How movements and gestures are grounded in our memories? 

There is evidence showing that multimodal encoding—processing information simultaneously through different senses (e.g., vision, hearing, touch) or movements—helps us recall information. Studies indicate that self-performed actions associated with linguistic tasks can support memorization of words and the learning of new word meanings. More specifically, it is easier to remember the sentence “grasp the orange” if we perform the action of grasping an (imaginary) orange ourselves, rather than observing someone else perform the action or simply reading or hearing the sentence.