seeking, antisociality/unconventionality, externalizing problems, and behavioral disinhibition provided the best comparative fit among the models.4 In addition, the RMSEA score of the hierarchical four-factor model indicated a reasonable error of approximation. The CFI score also suggested good fit, indicating more than 96 % of the covariation in the data was reproduced by the four-factor model.

Table 3.

Four-Factor Hierarchical Model of Disinhibited Personality Traits and Externalizing Problems Provides Best Fit Across Samples 1 and 2

Fit Statistics and Indices
χ2 df RMSEA CFI BIC AIC
Sample 1 (N = 541)
 One-Factor Model 405.14* 32 .147 .854 203.75 341.14
 Four-Factor Model 115.05* 29 .074 .966 −67.46 57.05
 Five-Factor Model 339.68* 32 .133 .880 138.29 275.68
Sample 2 (N = 463)
 One-Factor Model 434.17* 32 .165 .832 237.76 370.17
 Four-Factor Model 95.85* 29 .071 .972 −82.14 37.85
 Five-Factor Model 351.76* 32 .147 .867 155.35 287.76
Open in a new tab

Note. RMSEA = Root Mean Square Error of Approximation; CFI = Comparative Fit Index; BIC = Bayesian Information Criterion; AIC = Akaike Information Criterion. Lower RMSEA indicates better closeness of fit for each model in relation to its own degrees of freedom. CFI above .90 indicates good fit (i.e., covariation in the data is reproduced by the model). Lower (i.e., more negative) BIC scores and lower AIC scores indicate better comparative fit.

*

p < .05

Figure 1 shows path weights (single arrows) of three sets of endogenous indicators for impulsive sensation-seeking, anti-sociality/unconventionality, and externalizing problems, as well as the path weights for the three subfactors from the behavioral disinhibition superfactor (all p < .05). The expected pattern of covariation between the residual terms of the substance use indicators was found. The correlations for the residual terms of the substance use indicators are illustrated by the double arrows among the residual terms (small circles) in Figure 1.

Figure 1.

Figure 1

Open in a new tab

Hierarchical four-factor structure of disinhibited personality traits and externalizing problems for sample 1 (N = 541). EYS-Imp = Impulsivity scale from Eysenck Impulsivity-Venturesomeness test , MPQ-Contr = Control subscale from Multidimensional Personality Questionnaire, SSS-BS = Boredom Susceptibility scale from the Sensation Seeking Scale, SSS-Dis = Disinhibition scale from the Sensation Seeking Scale, CPI-Soc = Socialization scale from California Psychological Inventory, MMPI-PD = Psychopathic Deviate scale form Minnesota Multiphasic Personality Inventory-2, CDASPD = Sum of problem counts for conduct disorder and antisocial personality disorder from the Semi-structured Assessment for the Genetics of Alcoholism (SSAGA), ALC problems = Sum of alcohol problem counts from the SSAGA, MARJ Problems = Sum of marijuana problem counts from the SSAGA, Other Drug Problems = Sum of other drug problem counts from the SSAGA. All paths (single-arrowed lines) and correlations (double-arrowed lines) are statistically significant (p < .05).

Sample 2 Model Comparisons

Table 3 also displays the fit statistics and indices associated with the one-, four-, and five-factor models in sample 2. As is indicated by its more negative BIC value and lower AIC value, the hierarchical four-factor model of impulsive sensation-seeking, antisociality/unconventionality, externalizing problems, and behavioral disinhibition provided the best comparative fit among the models. In addition, the RMSEA score of the hierarchical four-factor model indicated a reasonable error of approximation. The CFI score also suggested good fit, indicating more than 97 % of the covariation in the data was reproduced by the four-factor model. These results replicate the findings from sample 1, which also indicated the comparative superiority of the hierarchical four-factor model.

Sample 2 Correlated Model of Behavioral Disinhibition and Cognitive Capacity

Figure 2 shows the path weights of impulsive sensation-seeking, anti-sociality/unconventionality, and externalizing problems from the behavioral disinhibition superfactor (all p < .05). The initial correlated model from the hierarchical four-factor model to the cognitive capacity variables designated bi-directional paths from the higher-order behavioral disinhibition factor to IQ, short-term memory, and working memory [χ2(93, N = 463) = 195.39, p < .05, RMSEA = 0.05], all of which resulted in significant correlations (p < .05). Three subsequent models examined bi-directional paths from the residual terms of the three subfactors (i.e., impulsive sensation-seeking, anti-sociality/unconventionality, and externalizing problems) to the cognitive capacity variables. None of these models resulted in statistically significant correlations (rs = .01 to −.11, all p > .05). As a result, the final predictive model depicted in Figure 2 excludes these nonsignificant paths. The final model shows that increased behavioral disinhibition, as indicated by impulsive sensation-seeking, anti-sociality/unconventionality, and externalizing problems, is associated with reduced cognitive capacity in the forms of reduced working memory capacity, lower IQ, and reduced short-term memory capacity.

Figure 2.

Figure 2

Open in a new tab

Final correlated model of Behavioral Disinhibition and the cognitive capacity constructs for sample 2 (N = 463). ACT = Auditory Consonant Trigram score, OPW = Operation Word Span score, DIG FOR = WAIS-R digits forward score, DIG BAC = WAIS-R digits backward score, LTR NMBR = WAIS-III letter-number sequencing score. All paths (single-arrowed lines) and correlations (double-arrowed lines) are statistically significant (p < .05).

Discussion

The goal of the present research was to evaluate a self-regulatory model of behavioral disinhibition in late adolescence/early adulthood. Two large samples were administered multiple personality scales assessing impulsive, norm-violating, and sensation-seeking tendencies (i.e., disinhibited personality traits) and were assessed for externalizing problems. Structural modeling identified a hierarchical four-factor structure of impulsive sensation-seeking, antisociality/unconventionality, externalizing problems, and a higher-order superfactor of disinhibition across both samples. Furthermore, a negative association between behavioral disinhibition and multiple domains of cognitive capacity was found, including IQ and indicators of short-term memory and working memory. The approach used in the current study augments the burgeoning literature examining psychopathology using continuous models of problems and symptoms by incorporating components of the larger system of self-regulatory influence that are known to be integrally related to behavioral disinhibition; namely, disinhibited personality traits and executive cognitive capacity.

The hierarchical four-factor structure of disinhibited personality traits and externalizing psychopathology found in the present research conforms to previous research on the etiology of these traits and problems (Krueger et al., 2002). Although these traits and problems have been shown to share a component of additive genetic influence, it also has been shown that the traits (measured as constraint) retain a substantial component of unique additive genetic influence. The modeling of the current research reflects these findings in the parsing of the various indicators as distinguishable factors of impulsive sensation-seeking, antisociality/unconventionality, and externalizing problems.

The finding of the divisibility of impulsive/sensation-seeking and antisocial traits also supports, in part, previous research on the lower-order structure of the personality trait domain of conscientiousness (Roberts et al., 2005), as well as theoretical positions and empirical evidence related to the division and content of the scales used in the present study. Specifically, narrative and factor-analytic depictions of the content of the CPI-Socialization and MMPI-2 Psychopathic Deviate scales converge on a description of these scales as assessing a heterogeneous pattern of alienated, antisocial, and norm-violating tendencies (Almagor & Koren, 2001; Gough, 1994), while Reinforcement Sensitivity Theory suggests the cluster of approach-oriented, uninhibited, and undercontrolled traits assessed via the MPQ-Control, EIV-Impulsiveness, SSS-Boredom Susceptibility, and SSS-Disinhibition scales represents a coherent domain (Pickering & Gray, 1999). The anti-sociality/unconventionality and impulsive sensation-seeking factors identified in the current research reflect these perspectives and provide some support for the separation of disinhibited traits along these two dimensions.

Although the structure of the subfactors represents a defensible division of related disinhibited tendencies, the results of the correlation analyses in sample 2 point to the important role of a behavioral disinhibition superfactor in an account of reduced cognitive capacity in the forms of working memory, short-term memory, and IQ. These results suggest that it is not any of the subfactors per se that is significantly related to the cognitive capacity outcomes, but the covariation among the subfactors. Consistent with Iacono et al.’s (1999) view of behavioral disinhibition as a generalized risk factor for various problems and disorders, the current research shows that it is a broad disposition of behavioral disinhibition that is associated with reduced working memory and short-term memory capacity, as well as lower IQ.

Keeping in mind the self-regulatory influence of working memory, the implication of this relation is that being behaviorally disinhibited means, in part, having a decreased capacity to keep something in mind (e.g., a behavioral norm) while being required to monitor and make decisions about the environment (e.g., a situational distraction that might interfere with keeping a behavioral norm in mind). Greater behavioral disinhibition increases the likelihood of an individual experiencing reduced capacity in his or her ability to retain sufficient attentional control to mitigate the influence of persistent distractions. An individual with greater behavioral disinhibition also is more likely to have reduced attentional capacity, as well as lower general cognitive ability.

More generally, the structural and correlated modeling results suggest behavioral disinhibition: 1) is a coherent global dispositional tendency, with strongly related components and manifestations; 2) demonstrates a global pattern of relations to three interrelated components of cognitive capacity; and 3) via its pattern of relations to the cognitive capacity constructs, provides a useful depiction of a larger system of self-regulatory influences, one which recognizes that cognitive capacity, disinhibited personality traits, and externalizing psychopathology are mutually informing and reinforcing.

Limitations and Conclusions

The current research is not without limitations and caveats. Primary among them is its cross-sectional design. The analyses in sample 2, in particular, do nothing to establish the predictive primacy of behavioral disinhibition or cognitive capacity. The predictive status that might be afforded these domains requires a longitudinal design. Such a design would 1) better account for cumulative development and transactions among disinhibited personality traits, externalizing psychopathology, and cognitive capacity; and 2) establish the predictive ordering of one or more sets of these constructs—assuming such an issue proves relevant. In addition to the limitation of a cross-sectional design was the targeted sampling scheme used in the current study. Although the sampling procedure was successful in recruiting disinhibited individuals, the resulting samples do not reflect the prevalence of these trait levels or problems in a ‘natural’ population. A large-scale, population-based longitudinal design would be better suited to establish a more precise structure as well as better estimates of the magnitude of the relations within that structure.

A second limitation concerns the network of individual difference constructs appropriate for inclusion in a depiction of relevant self-regulatory factors underlying behavioral disinhibition. Although somewhat more comprehensive than most previous research investigating aspects of behavioral disinhibition, the assessment of disinhibited personality traits, externalizing psychopathology, and cognitive capacity in the current research does not provide full coverage of the self-regulatory influences underlying behavioral disinhibition. Additional relevant individual differences factors include agreeableness, hostility/irritability/trait anger, neuroticism, and negative affect/emotionality, among others (Bettencourt, Talley, Benjamin, & Valentine, 2006; Elkins, King, McGue, & Iacono, 2006; Ohannessian & Hesselbrock, 2008). Similarly, the current study did not incorporate an appetitive or incentive (i.e., reward) structure or component to the cognitive tasks. Unlike the Iowa gambling task (Bechara, Damasio, Damasio, & Anderson, 1994), for example, the short-term memory and working memory tasks are not designed to assess or account for the influence of appetitive influences that call upon self-control for optimal performance. The inclusion of such tasks would undoubtedly aid in a more contextualized understanding of the relationship between cognitive capacity and behavioral disinhibition. Moreover, in any modeling of factors affecting the expression of any observable psychological features, one must also take into account the interplay of genetic and environmental influences. The above concerns do not invalidate the approach or findings of the present research so much as call attention to the array of influences underlying behavioral disinhibition—most of which could not possibly be assessed in a single design, but which deserve further integration so as to arrive at a better understanding of the self-regulatory substrata of behavioral disinhibition.

In spite of the limitations, the results of the studies suggest three important trends. First, across two samples, a dimensional approach to behavioral disinhibition yielded a structure of impulsive sensation-seeking, anti-sociality/unconventionality, and externalizing problems, with a behavioral disinhibition superfactor, that reflects the empirical and conceptual rendering of these tendencies found in diagnostic interviews and manuals, the findings of etiologic and factor-analytic research, and long-standing theoretical perspectives. Second, the behavioral disinhibition superfactor, but not the subfactors, was directly associated with reduced cognitive capacity in the forms of IQ, short-term memory, and working memory, revealing the global relations of this broad disposition. This second trend resulted from the finding that it was the covariation among the three lower-order latent indicators (represented as the behavioral disinhibition superfactor) that was related to cognitive capacity, rather than the indicators themselves. Third, and more broadly, the integration of disinhibited personality traits, externalizing problems, and cognitive capacity illuminates a larger system of interrelated self-regulatory influences underlying behavioral disinhibition—one of the most individually and interpersonally problematic patterns of behavioral expression.

Acknowledgments

This research was supported by National Institutes of Alcohol Abuse and Alcoholism grants R01 AA13650 and R01 AA10120 to Peter R. Finn.

Footnotes

1

Blom transformations rank order raw scores (settling ties by using the mean of the contested ranks) and then transform the ranks to z scores using the normal distribution. Simulation research comparing various transformations has shown that a Blom transformation of psychiatric symptom count data allowed for a more accurate selection of a true model from a set of alternative models (van den Oord et al., 2000).

2

The sample 1 assessment was conducted in the mid-1990s. Subsequently, and as part of the process of culling older data sets, data from the sample 1 assessment were compiled into a summary database that did not retain individual item responses. This process was implemented without foresight for the emergence of dimensional models of externalizing problems and disinhibited personality traits (and the subsequent desire to conduct reliability analyses). Consequently, alpha coefficients for sample 1 are not reported.

3

The DIS and BS scales, and not the Experience Seeking (ES) and Thrill and Adventure (TAS) scales, were used because research indicates that the DIS and BS scales reflect an underlying subfactor of sensation seeking with common genetic origins (Koopmans, Boomsma, Heath, & van Doornen, 1995) that better reflects excitement seeking (Finn et al., 2000). The TAS scale reflects low harm avoidance, rather than excitement seeking (Finn et al., 2000; 2002) and the ES scale reflects a preference for different types of experiences, rather than excitement seeking per se (Finn et al., 2000; 2002).

4

Although the approach for modeling the disinhibited personality scales and lifetime problem counts was guided by conceptual and/or theoretical arguments (i.e., confirmatory approach), one could argue that an exploratory approach is an appropriate alternative or complement. To address this concern, two exploratory approaches were used to investigate other possible structures of the personality scales and lifetime problem counts. The first approach used principal components analysis with oblique (Oblimin) rotation to identify two factors (via visual examination of the scree plot and Eigenvalues > 1.0) that explained 60–64 % of the variance across the two samples. The first factor was comprised of lifetime CDASPD, alcohol, marijuana, other drug problem counts, as well as the MMPI-PD and CPI-Soc scales; the second factor was comprised of the MPQ-Control, EIV-Imp, SSS-BS, and SSS-Dis scales. When analyzed in a latent variable framework, this correlated two-factor structure demonstrated poorer model-specific (RMSEA > .12, CFI < .91) and relative fit (ΔBIC > 10) than the hierarchical four-factor model examined using confirmatory analyses. The second exploratory approach used exploratory two-factor model specification searches (i.e., all possible indicator-factor combinations are analyzed—where 10 indicator variables and two latent variables yield 1,048,576 possible models). The searches produced inconsistent models across the two samples (likely due to sensitivity to smaller effects in these larger samples), as well as poorer model-specific and relative fit than the hierarchical four-factor model analyzed using the confirmatory approach. As a result of the poorer fit and inconsistent pattern of results for the exploratory approaches, and in keeping with the a priori conceptual rationales guiding the approaches to modeling the personality scales and lifetime problem counts, only the results of the confirmatory analyses are presented and discussed in the body of the report. Interested readers are invited to contact the first author regarding the specific findings (e.g., factor loadings) of the exploratory analyses.

References

  1. Akaike H. Factor analysis and AIC. Psychometrika. 1987;52:317–332. [Google Scholar]
  2. Almagor M, Koren D. The adequacy of the MMPI-2 Harris-Lingoes subscales: A cross-cultural factor analytic study of scales D, Hy, Pd, Pa, Sc, and Ma. Psychological Assessment. 2001;13:199–215. [PubMed] [Google Scholar]
  3. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4. Washington, DC: American Psychiatric Association; 1994. [Google Scholar]
  4. Aytaclar A, Tarter RE, Kirisci L, Lu S. Association between hyperactivity and executive cognitive function functioning in childhood and substance use in early adolescence. Journal of the American Academy of Child and Adolescent Psychiatry. 1999;38:172–178. doi: 10.1097/00004583-199902000-00016. [DOI] [PubMed] [Google Scholar]
  5. Baddeley AD, Logie R. Working memory: The multicomponent model. In: Mijake A, Shah P, editors. Models of Working Memory: Mechanisms of Active Maintenance and Executive Control. Cambridge, MA: Cambridge University Press; 1999. pp. 28–62. [Google Scholar]
  6. Barkley RA. Behavioral inhibition, sustained attention, and executive functions: Constructing a unifying theory or ADHD. Psychological Bulletin. 1997;121:65–94. doi: 10.1037/0033-2909.121.1.65. [DOI] [PubMed] [Google Scholar]
  7. Barkley RA. The executive functions and self-regulation: An evolutionary neuropsychological perspective. Neuropsychological Review. 2001;11:1–29. doi: 10.1023/a:1009085417776. [DOI] [PubMed] [Google Scholar]
  8. Bechara A, Damasio AR, Damasio H, Anderson SW. Insensitivity to future consequences following damage to human prefrontal cortex. Cognition. 1994;50:7–15. doi: 10.1016/0010-0277(94)90018-3. [DOI] [PubMed] [Google Scholar]
  9. Bechara A, Martin EM. Impaired decision making related to working memory deficits in individuals with substance addictions. Neuropsychology. 2004;18:152–162. doi: 10.1037/0894-4105.18.1.152. [DOI] [PubMed] [Google Scholar]
  10. Bentler PM. Comparative fit indexes in structural models. Psychological Bulletin. 1990;107:238–246. doi: 10.1037/0033-2909.107.2.238. [DOI] [PubMed] [Google Scholar]
  11. Bettencourt BA, Talley A, Benjamin AJ, Valentine J. Personality and aggressive behavior under provoking and neutral conditions: A meta-analytic review. Psychological Bulletin. 2006;132:751–777. doi: 10.1037/0033-2909.132.5.751. [DOI] [PubMed] [Google Scholar]
  12. Bogg T, Roberts BW. Conscientiousness and health-related behaviors: A meta-analysis of the leading behavioral contributors to mortality. Psychological Bulletin. 2004;130:887–919. doi: 10.1037/0033-2909.130.6.887. [DOI] [PubMed] [Google Scholar]
  13. Brown J. Some tests of the decay of immediate memory. Quarterly Journal of Experimental Psychology. 1958;10:12–21. [Google Scholar]
  14. Browne MW, Cudeck R. Alternative ways of assessing model fit. In: Bollen KA, Long JS, editors. Testing structural equation models. Newbury Park, CA: Sage; 1993. pp. 136–162. [Google Scholar]
  15. Bucholz K, Cadoret R, Cloninger CR, Dinwiddie S, Hasselbrock V, Nurnberger J, Reich T, Schmit I, Schuckit M. A new semistructured psychiatric interview for use in genetic linkage studies: A report of the reliability of the SSAGA. Journal of Studies on Alcohol. 1994;55:149–158. doi: 10.15288/jsa.1994.55.149. [DOI] [PubMed] [Google Scholar]
  16. Clark LA, Watson D. Temperament: A new paradigm for trait psychology. In: Pervin L, John O, editors. Handbook of personality: Theory and research. 2. New York: Guilford Press; 1999. pp. 399–423. [Google Scholar]
  17. Conway ARA, Engle RW. Working memory and retrieval: A resource-dependent inhibition model. Journal of Experimental Psychology: General. 1994;123:354–373. doi: 10.1037//0096-3445.123.4.354. [DOI] [PubMed] [Google Scholar]
  18. Cowan N. An embedded-process model of working memory. In: Mijake A, Shah P, editors. Models of Working Memory: Mechanisms of Active Maintenance and Executive Control. Cambridge, MA: Cambridge University Press; 1999. pp. 62–101. [Google Scholar]
  19. Donovan JE, Jessor R, Costa FM. Adolescent health behavior and conventionality–unconventionality: An extension of problem-behavior therapy. Health Psychology. 1991;10:52–61. [PubMed] [Google Scholar]
  20. Elkins IJ, King SM, McGue M, Iacono WG. Personality traits and the development of nicotine, alcohol, and illicit drug disorders: Prospective links from adolescence to young adulthood. Journal of Abnormal Psychology. 2006;115:26–39. doi: 10.1037/0021-843X.115.1.26. [DOI] [PubMed] [Google Scholar]
  21. Engle RW, Tuholski SW, Laughlin JE, Conway ARA. Working memory, short-term memory, and general fluid intelligence: A latent variable approach. Journal of Experimental Psychology: General. 1999;128:309–331. doi: 10.1037//0096-3445.128.3.309. [DOI] [PubMed] [Google Scholar]
  22. Eysenck SBG, Eysenck HJ. Impulsiveness and venturesomeness: Their position in a dimensional system of personality description. Psychological Reports. 1978;43:1247–1255. doi: 10.2466/pr0.1978.43.3f.1247. [DOI] [PubMed] [Google Scholar]
  23. Finn PR. Motivation, working memory, and decision making: A cognitive-motivational theory of personality vulnerability to alcoholism. Behavioral and Cognitive Neuroscience Review. 2002;1:183–205. doi: 10.1177/1534582302001003001. [DOI] [PubMed] [Google Scholar]
  24. Finn PR, Hall J. Cognitive ability and risk for alcoholism: short-term memory capacity and intelligence moderate personality risk for alcohol problems. Journal Abnormal Psychology. 2004;113:569–581. doi: 10.1037/0021-843X.113.4.569. [DOI] [PubMed] [Google Scholar]
  25. Finn PR, Justus A, Mazas C, Steinmetz JE. Working memory, executive processes, and the effects of alcohol on go/no-go learning: Testing a model of behavioral regulation and impulsivity. Psychopharmacology. 1999;146:465–472. doi: 10.1007/pl00005492. [DOI] [PubMed] [Google Scholar]
  26. Finn PR, Mazas C, Justus A, Steinmetz JE. Early-onset alcoholism with conduct disorder: Go/No-Go learning deficits, working memory capacity, and personality. Alcoholism: Clinical and Experimental Research. 2002;26:186–206. [PubMed] [Google Scholar]
  27. Finn PR, Sharkansky EJ, Brandt KM, Turcotte N. The effects of familial risk, personality, and expectancies on alcohol use and abuse. Journal of Abnormal Psychology. 2000;109:122–133. doi: 10.1037//0021-843x.109.1.122. [DOI] [PubMed] [Google Scholar]
  28. Giancola PR, Zeichner A, Yarnell JE, Dickenson KE. Relation between executive functioning and the adverse consequence of alcohol use in social drinkers. Alcoholism: Clinical and Experimental Research. 1996;20:1094–1098. doi: 10.1111/j.1530-0277.1996.tb01952.x. [DOI] [PubMed] [Google Scholar]
  29. Gough HG. Manual for the California Psychological Inventory. Palo Alto, CA: Consulting Psychologists Press; 1969. [Google Scholar]
  30. Gough HG. Theory, development, and interpretation of the CPI Socialization scale. Psychological Reports. 1994;75:651–700. doi: 10.2466/pr0.1994.75.1.651. [DOI] [PubMed] [Google Scholar]
  31. Harden PW, Pihl RO. Cognitive function, cardiovascular reactivity, and behavior in boys at high risk for alcoholism. Journal of Abnormal Psychology. 1995;104:94–103. doi: 10.1037//0021-843x.104.1.94. [DOI] [PubMed] [Google Scholar]
  32. Hathaway SR, McKinley JC. MMPI-2: Minnesota Multiphasic Personality Inventory. Minneapolis: The University of Minnesota Press; 1989. [Google Scholar]
  33. Hinson JM, Jameson TL, Whitney P. Impulsive decision making and working memory. Journal of Experimental Psychology: Learning Memory and Cognition. 2003;29:298–306. doi: 10.1037/0278-7393.29.2.298. [DOI] [PubMed] [Google Scholar]
  34. Iacono WG, Carlson SR, Taylor J, Elkins IJ, McGue M. Behavioral disinhibition and the development of substance-use disorders: Findings from the Minnesota Twin Family Study. Development and Psychopathology. 1999;11:869–900. doi: 10.1017/s0954579499002369. [DOI] [PubMed] [Google Scholar]
  35. Justus A, Finn PR, Steinmetz JE. The influence of traits of disinhibition on the association between alcohol use and risky sexual behavior. Alcoholism: Clinical and Experimental Research. 2000;24:1028–1035. [PubMed] [Google Scholar]
  36. Kendler KS, Prescott CA, Myers J, Neale MC. The structure of genetic and environmental risk factors for common psychiatric and substance use disorders in men and women. Archives of General Psychiatry. 2003;60:929–937. doi: 10.1001/archpsyc.60.9.929. [DOI] [PubMed] [Google Scholar]
  37. Kimberg DY, Farah MJ. A unified account of cognitive impairments following frontal lobe damage: The role of working memory in complex, organized behavior. Journal of Experimental Psychology: General. 1993;122:411–428. doi: 10.1037//0096-3445.122.4.411. [DOI] [PubMed] [Google Scholar]
  38. Koopmans JR, Boomsma DI, Heath AC, van Doornen LJP. A multivariate genetic analysis of sensation seeking. Behavior Genetics. 1995;25:349–356. doi: 10.1007/BF02197284. [DOI] [PubMed] [Google Scholar]
  39. Krueger RF, Hicks BM, Patrick PJ, Carlson SR, Iacono WG, McGue M. Etiologic connections among substance dependence, antisocial behavior, and personality: Modeling the externalizing spectrum. Journal of Abnormal Psychology. 2002;111:411–424. [PubMed] [Google Scholar]
  40. Krueger RF, Markon KE. Reinterpreting comorbidity: A model-based approach to understanding and classifying psychopathology. Annual Review of Clinical Psychology. 2006;2:111–133. doi: 10.1146/annurev.clinpsy.2.022305.095213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Krueger RF, Markon KE, Patrick PJ, Benning SD, Kramer MD. Linking antisocial behavior, substance use, and personality: An integrative quantitative model of the adult externalizing spectrum. Journal of Abnormal Psychology. 2007;116:645–666. doi: 10.1037/0021-843X.116.4.645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Krueger RF, Markon KE, Patrick PJ, Iacono WG. Externalizing psychopathology in adulthood: A developmental-spectrum conceptualization and its implications for DSM-V. Journal of Abnormal Psychology. 2005;114:537–550. doi: 10.1037/0021-843X.114.4.537. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Lubman DI, Yücel M, Hall WD. Substance use and the adolescent brain: A toxic combination? Journal of Psychopharmacology. 2007;21:792–794. doi: 10.1177/0269881107078309. [DOI] [PubMed] [Google Scholar]
  44. Markon KE, Krueger RF. Categorical and continuous models of liability to externalizing disorders. Archives of General Psychiatry. 2005;62:1352–1359. doi: 10.1001/archpsyc.62.12.1352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. McCrae RR, Costa PT. Personality in adulthood: A five-factor theory perspective. 2. New York: Guilford Press; 2003. [Google Scholar]
  46. McGue M, Iacono WG, Krueger RF. The association of early adolescent problem behavior and adult psychopathology: A multivariate behavioral genetic perspective. Behavior Genetics. 2006;36:591–602. doi: 10.1007/s10519-006-9061-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. McGue M, Slutske W, Iacono WG. Personality and substance use disorders: II. Alcoholism versus drug use disorders. Journal of Consulting and Clinical Psychology. 1999;67:394–404. doi: 10.1037//0022-006x.67.3.394. [DOI] [PubMed] [Google Scholar]
  48. Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States, 2000. Journal of the American Medical Association. 2004;291:1238–1245. doi: 10.1001/jama.291.10.1238. [DOI] [PubMed] [Google Scholar]
  49. Monti PM, Miranda R, Jr, Nixon K, Sher KJ, Swartzwelder HS, Tapert SF, White A, Crews FT. Adolescence: Booze, brains, and behavior. Alcoholism: Clinical and Experimental Research. 2005;29:207–220. doi: 10.1097/01.alc.0000153551.11000.f3. [DOI] [PubMed] [Google Scholar]
  50. Ohannessian CM, Hesselbrock VM. Paternal alcoholism and youth substance abuse: The indirect effects of negative affect, conduct problems, and risk taking. Journal of Adolescent Health. 2008;42:198–200. doi: 10.1016/j.jadohealth.2007.08.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Oberauer K. Access to information in working memory: Exploring the focus of attention. Journal of Experimental Psychology: Learning, Memory, and Cognition. 2002;28:411–421. [PubMed] [Google Scholar]
  52. Pickering AD, Gray JA. The neuroscience of personality. In: Pervin LA, John OP, editors. Handbook of personality: Theory and Research. 2. New York: The Guilford Press; 1999. pp. 277–299. [Google Scholar]
  53. Pihl RO, Peterson JB, Finn PR. Inherited predisposition to alcoholism: Characteristics of sons of male alcoholics. Journal of Abnormal Psychology. 1990;99:291–301. doi: 10.1037//0021-843x.99.3.291. [DOI] [PubMed] [Google Scholar]
  54. Poon E, Ellis DA, Fitzgerald HE, Zucker RA. Intellectual, cognitive, and academic performance among sons of alcoholics during the early school years: Differences related to subtypes of familial alcoholism. Alcoholism: Clinical and Experimental Research. 2000;24:1020–1027. [PubMed] [Google Scholar]
  55. Raftery AE. Bayesian model selection in social research. Sociological Methodology. 1995;25:111–163. [Google Scholar]
  56. Roberts BW, Bogg T. A longitudinal study of the relationships between conscientiousness and the social environmental factors and substance use behaviors that influence health. Journal of Personality. 2004;72:325–353. doi: 10.1111/j.0022-3506.2004.00264.x. [DOI] [PubMed] [Google Scholar]
  57. Roberts BW, Chernyshenko OS, Stark S, Goldberg LR. The structure of conscientiousness: An empirical investigation based on seven major personality questionnaires. Personnel Psychology. 2005;58:103–139. [Google Scholar]
  58. Slutske WS, Heath AC, Madden PAF, Bucholtz KK, Stathamm DJ, Martin NG. Personality and the genetic risk for alcohol dependence. Journal of Abnormal Psychology. 2002;111:124–133. [PubMed] [Google Scholar]
  59. Stuss DT, Seethem LL, Poirier CA. Comparison of three tests of attention and rapid information processing across six age groups. Clinical Neuropsychology. 1987;1:139–152. [Google Scholar]
  60. Tellegen A. Unpublished manuscript. University of Minnesota; 1982. Brief Manual of the Multidimensional Personality Questionnaire. [Google Scholar]
  61. van den Oord EJCG, Simonoff E, Eaves LJ, Pickles A, Silberg J, Maes H. An evaluation of different approaches for behavior genetic analyses with psychiatric symptom scores. Behavior Genetics. 2000;30:1–18. doi: 10.1023/a:1002095608946. [DOI] [PubMed] [Google Scholar]
  62. Wechsler D. Manual for the Wechsler Adult Intelligence Scale (Rev) New York: The Psychological Corporation; 1981. [Google Scholar]
  63. Wechsler D. Wechsler Adult Intelligence Test-III. Psychological Corporation, Harcourt Brace and Co; 1997. [Google Scholar]
  64. Wechsler D. Wechsler Abbreviated Scale of Intelligence (WASI) Psychological Corporation, Harcourt Brace and Co; 1999. [Google Scholar]
  65. Widom CS. A method for studying noninstitutionalized psychopaths. Journal of Consulting and Clinical Psychology. 1977;45:674–683. [PubMed] [Google Scholar]
  66. Zachary RA. Shipley Institute of Living Scale: Revised Manual. Los Angeles, CA: Western Psychological Services; 1986. [Google Scholar]
  67. Zelazo PD, Frye D. Cognitive complexity and control: II. The development of executive function in childhood. Current Directions in Psychological Science. 1998;7:121–126. [Google Scholar]
  68. Zucker RA, Wong MM, Clark DB, Leonard KE, Schulenberg JE, Cornelius JR, Fitzgerald HE, Homish GG, Merline A, Nigg JT, O'Malley PM, Puttler LI. Predicting risky drinking outcomes longitudinally: What kind of advance notice can we get? Alcoholism: Clinical and Experimental Research. 2006;30:243–252. doi: 10.1111/j.1530-0277.2006.00033.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Zuckerman M. The neurobiology of impulsive sensation seeking: Genetics, brain physiology, biochemistry, and neurology. In: Stough C, editor. Neurobiology of exceptionality. New York: Kluwer Academic/Plenum Publishers; 2005. pp. 31–52. [Google Scholar]
  70. Zuckerman M. Biological bases of personality. In: Millon T, Lerner MJ, editors. Handbook of psychology: Personality and social psychology. Vol. 5. Hoboken, NJ: John Wiley & Sons, Inc; 2003. pp. 85–116. [Google Scholar]
  71. Zuckerman M. Sensation Seeking: Beyond the Optimal Level of Arousal. Hillsdale, NJ: Erlbaum; 1979. [Google Scholar]

RESOURCES