Cannabis intoxication does not impair eyewitness recall with the Sketch Cognitive Interview but increases confidence in lineup errors.

Previous research has shown that cannabis intoxication may increase false memories, thereby affecting the validity of eyewitness testimony. Evidence-based investigative interview methods might counteract these cannabis-induced memory impairments. In this quasi-experimental study, we used the Sketch Reinstatement of Context Cognitive Interview (Sketch-CI) with cannabis-using and non-using participants, examining how the timing of intoxication affected eyewitness memory reports and identification. We tested four groups (N = 131): regular users intoxicated at both encoding and retrieval, regular users intoxicated only at retrieval, regular sober users, and sober non-users. During an online video call, participants viewed a mock crime video, were interviewed using the Sketch-CI and completed three target-absent lineups. The groups did not statistically differ for memory performance, measured by the number of correct and incorrect details, confabulations, recall accuracy, recall completeness, and identification performance. This suggests the potential usefulness of the Sketch-CI for interviewing cannabis-using eyewitnesses. However, stronger levels of subjective intoxication were associated with poorer memory performance, implicating that individual sensitivity to cannabis might determine real-life memory impairment. Participants intoxicated at encoding and retrieval also exhibited inflated confidence in lineup errors. Future studies could replicate these findings using additional retrieval methods and a control condition without external context reinstatement support.