This review advances the case that bidirectional, cross-species translation of findings from experimental animals to and from humans is an important strategy for drug discovery. Animal models of mental disorders require appropriate behavioural or cognitive outcome variables that can be generalized cross-species. One example is the treatment of impulsive behaviour in attention deficit hyperactivity disorder (ADHD) with stimulant drugs. Performance on the stop signal reaction task as an index of impulsivity is improved both in healthy human volunteers and in patients with adult ADHD by stimulant drugs and also by the selective noradrenaline reuptake blocker atomoxetine. Functional neuroimaging evidence suggests a modulation of circuitry including the inferior prefrontal cortex by this drug. Parallel work in rats had shown that atomoxetine improves stop signal performance by affecting possibly homologous regions of the rodent prefrontal cortex. This parallel effect of atomoxetine in rodents and humans could potentially be exploited in other disorders in which impulsivity plays a role, such as stimulant abuse and Parkinson’s disease. A contrasting relative lack of involvement of 5-HT mechanisms in the stop signal reaction time task will also be described. Research in humans and experimental animals that demonstrate effects of serotoninergic agents such as the selective serotonin (5-HT) reuptake inhibitor citalopram on probabilistic learning and reversal (upon which atomoxetine has little effect) will also be reviewed, possibly relevant to the treatment of clinical depression, Finally, other promising examples of parallel studies of behavioural effects of CNS-active drugs in animals and humans will also be described.