Examining the Reliability and Utility of MouseView.js as a Method to Measure Attentional Bias for Cannabis Use

Abstract

Objective: Current methods of assessing cannabis-related attentional bias such as modified Stroop tasks are characterized by several limitations, including low reliability. The goal of the present study was to explore the reliability and utility of a novel methodological tool that is a proxy for eye-tracking—MouseView.js—to examine cannabis-related attentional bias. Method: Canadian postsecondary students (N = 580) freely viewed 30 image pairs of neutral and cannabis stimuli using MouseView.js. Participants also completed self-report measures of cannabis use, including problematic cannabis use. Reliability coefficients (Cronbach’s alpha and split-half) were calculated to estimate the internal consistency of cannabis images, neutral images, and dwell difference scores. A hierarchical testing strategy was used to examine whether image (neutral vs. cannabis) and cannabis use status (non-use, recreational use, problematic use), as well as the interaction between image and cannabis use status, explain variation in dwell times. Results: A total of 368 participants (64.4%) did not use cannabis, 138 (23.8%) used cannabis recreationally, and 74 (12.8%) used cannabis at problematic levels. The reliability estimates for cannabis images, neutral images, and attentional bias scores ranged from acceptable to excellent. There was a main effect of image, such that all participants spent more time viewing cannabis relative to neutral images, indicative of an attentional bias. The main effects for cannabis use status and interactions between cannabis use status and image type were not statistically significant. Conclusions: Taken together, the present findings suggest MouseView.js may be a reliable method to assess cannabis-related attentional biases.