Data crowdsourcing on psychomotor and cognitive effects of cannabis via mobile app.

In an effort to educate consumers of cannabis, we created a downloadable application (app) for mobile phones that collects data on the neuropsychological effects related to cannabis use. In particular, the app assessed four domains, these being: (i) psychomotor compensation, (ii) time estimation, (iii) sustained attention, and (iv) response inhibition. These tests were presented as a sequence of video games to be completed in under 10 min. Included in the analysis were 213 users who indicated that they were intoxicated from cannabis at the moment of app use. The control group contained individuals who reported using the app while sober ( = 137). A machine learning model was applied to the data to determine whether a particular pattern of performance was predictive of intoxication, and these results were used to inform the creation of a composite score that reflected aggregate performance for all four (i-iv) video games. Relative to the control group, the largest performance decrements were discovered within the initial 120 min after self-administration. These deficits abated as the time-since-use lengthened, and this pattern was consistent with the time-course of subjectively reported intoxication. Although significant limitations in interpretation exist due to the naturalistic and self-report data collection method, this proof-of-concept study points toward the potential utility of mobile app detection of cannabis effect. (PsycInfo Database Record (c) 2023 APA, all rights reserved).