Category: Addictive Behaviors

PS1- #A10 - Development and Evaluation of Measures of Implicit Cannabis Associations: Implicit risk and Associations Predict Cannabis Use

Friday, Nov 17
8:30 AM – 9:30 AM
Location: Indigo Ballroom CDGH, Level 2, Indigo Level

Keywords: Addictive Behaviors | Cognitive Biases / Distortions | Health Psychology

Dual process models suggest that both fast/reflexive/impulsive (implicit) and slow/reflective/deliberate (explicit) cognitive processes are important in substance use behaviors. These models have been most commonly applied to alcohol and tobacco, and evidence suggests that both implicit and explicit cognitions uniquely predict alcohol and tobacco use and misuse. Application to cannabis, including the development of implicit measures related to cannabis, is scant. The aims of this study were, therefore, to develop and test two implicit measures of cannabis cognitions: one that evaluated the strength of associations with cannabis and risk and the other with cannabis and identity. Their utility as predictors of cannabis use cross-sectionally and prospectively was investigated. A community sample of 380 US young adults (54% female, all between the ages of 18-24) who live in a Western state with legal recreational cannabis use and who also reported using alcohol at least once in the past year, completed measures of cannabis use and Implicit Association Tests (IATs) evaluating implicit cannabis risk and cannabis identity. Measures of cannabis use were completed at 3-month (n= 299) and 6-month (n = 313) follow up assessments. At baseline, 75% of participants reported using cannabis in some form at least once in their lifetime; 54% reported using cannabis in some form at least once in the last 30 days. Both IATs had good internal consistencies and were correlated with cannabis use, both cross-sectionally and at 3- and 6-month follow up assessments. When evaluated simultaneously in regression models, cannabis risk and cannabis identity IAT scores were significantly and positively associated with cannabis use. Preliminary results from models evaluating use longitudinally indicated greater support for cannabis risk, versus cannabis identity, IAT scores as predictors of use. Collectively, these findings provide preliminary evidence for the validity of the cannabis risk and cannabis identity IATs. They further suggest that implicit cannabis cognitions, particularly cannabis risk cognitions, may be important risk factors for cannabis use and misuse, and ultimately, that they may be targets for intervention efforts.

Kristen P. Lindgren

University of Washington

Cecilia C. Olin

Research Coordinator
University of Washington
Seattle, Washington

Jason J. Ramirez

University of Washington

Devon Abdallah

University of Washington

Christine Lee

University of Washington