[Paper Review] Personality Traits and Drug Consumption. A Story Told by Data

Elaine Fehrman, Egan, Vincent+4 more|arXiv (Cornell University)|Jan 17, 2020

Mental Health Research Topics7 references4 citations

TL;DR

This study uses machine learning and a large dataset of 1,885 respondents to show that personality traits (Five Factor Model, impulsivity, sensation-seeking) combined with basic demographics predict individual drug use with over 70% sensitivity and specificity for most substances. It identifies distinct personality profiles for users of different drugs and reveals three clusters of correlated drug use, known as 'correlation pleiades'.

ABSTRACT

This is a preprint version of the first book from the series: "Stories told by data". In this book a story is told about the psychological traits associated with drug consumption. The book includes: - A review of published works on the psychological profiles of drug users. - Analysis of a new original database with information on 1885 respondents and usage of 18 drugs. (Database is available online.) - An introductory description of the data mining and machine learning methods used for the analysis of this dataset. - The demonstration that the personality traits (five factor model, impulsivity, and sensation seeking), together with simple demographic data, give the possibility of predicting the risk of consumption of individual drugs with sensitivity and specificity above 70% for most drugs. - The analysis of correlations of use of different substances and the description of the groups of drugs with correlated use (correlation pleiades). - Proof of significant differences of personality profiles for users of different drugs. This is explicitly proved for benzodiazepines, ecstasy, and heroin. - Tables of personality profiles for users and non-users of 18 substances. The book is aimed at advanced undergraduates or first-year PhD students, as well as researchers and practitioners. No previous knowledge of machine learning, advanced data mining concepts or modern psychology of personality is assumed. For more detailed introduction into statistical methods we recommend several undergraduate textbooks. Familiarity with basic statistics and some experience in the use of probabilities would be helpful as well as some basic technical understanding of psychology.

Motivation & Objective

• To investigate the link between personality traits and drug consumption using empirical data.
• To develop predictive models for individual drug use based on personality and demographic factors.
• To identify distinct psychological profiles associated with users of different psychoactive substances.
• To uncover patterns of correlated drug use through data-driven clustering.
• To create an open-access database and visualization tools (risk maps) for future research.

Proposed method

• Collected a dataset of 1,885 respondents with self-reported use of 18 psychoactive drugs and personality trait assessments.
• Applied multiple data mining techniques including linear discriminant analysis, decision trees, random forests, k-nearest neighbors, and logistic regression.
• Used advanced preprocessing methods such as polychoric correlation, nonlinear CatPCA, sparse PCA, and double Kaiser’s feature selection.
• Employed probability density estimation with radial basis functions and Gaussian mixture models for classification.
• Validated results by excluding spurious correlations and ensuring robustness across multiple algorithms.
• Developed risk map visualizations to represent the probability of drug consumption based on personality and demographic profiles.

Experimental results

Research questions

• RQ1 Can personality traits and demographic data predict the risk of consuming specific drugs with high accuracy?
• RQ2 Are there distinct personality profiles associated with users of different psychoactive substances?
• RQ3 Do patterns of drug use cluster together, indicating shared underlying behavioral or psychological drivers?
• RQ4 How do the personality profiles of users of heroin, ecstasy, and benzodiazepines differ significantly?
• RQ5 What are the most predictive personality traits for the use of individual drugs, and do these vary across substances?

Key findings

• Personality traits (Five Factor Model, impulsivity, sensation-seeking) combined with basic demographics achieved sensitivity and specificity above 70% for predicting use of most of the 18 drugs studied.
• Significant differences in personality profiles were found between users of heroin, ecstasy, and benzodiazepines, particularly in Neuroticism, Extraversion, Agreeableness, and Impulsivity.
• Users of ecstasy showed higher levels of Extraversion, Agreeableness, and Sensation-Seeking compared to heroin users.
• Heroin users exhibited higher Neuroticism and Impulsivity than ecstasy users, while benzodiazepine users showed higher Agreeableness and lower Sensation-Seeking.
• Three distinct clusters of correlated drug use—centered on heroin, ecstasy, and benzodiazepines—were identified, forming 'correlation pleiades' that align with longitudinal patterns of drug involvement.
• An open-access database of 1,885 respondents with detailed drug use and personality data is publicly available for further research.