$Rationale:$ Biased attention towards drug-related cues and reduced inhibitory control over the regulation of drug-intake characterize drug addiction. The noradrenaline system has been critically implicated in both attentional and response inhibitory processes and is directly affected by drugs such as cocaine.

$Objectives:$ We examined the potentially beneficial effects of the noradrenaline reuptake inhibitor atomoxetine in improving cognitive control during two tasks that used cocaine- and non-cocaine-related stimuli.

$Methods:$ A double-blind, placebo-controlled, and cross-over psycho-pharmacological design was employed. A single oral dose of atomoxetine (40 mg) was administered to 28 cocaine-dependent individuals (CDIs) and 28 healthy controls. All participants performed a pictorial attentional bias task involving both cocaine- and non-cocaine-related pictures as well as a verbal go/no-go task composed of cocaine- and food-related words.

$Results:$ As expected, CDIs showed attentional bias to cocaine-related cues whilst controls did not. More importantly, however, atomoxetine, relative to placebo, significantly attenuated attentional bias in CDIs ($F26$ = 6.73, $P$ = 0.01). During the go/no-go task, there was a treatment × trial × group interaction, although this finding only showed a trend towards statistical significance ($F26$ = 3.38, $P$ = 0.07).

$Conclusions:$ Our findings suggest that atomoxetine reduces attentional bias to drug-related cues in CDIs. This may result from atomoxetine's modulation of the balance between tonic/phasic activity in the locus coeruleus and the possibly parallel enhancement of noradrenergic neurotransmission within the prefrontal cortex. Studying how cognitive enhancers such as atomoxetine influence key neurocognitive indices in cocaine addiction may help to develop reliable biomarkers for patient stratification in future clinical trials.