Motor imagery is a mental representation of motor behavior which has been widely used to study the cognitive basis of movement. The assumption that real movements and motor imagery (virtual movements) use the same neurobiological basis has been questioned by functional magnetic resonance data. The functional similarity in the planning of real and virtual movements was studied here by analyzing event-related EEG recordings of the Mu-activity in the sensitive-motor cortex, pre-motor cortex and supplementary motor cortex. A visual stimulus (an arrow) which displayed the information needed for planning a motion (which can be executed or imaged later after the display of a second stimulus) induced a short-lasting phase-locked Mu-response (PLr) which was wider and more widespread when it was used for the motor planning of real or virtual movements than when it was passively watched. The phase-locked Mu-response was accompanied by a persistent decrease of the Mu-rhythms which were not phase-locked to stimuli (NPLr), a response which also was more marked and generalized when stimuli were used for motor planning than when they were passively observed. PLr and NPLr were similar during motor testing and imagery testing, suggesting that both tasks activated the Mu rhythms to a similar degree. This congruency between real and virtual movements was observed in the three cortical areas studied, where the amplitude, latency and duration of the phase-locked and non-phase-locked Mu response was similar in both cases. These noticeable similarities support the idea that the same cortical mechanisms are recruited during the planning of real and virtual movements, a fact that can be analyzed better when an event-related paradigm and a high time-resolution method are used. Copyright © 2013 Elsevier Ltd. All rights reserved.