Decision-making, not impulse control, predicts cannabis use

This groundbreaking study challenges a long-held assumption in addiction research: that poor inhibitory control—the ability to stop unwanted actions—is the primary driver of substance use vulnerability. Using advanced computational modeling on data from over 1,000 participants tracked from ages 19 to 23, researchers found that inhibitory control measures had virtually no connection to cannabis or cigarette use. Instead, they discovered that efficiency of evidence accumulation—how quickly and effectively a person's brain processes decision-making information—emerged as a far more reliable predictor of substance use risk than the traditional inhibitory control measures that addiction researchers have focused on for decades.

The study employed the sophisticated Racing Diffusion Ex-Gaussian ABCD (RDEX-ABCD) model, which separates decision-making into distinct processes rather than treating it as a single measure. Participants with slower evidence accumulation, lower decision thresholds (being more impulsive), and higher rates of decision failures showed stronger associations with cannabis and cigarette use. When researchers mapped these cognitive parameters to brain activity using connectome-based predictive modeling, they found that brain networks predicting decision-making efficiency were robustly connected to substance use, while networks associated with inhibition were not.

These findings have significant implications for understanding and preventing substance use disorders. Rather than targeting inhibitory control training—the focus of many intervention programs—the research suggests interventions should target the underlying decision-making processes that affect how people weigh risks and make choices across various contexts. This represents a fundamental shift in how we think about addiction vulnerability, pointing toward more precise, mechanism-based approaches to prevention and treatment.