An fMRI study in stimulant users showed attention bias to drug-related words: addicted individuals, but not controls, showed more attention bias to drug-related words (measured as the median response latency of correctly identified colours of drug-related words minus the median response latency of correctly identified colours of matched neutral words), which was correlated with enhanced left ventral PFC responses. Such responses were not observed for the colour–word Stroop task126. Similarly, drug-related pictures amplified dACC responses to task-relevant information in cigarette smokers127. These findings suggest that in addiction, more top-down resources are needed to focus on cognitive tasks when drug-related cues are present as distractors (thus biasing attention) during the task. Conflicting with these and other results128 are studies in current cocaine users, in which drug-related words were not associated with slower performance or more errors83,129. This disparity could be related to task design or the treatment-seeking status of the study participants; we predict that enhanced conflict between drug-related words and neutral words characterizes those individuals who are trying to abstain from drugs. Evidence for such an effect in cigarette smokers was recently published130.
