Objective <p>Alcohol and cannabis co-use is the most common form of polysubstance use in the U.S. Some evidence suggests that there is an association between cannabis use and reduced alcohol consumption, however data on this topic are mixed. Further research is needed to elucidate the individual factors that determine whether cannabis substitutes for alcohol.</p> Method <p>This study is an exploratory analysis of data from a within-subjects, crossover laboratory study on alcohol and cannabis co-administration. Participants completed two sessions: one in which they self-administered alcohol after using cannabis, and one in which they self-administered alcohol but did not use cannabis. We aim to compare behavioral phenotypes of individuals who self-administered less alcohol after cannabis (“substituters”; <i>n</i> = 23), more alcohol after cannabis (“complementers”; <i>n</i> = 7), or the same amount of alcohol regardless of whether cannabis was used (“non-substituters”; <i>n</i> = 16). We utilized Welch’s ANOVAs and Kruskal-Wallis tests to compare baseline group differences, drawing on domains from the Addictions Neuroclinical Assessment (ANA) framework. Wilcoxon tests were employed to test differences in laboratory self-administration measures between sessions for each group.</p> Results <p>Substituters had significantly lower scores on measures corresponding to ANA domains of negative emotionality (BDI-II, DASS-Depression) and executive function (UPPS-P Lack of Premeditation) than non-substituters. Substituters endorsed higher positive cannabis expectancies than non-substituters, though expectancy differences did not reach significance.</p> Conclusions <p>Individuals who use cannabis and alcohol and have lower levels of negative emotionality and impulsivity may be more likely to substitute cannabis for alcohol. Future research is needed to explore long-term outcomes of cannabis substitution.</p> Clinical trials registration number <p>NCT04998006.</p>