Purpose: To compare problems of attention in schoolchildren with newly diagnosed idiopathic or cryptogenic epilepsy with those in healthy classmates. Methods: A computerized battery of tasks comprised Reaction Time (RT) measurement, Trail making (Color Trails 1 and 2), Manual Tapping and Steadiness, and a newly developed task of sustained attention (Balloon Piercing). Subjects: Fifty-one children with epilepsy (age 7-16 years) and 48 gender- and age-matched classmates were assessed thrice: within 48 h after diagnosis [before start of antiepileptic drugs (AEDs)], and 3 and 12 months later. Significantly more children with epilepsy (51%) than control children (27%) had required special educational assistance at school. Results: Children with epilepsy could not be distinguished from controls in execution times or motor speed. However, errors were more frequent among patients in a "go-no-go" RT task, and errors of omission in a task requiring sustained attention. Within the group of children with epilepsy, those with prior school or behavior difficulties and those whose parents reacted maladaptively to the onset and diagnosis of epilepsy performed worse than those without these adversities, in the sense that their RT increased inordinately with increasing task difficulty. Epilepsy-related variables did not explain any variance. Transient inordinately poor performances were found in 69% of patients and 40% of controls. Conclusions: Children with newly diagnosed "epilepsy only" do not have persistent attention deficits. AED treatment has no detrimental effect on attention. Prior school and behavior difficulties and a maladaptive reaction to the onset of epilepsy rather than epilepsy variables are related to decreased attentional efficiency.