Testing a multistage model of risk factors for cannabis use utilizing a measurement burst design - Cannabis use disorder (CUD) is most prevalent in young adulthood (ages 18-25), with 47% of CUDs developing during this period. Alarmingly, the proportion of young adults who use cannabis on a daily basis nearly doubled from 2013-2021, which is linked to a 17-fold increase in risk for CUD. Combined with the paucity of interventions for CUD, this highlights the urgent need for CUD research among young adults. Examining event-level risk processes clarifies the temporal order of risk factors and cannabis use (CU) as they unfold in near-real time. However, ecological momentary assessment (EMA) studies, including our own, provide mixed evidence for two sets of risk factors for CU: risk factors related to increasing positive affect (positive reinforcement; e.g., enhancement motives) and decreasing negative affect (negative reinforcement; e.g., coping motives). Providing a potential explanation for mixed evidence, our preliminary results and the multistage model of substance use suggest these two sets of risk factors may be relevant at different stages of CUD, with positive reinforcement most relevant when CUD symptoms are absent to mild and negative reinforcement becoming prominent when CUD symptoms become more severe. We propose to test the multistage model using a measurement burst design, with 5 semi-annual bursts of 14-days of EMA (2 surveys per day), allowing us to examine changes in event-level associations between risk factors and CU as a function of longer-term (i.e., semi-annual) changes in CUD symptoms. While risk processes discussed previously unfold at the event-level, these processes contribute to broader patterns of CU frequency. These patterns of change in CU result in trajectories of CU frequency, which vary across individuals. For example, some individuals may start with a low CU frequency that increases over time, while others may maintain a high CU frequency over time. These patterns of CU frequency are theorized to be predictive of the development of CUD. Therefore, we propose to use semi-annual longitudinal data to identify multiple trajectories of CU frequency and examine differences across trajectories in the likelihood of having a CUD diagnosis at the final wave of data collection. This will help to inform preventive efforts by identifying individuals likely to develop CUD. The purpose of this R01 is to advance our understanding of CUD, with a focus on how event-level associations between risk factors and CU change as CUD develops. We will use a measurement burst design, with 600 young adults (age 18-25) who use cannabis at least four times in the past month at the screening survey. The proposed study will accomplish three specific aims: 1) identify changes in the effects of positive reinforcement event-level risk factors on CU as CUD develops, 2) test for changes in the effects of negative reinforcement event-level risk factors on CU as CUD develops, and 3) examine associations between trajectories of CUD symptoms over a 2 year period and the likelihood of having a CUD at the final wave of data collection. The proposed study will inform efforts to increase efficacy of existing CUD interventions.