ABSTRACT Cannabis is the third most commonly used substance in the United States (US), with almost half of all US adults reporting lifetime use. Cannabis use is associated with physical harms (e.g., respiratory diseases), psychological harms (e.g., cognitive deficits), and other types of harms (e.g., car crashes). Despite these harms, many adults underestimate or do not know about the risks of cannabis use. Warning labels on cannabis packages are a population-level intervention that can increase risk perceptions, knowledge, and recall of cannabis harms, however, there is a lack of research on how cannabis warnings can be improved. Currently, most US states with legalized cannabis require text-only cannabis warnings, but they are often placed on the back of packages, are small (e.g., 6-point font), are wordy (e.g., more than 100 words), and read like a legal disclaimer. As a result, current cannabis warnings are unnoticed, hard to read, and confusing. Accordingly, the long-term goal of this research is to develop cannabis warnings that inform people about cannabis harms and that states can implement into their cannabis warning regulations. The overall objective of this research is to rigorously examine the current landscape of cannabis warning regulations and experimentally determine which warning characteristics most effectively communicate the harms of cannabis use. The central hypothesis is that large cannabis warnings with characteristics found to be promising (e.g., have icons, include colors) will increase risk perceptions of cannabis harms. To accomplish our long-term goal and objective, the proposed study will include three specific aims: Aim 1: Examine the legal and regulatory landscape of US cannabis warnings; Aim 2: Develop a set of evidence-based cannabis warnings and identify which warning characteristics increase perceived warning effectiveness; and Aim 3: Experimentally determine if large cannabis warnings with characteristics found to be promising in Aim 2 increase risk perceptions, knowledge, and recall of cannabis harms. Included within these aims are rigorous and innovative methods: a comprehensive legal analysis, key informant interviews with cannabis regulators, an expert panel review, online experiments, and a discrete choice experiment. Throughout the project, we will disseminate findings from our study to regulators so that they can implement evidence-based warnings and impact change in the cannabis regulatory landscape. This project is significant because each state that legalizes cannabis is tasked with developing regulations on warning labels, but there is a lack of rigorous published research, especially from the US, on the effectiveness of different cannabis warning themes (e.g., content), characteristics (e.g., text length, color) and format (e.g., size). Therefore, our project can help meet a critical need for a comprehensive and rigorous evaluation of cannabis warnings. The proposed study directly addresses NIDA’s p...