The various advantages of using virtual environments for training have been emphasized in previous chapters. In addition, studies primarily concerned with learning have consistently shown the bene–ts of multisensory learning (Shams and Seitz, 2008). Because human interaction with the real world often, if not always, involves multisensory stimulation, it seems reasonable to argue that multimodal interfaces in virtual environment training must be the rule. However, in the process of designing multimodal interfaces for training, we must bear the importance of evaluating the signi–cance of the individual modalities so that we can maximize the effectiveness of the multimodal virtual environment for training. In plain terms, we should use each modality for what it does best.