, 2011), foveal position (Hasson et al., 2002), and high spatial frequencies (Woodhead et al., 2011). How dependent is VWFA selectivity

on such visual sensory features? It was recently shown (Rauschecker et al., 2011) that reading activates VWFA even when the shape of the letters is derived from atypical features such as the movement or the luminance of sets of dots. This suggests that within vision there is remarkable feature tolerance. Here we explored whether this tolerance of VWFA activation for reading can be generalized as far as to reading in a new nonvisual sensory modality, and further, when reading in this novel modality is learned in adulthood, well after reading skills are usually acquired. We took advantage of a unique setup to probe these questions and several other related key issues. We studied a group of congenitally fully blind adults trained to read through a visual-to-auditory sensory Selleckchem PI3K inhibitor substitution device (SSD; Bach-y-Rita and W Kercel, 2003), which converts visual images to auditory “soundscapes” using a predetermined consistent algorithm (The vOICe; Meijer, 1992). This enabled the blind to perceive high-resolution visual information (Striem-Amit et al., 2012b) and, in this case, to learn to read, with sounds

topographically representing visual images of letters (see Figure 1). Moreover, subjects also learned to recognize soundscapes of other visually complex object categories such as faces, houses, and body parts. This ability helped us test a fundamental buy VX-770 question: can the “visual” category selectivity of the VWFA develop in the absence of any visual experience? In sighted subjects, the VWFA is characterized by increased responses to

letters as compared to different visual object categories (such as faces, houses, and objects; Cohen and Dehaene, 2004; Dehaene and Cohen, 2011; Sclareol Dehaene et al., 2010; Hasson et al., 2002; Puce et al., 1996; Szwed et al., 2011; Tsapkini and Rapp, 2010), similar to the preferential activation of the neighboring regions for faces, scenes, objects, and body shapes (Kanwisher, 2010). Can full category selectivity in the VWFA also emerge without visual experience and by using auditory sensory substitution, and if so, what is the basis for such a robust preference pattern? Finally, another key question relates to the developmental origin of the VWFA (and the ventral stream more generally). Even assuming that the VWFA can develop specialization and selectivity for reading in nonvisual modalities, how dependent is such specialization on the age and amount of training? Reading in the visual modality is usually learned in childhood and improves over many years of practice (Aghababian and Nazir, 2000). Recent evidence shows that plasticity of the ventral visual cortex extends into adolescence and beyond (Dehaene et al., 2010; Golarai et al., 2007).