Dijkerman & de Haan (D&dH) propose a model that takes into account the role of the somatosensory system in perception and action, in analogy with the dual-route model of the visual cortical system. In our commentary, we address two issues that are not developed enough in the target article.

One of the core aspects of D&dH's model is that, differently from the dual-route model of the visual system, it distinguishes between internal and external stimuli, that is, between somatosensory processing of one's own body and of external objects. First, despite the target article's wide review of literature on body image, we believe the model does not clearly address the distinctions among external objects and external body parts, and internal bodies as regards body image. Second, the target article should have discussed further the role played by body schema. The authors do not sufficiently underline the dynamic character of body schema. In addition, even though they admit that the neural correlates of visual representation of the observers' own body has received scarce attention, they do not address the behavioral and neural literature focusing on the role of one's own as well as others' perspective in body schema. Finally, we think that D&dH should take into account the role of the body schema in the imitation of action. Here, we consider these two aspects in turn.

D&dH do not clearly discuss how the bodies of others are processed: Are they considered as external objects or as external bodies? For example, their model is not fully able to explain how tactile recognition of others' body parts takes place. This is an important issue, because both neuropsychological and experimental studies on body image suggest that a common supramodal representation is used for representing the bodies of others, as well as one's own body (Bosbach et al. Reference Bosbach, Knoblich, Reed, Cole and Prinz2006; Buxbaum & Coslett Reference Buxbaum and Coslett2001). This is suggested, for example, by the inability of autotopagnosic patients to locate body parts on their own body, on another's body, or on a mannequins' body (Buxbaum & Coslett Reference Buxbaum and Coslett2001; Ogden Reference Ogden1985; Semenza Reference Semenza1988; Sirigu et al. Reference Sirigu, Grafman, Bressler and Sunderland1991), and by the results of a test with healthy participants by Reed and Farah (Reference Reed and Farah1995). A possible way for explaining how this recognition takes place might be to assume that an external hand is tactilely recognized and activates its corresponding representation in semantics (as suggested by the link between the tactile object-recognition module and the semantics module, which also includes the semantic knowledge about the body). However, because of the lack of a direct link between semantics and the internal body image, it is difficult to understand how another body's hand can, for example, be recognized as being a right or a left hand. D&dH should at least add a link between the internal body image and semantics. In this regard, a clarification is needed concerning terminology. The authors distinguish between body schema, body image, and some form of semantic and conceptual representation of the body. However, in the recent literature a more clear terminology is proposed. For example, Schwoebel et al. (Reference Schwoebel, Buxbaum and Coslett2004) distinguish between body schema, body image, or body semantics, and body structural description. In sum, even though D&dH discuss the existence of a semantic and conceptual representation of the body, they do not relate body image to semantics in their model.

The authors do not discuss enough the role of body schema as far as three aspects are concerned: its dynamic character, the role of perspective, and the role of body schema in imitation. The dynamicity of body schema is clearly demonstrated by the neural and behavioral literature focusing on the enlargement of the body schema on account of tool use (for a review, see Maravita & Iriki Reference Maravita and Iriki2004). We believe that this interaction between the body and objects conceived of as extensions of the body should be at least briefly considered.

Moreover, the target article does not develop enough the role played by perspective-taking. The authors dismiss the literature in this field by arguing that there is ample evidence for egocentric coding of external targets, and that the neural correlates of visual representation of the observers' own body have received less attention. However, the role of perspective is an important issue: Perspective is one of the visual characteristics that enables us to distinguish our own body from the body of someone else. According to the in-use terminology, “egocentric” refers to the perspective consistent with looking at one's own body, whereas “allocentric” refers to the perspective that is consistent only with looking at someone else's body (Saxe et al. Reference Saxe, Jamal and Powell2006). Both neural and behavioral studies have provided evidence for distinct representations for self and others with respect to visuo-spatial perspective taking. Much neural evidence confirms our sensitivity to action perspective (Carey et al. Reference Carey, Perrett, Oram, Boller and Grafman1997; David et al. Reference David, Bewernick, Cohen, Newen, Lux, Fink, Shah and Vogeley2006; Jackson et al. Reference Jackson, Meltzoff and Decety2006; Knoblich Reference Knoblich2002; Ruby & Decety Reference Ruby and Decety2001). On the behavioral side, evidence of the role of perspective is more scarce. However, Vogt et al. (Reference Vogt, Taylor and Hopkins2003) suggest the existence of two different priming effects: a visuomotor priming effect, driven by a visual stimulus (a hand) that automatically evokes a motor response, and a motor visual priming, driven by planning, which enhances the visual processing of body parts in the egocentric perspective.

Finally, we think that the authors should at least briefly consider the role of dynamic body schema in action imitation, as shown by recent neuropsychological and brain-imaging studies (Buxbaum et al. Reference Buxbaum, Giovanetti and Libon2000; Chaminade et al. Reference Chaminade, Meltzoff and Decety2005; Goldenberg Reference Goldenberg1996). Indeed, there is a common code linking different body parts and their spatial configuration that is independent of different modalities and perspective (Goldenberg & Karnath Reference Goldenberg and Karnath2006).