IntroductionDrug addiction is a neuropsychiatric disorder characterized by attentional abnormalities whereby enhanced attention is afforded to drugs and drug-related cues at the expense of other (including intrinsically pleasant) reinforcers (e.g., food, sex or money).1,2 This attention bias toward drug-related cues is posited to result from conditioning to drug-related cues, which become excessively and motivationally salient through habitual use. 1,3,4 The underlying mechanism invokes changes to mesolimbic dopamine transmission, associated with an increased incentive salience that is automatically attributed to drug-related cues. 5,6 Ultimately, this drug-related attention bias influences drug-seeking behaviour, which in turn is associated with increased craving 7-9 and relapse susceptibility. [10][11][12][13] Abstinence from drug use improves cognitive and affective functioning, including attentional bias/ dysregulation, in drug-addicted individuals.14 To objectively quantify attention to salient cues, several prior studies have used the late positive potential (LPP), an event-related potential (ERP) component that marks motivated attention to emotionally salient stimuli. 15,16 Specifically, increased LPP amplitude in response to drug-related compared with neutral cues has consistently been shown across all substance use disorders [17][18][19][20][21][22][23] and has further been linked with cue-induced craving.24 Importantly, cross-sectional studies have shown that the LPP amplitude to drug-related cues is decreased 25,26 while that to non-drug-related cues is increased 19,27 after a period of reduced drug use (3 d to about 1 yr). Moreover, prospective neuroimaging studies modelling relapse vulnerability and future drug use have shown that reduced attention toward drug-related cues 11,12,[28][29][30] and increased attention toward non-drug-related reinforcers [31][32][33] predicted longer abstinence durations in individuals with substance use disorders. Although these studies are highly informative, many have been limited to the cross-sectional Background: Increased attention bias toward drug-related cues over non-drug-related intrinsically pleasant reinforcers is a hallmark of drug addiction. In this study we used the late positive potential (LPP) to investigate whether such increased attention bias toward drugrelated relative to non-drug-related cues changes over a protracted period of reduced drug use in treatment-seeking individuals with a cocaine use disorder (CUD). Methods: Treatment-seeking individuals with CUD and matched healthy controls passively viewed a series of pleasant, neutral and drug-related pictures while their event-related potentials were recorded at baseline (≤ 3 weeks after treatment initiation) and at 6-month follow-up (only CUD). Results: We included 19 treatment-seeking individuals with CUD and 18 matched controls in our analyses. The results showed a reversal in attention bias (i.e., LPP amplitude) from baseline (i.e., drug > pleasant) to follow-up (i.e., pleasant > drug) driven by...