Background: In this study several aspects of attention were studied in 237 nearly 6-year-old twin pairs. Specifically, the ability to sustain attention and inhibition were investigated using a computerized test battery (Amsterdam Neuropsychological Tasks). Furthermore, the Teacher's Report Form (TRF) was filled out by the teacher of the child and the attention subscale of this questionnaire was analyzed. Methods: The variance in performance on the different tasks of the test battery and the score on the attention scale of the TRF were decomposed into a contribution of the additive effects of many genes (A), environmental effects that are shared by twins (C) and unique environmental influences not shared by twins (E) by using data from MZ and DZ twins. Results: The genetic model fitting results showed an effect of A and E for the attention scale of the TRF, and for some of the inhibition and sustained attention measures. For most of the attention variables, however, it was not possible to decide between a model with A and E or a model with C and E. Time-on-task effects on reaction time or number of errors and the delay after making an error did not show familial resemblances. A remarkable finding was that the heritability of the attention scale of the TRF was found to be higher than the heritability of indices that can be considered to be more direct measures of attention, such as mean tempo in the sustained attention task and response speed in the Go-NoGo task. Conclusion: In preschoolers, familial resemblances on sustained attention and inhibition were observed.