Do 9-month-old infants motorically simulate actions they perceive others perform? Two experiments tested whether action observation, like overt reaching, is sufficient to elicit the Piagetian A-not-B error. Infants recovered a toy hidden at location A or observed an experimenter recover the toy. After the toy was hidden at location B, infants in both conditions perseverated in reaching to A, demonstrating that active search by the infant is not necessary for the A-not-B error. Consistent with prior research, infants displayed an ipsilateral bias when reaching, the so-called "mysterious midline barrier." A similar ipsilateral bias was also observed depending on the manner in which "the experimenter" reached; infants perseverated following observation of ipsi- but not contralateral reaches by the experimenter. Thus, infants perseverated only following observation of actions they themselves were able to perform, suggesting that they coded others' actions in terms of motor simulation. (Contains 3 figures, 1 table, and 7 footnotes.)