However, to date, only few studies have examined personality factors that distinguish marijuana from nicotine users and even fewer differentiate isolated use from combined use. In terms of isolated use, high openness but lower agreeableness and conscientiousness in marijuana users relative to non-users has been noted , suggesting that marijuana users differ from non-users on dimensions of normal personality traits as measured by the Big Five model of personality. Conversely, greater extraversion is widely reported in nicotine-only users , as well as high neuroticism and impulsivity. Studies that have performed direct contrasts between isolated marijuana and nicotine users have also shown differences between the two groups. For example, using the wide spectrum Five-Factor Model of personality, Terracciano et al. showed that nicotine users had lower conscientiousness and higher neuroticism whereas marijuana users had high openness, average neuroticism, and low agreeableness and conscientiousness. However, these studies did not examine personality factors in co-morbid nicotine and marijuana users. These traits together suggest that co-morbid users would have a personality profile endorsing high openness and neuroticism, but comparatively less of these traits than isolated users. Personality factors are markers that can be used as endopheno types for substance use disorders particularly because brain circuits involved in personality traits are also implicated in SUD. For example, emergent literature has classified the Big Five personality model via machine learning techniques from resting state fMRI data. These studies indicate that neuroticism negatively correlated with activity in the middle frontal gyrus and precuneus; extraversion correlated positively with regional activity in the striatum, precuneus, and superior frontal gyrus; openness correlated positively with activity in the thalamus and amygdala, indoor plant table and negatively with the superior frontal gyrus; conscientiousness correlated positively with regional activity of the middle frontal gyrus and correlated negatively with the cerebellum. While these findings have not been consistent across studies, they suggest underlying neurobiological mechanisms/pathways that confer personality factors particularly in similar neural substrates implicated in SUD.
Altogether, better understanding of the links between personality and SUD can provide understanding of the brain circuits implicated in SUD that could improve prevention and intervention. Given the paucity in the literature on personality factors that discriminate co-morbid from isolated marijuana and nicotine use, this study examined differential NEO personality profiles in marijuana only, nicotine only, co-morbid marijuana and nicotine use and non-using controls. Because the existing literature has shown that marijuana users and nicotine users differ on openness and neuroticism, we predict that comorbid users would have a personality profile high on these two personality traits, but intermediate to that of the isolated users.Participants were recruited from the general community through flyers and newspaper advertisements to participate in a study to determine behavioral and neural associations of substances at the Mind Research Network in Albuquerque, New Mexico. All participants were between the ages of 18 55, without current Axis I disorders, not currently taking any psychotropic medications, and, have no history of brain injury. Because these data were collected as part of a larger fMRI study, participants were further required to be free of MRI contraindications and be right-handed. Of the 224 individuals who met study criteria, 80 participants were excluded for having a lifetime substance use disorder other than marijuana and nicotine. Two participants were also excluded due to missing data. Thus, analyses for this study were conducted on a sample size of 142. We then categorized the participants into four groups based on their primary and regularly-used substance: marijuana-only , nicotine-only , co-morbid marijuana and nicotine and non-using control groups. For the marijuana only group, regular marijuana use was defined as at least four times a week for the previous six months. For the nicotine-only group, regular nicotine use was defined as smoking at least ten cigarettes per day. The combined marijuana and nicotine group consisted of those who use both marijuana and nicotine regularly, as defined by 60 days out of the past 90 of concurrent use. The non-using control group consisted of participants that were neither regular users of marijuana or nicotine. Table 1 summarizes the substance use characteristics for all of the groups.All assessments took place in the laboratory and were not time constrained. The typical length of time that participants took to complete questionnaires was 1.5 h.
The study outcome variables consisted of groups of marijuana only users, nicotine-only users, and, co-morbid users of marijuana and nicotine. Co-variates included sex, race , age, and years of education. Marijuana, nicotine and alcohol use was evaluated using the participant’s self-reported use on the Timeline Follow back as well as the Marijuana History Questionnaire. Substance use disorders were evaluated using the Structured Clinical Interview for DSM-IV Disorders. Personality traits were measured by NEO PI-R. The NEO PI-R scale is based on the theory of “The Big Five Factors” and assesses personality from a dimensional point of view. Because we were interested in patterns of personality traits that distinguish co-morbid users of both marijuana and nicotine from the marijuana-only or nicotine-only users, we first used a factor analysis with VARIMAX orthogonal rotation method to identify unique relationships between each NEO personality dimension variable and unobserved latent factors. This method allowed us to combine these five factors into linear models that we could then test in linear regression given our sample size. The logistic regression model was used to estimate how much the two significant personality factors derived from factor analysis discriminate marijuana-only, nicotine-only, co-morbid marijuana and nicotine users, and, controls controlling for co-variates of sex, age, race, and education. Odds ratios and 95% confidence interval estimates were presented as results. Further, in order to test accuracy in personality factors’ discriminability of groups who use marijuana only from nicotine only and neither marijuana nor nicotine use, 70% of area under the Receiver Operating Characteristic curves was set as a minimum value for an accuracy of classification. In addition to personality factor patterns as predictors, we also examined the group variance in each NEO personality dimension using Analysis of Variance. ANOVA was run for all five dimensions separately controlling for co-variates of sex, race, age, and education. Post-hoc pairwise comparisons between two diagnosis groups were conducted if an overall group effect was statistically significant and Tukey-Kramer adjusted p values were then reported. SAS 9.4 version was used for all statistical analyses and p value less than 0.05 was set as a statistical significance level.
To investigate the discriminatory effectiveness of personality factors of marijuana use among four groups , we used three paired comparisons: marijuana-only vs. all other groups, marijuana-only vs. nicotine-only, and marijuana only vs. controls. A logistic regression model consisted of two personality factors and co-variates of sex, race, age, and education. Logistic regression results showed that as personality factor 2 score increased by 1, the odds of using marijuana increased by 180% compared to non-marijuana using groups, respectively. However, personality factor 1 did not significantly discriminate marijuana group from the other groups. Regarding demographic variables, these comparisons also demonstrated that older participants were less likely to use marijuana, and that females had significantly lower odds than males of using marijuana only compared to neither marijuana nor nicotine use. Additionally, as years of education increased by 1 year, the odds of marijuana use decreased by 74%. The study model was able to discriminate marijuana users from the rest with 83% accuracy, while each personality factor alone discriminated marijuana users from the rest of the sample with less than 60% accuracy. When comparing marijuana-only and nicotine-only groups, neither factor 1 nor factor 2 alone discriminated marijuana-only users from nicotine-only users with greater than 70% accuracy. However, a model with both personality factors and demographic covariates discriminated marijuana only use from nicotine users with 80% accuracy. When comparing the marijuana-only group with the control group, neither factor 1 nor factor 2 alone discriminated marijuana only users from the control group, while a model with both personality factors and demographic covariates discriminated marijuana only use from the control group with 85% accuracy. While no significant overall group effect was found in any of NEO agreeableness, neuroticism, extraversion, or conscientiousness t scores, ANOVA analysis specifically differentiated the marijuana-use group from the NIC and CON groups in the NEO openness dimension. Post-hoc pairwise group comparison showed that there was a difference between MJ and NIC groups, Tukey adjusted p¼0.0048, and a trend between MJ and Control groups, Tukey adjusted p¼0.0544.Together,plant growing stand these observations indicate clear personality differences between individuals who use marijuana, nicotine, or both, and implicate differences in treatment between these populations.
The aim of this study was to determine personality profiles that distinguish marijuana users, nicotine users, and, co-morbid marijuana-nicotine users. Factor analysis showed that a model with both personality and demographic factors discriminated marijuana users from non-users better than personality factors alone. Logistic regression found strong effects of age, sex and years of education in discriminating marijuana users from non-users whereby the odds of using marijuana increased by being male, younger and less educated. A model with both personality and demographic factors also discriminated marijuana users from nicotine users with high predictability. ANOVA results showed that the openness dimension discriminated the marijuana users from all other groups and the marijuana-nicotine group from the nicotine users. These findings suggest that the discriminability of the co-morbid group from the nicotine-only and non-using group is primarily due to the contributions of marijuana use. The larger contribution of marijuana in concomitant users in terms of risk is in line with studies that found that the association between comorbid use and negative emotion is largely driven by marijuana users. Our findings are concordant with the literature that showed openness discriminates marijuana users from other groups. Openness identifies the seeking of experiences for their own sake. Relative to marijuana, nicotine use does not have the burden of legal consequences and therefore may not require the same degree of openness as marijuana use. Our findings are also concordant with those suggesting that agreeableness and conscientiousness are lower in drug users. Our findings further add to this literature by showing that agreeableness and conscientiousness in marijuana users are intermediate to that of nonusing controls and nicotine users. Interestingly, studies have also reported that extraversion is traditionally lower in drug users compared to non-drug users suggesting that extraversion may confer resilience to the development of addiction disorders in general. However, in marijuana users, the inverse effect has been observed. For instance, in a study by Terracciano et al. , marijuana users exhibited greater “excitement-seeking” and “activity” compared to non-users.
Reports of marijuana use may also be more highly associated with extraversion due to the more communicative personality traits implicated in admission of illegal activity. However, in line with the studies mentioned above, the more robust effect is likely the similarly of the extraversion “excitement-seeking” facet, and sensation-seeking, which is a well established risk factor for drug use.Together with our findings of greater openness in marijuana users, we speculate potential specificity of personality traits in different substance using population. This underlines the need to characterize substance-using populations relative to each other. Additionally, given the widely accepted association between novelty-seeking and sensation seeking in marijuana users , it may be through extraversion that these risk factors lead to substance use disorders. Thus, greater sensation seeking in drug abusers would suggest that extraversion could be a risk factor as others may suggest in other SUDs. In addition to openness, other factors that discriminated marijuana users from nicotine users in our study have been linked to neurological systems responsible for appetitive-approach behaviors. Recent imaging work has shown specific brain mechanisms associated with similar traits including motivation, optimism, and enthusiasm. For example, DeYoung et al. showed that extraversion scores covaried with medial orbitofrontal cortex volume, which has also been shown to be altered in long-term marijuana users. Agreeableness was positively associated with retrosplenial PCC , which is implicated in altered sensory awareness in addiction. Similarly, genetic mechanisms have also been associated with personality traits. For instance, COMT Val158Met allele has been posited to modulate extraversion. Together, these results suggest potential mechanisms for the association between these personality traits and marijuana use.