This study was carried out to assess the behavioral effects of the non-psychostimulant drug atomoxetine, in rats prenatally-exposed to the organic compound trimethyltin chloride (TMT) and in spontaneously hypertensive rat (SHR), two rodent models of Attention Deficit/Hyperactivity Disorder (ADHD). At birth, neonatal reflexes (righting, cliff aversion, forelimb placing, forelimb grasping, bar holding and startle) had an earlier onset (i.e. percent of appearance) and completion (maximum appearance, i.e. 100% of the brood exhibiting each reflex) in prenatally TMT-exposed and SHR pups as compared to control groups. Two months after birth, TMT-exposed and SHR rats showed impaired cognitive performances in both the step-through passive avoidance test and the shuttle box active avoidance test. Atomoxetine (1, 3 and 6 mg/kg, i.p.), already at the lowest dose tested, improved learning and memory capacity of prenatally TMT-exposed rats and SHR; while methylphenidate (1, 3 and 6 mg/kg, i.p.), used here as positive control, elicited a significant cognitive enhancing effect only at the higher doses. In the open field test, both TMT-exposed rats and SHR displayed enhanced locomotor activity. Methylphenidate further increased locomotor activity in all groups, whereas atomoxetine reduced the enhanced locomotor activity of TMT-exposed rats and SHR down to the level of controls. These results suggest that prenatal TMT-exposure could be considered as a putative experimental model of ADHD and further support the effectiveness of atomoxetine in the ADHD pharmacotherapy. Furthermore, despite the similar effect of the two drugs on cognitive tasks, they exhibit distinct profiles of activity on locomotion, in ADHD models. (C) 2012 Elsevier B.V. All rights reserved.