Preserved Functional Specialization for Spatial Processing in the Middle Occipital Gyrus of the Early Blind

Transcription

1 Article Preserved Functional Specialization for Spatial Processing in the Middle Occipital Gyrus of the Early Blind Laurent A. Renier, 1,2 Irina Anurova, 1,3 Anne G. De Volder, 2 Synnöve Carlson, 3,4,5 John VanMeter, 6 and Josef P. Rauschecker 1,7, * 1 Laboratory for Integrative Neuroscience and Cognition, Department of Physiology and Biophysics, Georgetown University Medical Center, 3970 Reservoir Road, NW Washington, D.C , USA 2 Institute of Neuroscience, Université Catholique de Louvain, Avenue Hippocrate, 54, UCL-54.46, B-1200 Brussels, Belgium 3 Neuroscience Unit, Institute of Biomedicine/Physiology; University of Helsinki, Helsinki, Finland 4 Brain Research Unit, Low Temperature Laboratory, Aalto University School of Science and Technology, FI AALTO, Finland 5 Medical School, University of Tampere, Tampere, Finland 6 Center for Functional and Molecular Imaging, Georgetown University Medical Center, USA 7 Mind and Brain Laboratory, Centre of Excellence in Computational Complex Systems Research, Aalto University School of Science and Technology, FI AALTO, Finland *Correspondence: rauschej@georgetown.edu DOI /j.neuron SUMMARY The occipital cortex (OC) of early-blind humans is activated during various nonvisual perceptual and cognitive tasks, but little is known about its modular organization. Using functional MRI we tested whether processing of auditory versus tactile and spatial versus nonspatial information was dissociated in the OC of the early blind. No modality-specific OC activation was observed. However, the right middle occipital gyrus (MOG) showed a preference for spatial over nonspatial processing of both auditory and tactile stimuli. Furthermore, MOG activity was correlated with accuracy of individual sound localization performance. In sighted controls, most of extrastriate OC, including the MOG, was deactivated during auditory and tactile conditions, but the right MOG was more activated during spatial than nonspatial visual tasks. Thus, although the sensory modalities driving the neurons in the reorganized OC of blind individuals are altered, the functional specialization of extrastriate cortex is retained regardless of visual experience. INTRODUCTION A number of studies have demonstrated that visual deprivation leads to functional reorganization of deafferented visual cortex (Hyvärinen et al., 1981; Veraart et al., 1990; Rauschecker, 1995; Cohen et al., 1997; Bavelier and Neville, 2002; Merabet and Pascual-Leone, 2010). In early-blind humans, a cross-modal recruitment of several occipital areas has been observed during the processing of both auditory (Gougoux et al., 2005; Röder et al., 2002; Weeks et al., 2000) and tactile stimuli (Kujala et al., 1992; Sadato et al., 1996, 1998; Büchel et al., 1998; Burton et al., 2002, 2004; Hamilton et al., 2004; Ptito et al., 2005; Stilla et al., 2008) as well as during various nonvisual cognitive tasks, including stimulus discrimination and localization, verb generation, and verbal memory (Röder et al., 2001; Röder and Rösler, 2003; Burton et al., 2003 Amedi et al., 2003; Raz et al., 2005). This type of occipital cortex (OC) recruitment may provide an explanation for the enhanced perceptual abilities in the remaining senses commonly observed in visually deprived humans and animals (Pascual-Leone and Torres, 1993; Rauschecker and Korte, 1993; Rauschecker and Kniepert, 1994; Lessard et al., 1998; Gougoux et al., 2004; Lewald, 2002; Goldreich and Kanics, 2003; Voss et al., 2004; Collignon et al., 2006; Alary et al., 2009). Despite the increasing number of studies on cross-modal brain plasticity, little is known about the modular organization of the OC in blind individuals and its exact role in the processing of nonvisual information. In sighted individuals, the visual pathways are functionally divided into ventral/identification ( what ) and dorsal/localization ( where ) streams (Zeki, 1978; Ungerleider and Mishkin, 1982; Haxby et al., 1991; Goodale and Milner, 1992). As demonstrated by animal and human studies, there are also two distinct processing streams within the cerebral cortex for hearing (Rauschecker, 1998; Romanski et al., 1999; Kaas and Hackett, 2000; Rauschecker and Tian, 2000; Tian et al., 2001; Alain et al., 2001; Maeder et al., 2001) and, to a lesser extent, for touch (Deibert et al., 1999; Zangaladze et al., 1999; Stoesz et al., 2003; Prather et al., 2004; Van Boven et al., 2005). To date, there have been very few attempts to test dissociation in the OC of early-blind subjects for auditory and tactile stimulus modalities (Kujala et al., 1995; Röder et al., 1996; Weaver and Stevens, 2007) and for spatial and nonspatial tasks (De Volder et al., 2001; Vanlierde et al., 2003; Bonino et al., 2008). Since the visual cortex of early-blind subjects is widely activated during a variety of perceptual and cognitive tasks, the only way to identify functional specialization is to compare patterns of activation elicited by different conditions in the 138 Neuron 68, , October 7, 2010 ª2010 Elsevier Inc.

2 same subjects in a well-controlled paradigm. In the present study, we used functional magnetic resonance imaging (fmri) in early-blind individuals and sighted controls to compare brain activation patterns elicited by the processing of spatial and nonspatial attributes of equivalent auditory and vibrotactile stimuli. In addition, we ran a control experiment in a separate group of sighted subjects to identify the regions within the OC that are specialized for spatial processing in vision. RESULTS The main experiment was done in two groups of 12 early-blind (EB) volunteers and 12 sighted controls (SCs). Both auditory and vibrotactile stimuli varied in two dimensions: frequency and spatial location. The subjects were instructed to either detect the stimuli (Detection condition) or perform a one-back comparison task where they had to determine whether each stimulus was the same or different from the preceding one regarding either its frequency (Identification condition) or its location (Localization condition). Blind participants were totally blind from birth or by the second year of life (see Table 1). In a control experiment, a separate group of six sighted subjects performed one-back spatial and nonspatial visual tasks comparing either locations (Localization condition) or geometric shapes (Identification condition) of figures presented on a projection screen. Behavioral Results We performed separate analyses of performance accuracy in detection (Figure 1A) and identification/localization (Figures 1B and 1C) because of differences in cognitive demand between these tasks (simple detection versus one-back comparison). Detection scores did not differ between the EB and SC groups but were generally higher for the auditory than the tactile condition (F 1,22 = 18.76; p < 0.001). A three-way ANOVA performed on accuracy in the identification and localization tasks showed no effect of group (F 1,22 = 0.037; p = 0.085) but an effect of the task (F 1,22 = ; p < 0.001) and modality (F 1,22 = ; p < 0.001); tactile identification was more difficult than auditory identification and tactile localization in the two groups (all p values < 0.05). The interaction between task and modality was also significant (F 1,22 = ; p < 0.001). Performance accuracy differed between the two tasks in the tactile but not in the auditory modality. Performance accuracy during the visual conditions in the control study with SC participants was also satisfactory (>95% of accuracy). No difference was observed between the identification and localization conditions in vision (p > 0.05). Functional Imaging Results Main Effects Related to Sensory Modality In both groups, the contrast of Auditory conditions versus Rest showed the expected bilateral activation in the primary and secondary auditory cortices within the superior temporal gyri (Brodmann areas [BAs] 41, 42, 22), whereas the contrast of Tactile conditions versus Rest showed bilateral activation (more pronounced in the left hemisphere contralateral to the stimulated hand) in the primary and secondary somatosensory Table 1. Characteristics of Early-Blind Subjects Subject Sex Age Handedness Cause and onset of blindness EB01 M 45 Right Retinopathy* (before the 2 nd year) Onset of blindness <2 nd year EB02 F 56 Right Unknown* Congenital EB03 M 38 Right Detached <2 nd year retinas (before the 2 nd year) EB04 F 58 Right ROP Congenital EB05 M 54 Right ROP Congenital EB06 F 54 Right ROP Congenital EB07 F 34 Right Optic nerve <2 nd year damage* (before the 2 nd year) EB08 F 34 Right ROP Congenital EB09 F 55 Right Unknown* Congenital EB10 F 55 Left ROP Congenital EB11 F 56 Right ROP Congenital EB12 M 56 Left ROP Congenital * No additional detail available. ROP, retinopathy of prematurity; M, male; F, female. cortices within the postcentral gyri (BA1, 2, 3, 40, 43; Tables S1 and S2 available online and Figure 2). In the OC, the middle and inferior occipital gyri (MOG and IOG, BA19) were bilaterally activated in EB subjects and deactivated in SC subjects during both auditory and tactile conditions (Tables S3 and S4 and Figure 2). Among OC areas, only the lingual-gyrus/cuneus (LG, BA17/18) was activated during the auditory and tactile conditions in both groups, though to a larger extent in EB subjects. A between-group comparison for each modality showed that the frontal and occipital cortices, including parts of the cuneus, the LG (BA18), and the MOG and IOG, were overall more activated in EB than in SC subjects during both the auditory and the tactile conditions (q[fdr] < 0.05, Figure 3). The contrasts between the auditory and tactile modality in each group did not show any modality-specific activation in the OC in either EB or SC subjects even at a low threshold of p < (uncorrected for multiple comparisons in combination with a cluster size threshold of p < 0.05). Even at the individual level, no clear modality specificity was observed in the OC of EB subjects. While some subjects activated their OC more during the auditory than the tactile conditions, other subjects showed the opposite pattern. We did not find a double dissociation within the OC of any blind subject. Conjunction analyses using a threshold of p < (uncorrected) confirmed that the left and right MOG-IOG (BA19) and the LG/cuneus (BA17/18) were similarly activated by both modalities in EB subjects (Table S5). In SC subjects, the same conjunction analysis showed that the left (x: 40, y: 70, z: 1, 38 voxels) and right (x: 46, y: 67, z: 1, 131 voxels) Neuron 68, , October 7, 2010 ª2010 Elsevier Inc. 139

3 Figure 1. In-Scan Behavioral Performance (A) Percentage of correct responses during the detection conditions as a function of group and modality. Standard error bars (SEM) are displayed in the graphs. (B and C) Percentage of correct responses as a function of the group (EB and SC), modality (auditory and tactile), and task (identification and localization). MOG were deactivated by both modalities. Other brain regions, including the right middle frontal gyrus (MFG, BA6), the inferior parietal lobule (IPL), and the precuneus, were activated conjointly by both modalities in each group. However, in SC subjects the activation in the parietal cortex was confined to the left hemisphere, while in EB subjects it was bilateral. Main Effects Related to the Task The contrast of Identification minus Detection showed a similar activation pattern in the frontal cortex including the Figure 2. Brain Areas Recruited During Auditory and Tactile Processing in Blind and Sighted Subjects Functional brain activity in 12 early-blind subjects (EB) and 12 sighted control subjects (SC) during auditory and tactile processing was projected onto a 3D representation of the right and left hemispheres (RH and LH) of a representative brain of one subject. The activation maps resulting from the contrasts between the auditory and tactile conditions compared to the baseline were obtained by RFX with a threshold of q < 0.05, corrected for multiple comparisons using FDR in combination with a cluster size threshold of p < (A) Brain activation related to auditory processing in EB (in red) and in SC subjects (in green). (B) Brain deactivation, i.e., negative BOLD response, related to auditory processing in EB (in pink) and in SC subjects (in pale green). (C) Brain activation related to tactile processing in EB (in yellow) and in SC subjects (in blue). (D) Brain deactivation, i.e., negative BOLD response, related to tactile processing in EB (in pale yellow) and in SC subjects (in pale blue). The middle and inferior occipital gyri were bilaterally activated in EB subjects and deactivated in SC subjects during the auditory and tactile conditions. The lingual gyrus-cuneus (LingG, BA17-18) was bilaterally activated in EB and SC subjects during the auditory and tactile conditions, though to a much larger extent in EB subjects and during the auditory conditions. See also Tables S1 S Neuron 68, , October 7, 2010 ª2010 Elsevier Inc.

4 Figure 3. Between-Group Comparison for the Auditory and the Tactile Modality Activation maps resulting from a group comparison in each modality using RFX with a threshold of q < 0.05 corrected for multiple comparisons using a FDR in combination with a cluster size threshold of p < 0.05 and superimposed onto the brain of a representative subject. (A) Brain regions that were activated more in EB than in SC (in orange-yellow) and more in SC than in EB (in blue-green) subjects during auditory processing. (B) Brain regions that were activated more in EB than in SC (in orange-yellow) and more in SC than in EB (in blue-green) subjects during tactile processing. The color-gradient scale codes the t values. LH, left hemisphere; RH, right hemisphere. See also Table S8 and Figure 5. right middle frontal cortex and the insula bilaterally in both groups, but occipital activation in EB subjects only (p < uncorrected, Table S6). The contrast of Localization minus Detection showed a similar brain activation pattern in the parietal cortex including the right IPL in both groups (p < uncorrected, Table S7), but occipital activation in EB subjects only. The reversed contrasts, i.e., Detection minus Localization and Detection minus Identification, did not show any activation in the OC in either of the two groups. The between-group comparison for each task (Identification and Localization) confirmed that the OC, including the MOG bilaterally and the lingual gyrus (LG), was significantly more activated in EB than in SC subjects (q < 0.05, FDR corrected, Table S8). What- and Where-Specific Activation The contrast of Localization minus Identification showed that in addition to medial parietal areas (such as the precuneus and the superior parietal lobule [SPL]), part of the right MOG (BA19; x: 51, y: 64, z: 5; 101 voxels) was specifically more activated during localization than identification conditions in EB subjects (Table S9 and Figure 4). When we used a more liberal threshold of p < (uncorrected), a similar MOG focus appeared in the left hemisphere as well. With this more liberal threshold, the right MOG focus was also found in the contrast of Localization minus Identification in the auditory modality alone (where the behavioral performance was well balanced). The between-group comparison for the same contrast revealed bilateral activations in MOG (Figure 5). The activity profile observed in the right MOG focus was very similar to the one observed in the parietal regions (Figure 4). The same contrast in SC generated similar activation in the parietal cortex, particularly in the right hemisphere, but no activation in the OC (Table S9). The contrast of Identification minus Localization did not show any activation in the OC for either of the groups (even at a more relaxed threshold of p < uncorrected). Results from the control experiment in sighted subjects confirmed that a large portion of the MOG, which included the MOG focus observed in EB subjects, was more activated during the spatial than the nonspatial task in vision (Figure 6). Correlation analysis was performed in order to test the relationship between behavioral performance and brain activity within the OC. Correlations were computed in both groups between performance accuracy (in the six conditions: auditory detection, auditory identification, auditory localization, tactile detection, tactile identification, and tactile localization ) and percentage of signal change (beta values) in the right MOG, and the left and right precuneus as identified in the contrast of Identification minus Localization in EB subjects (see Neuron 68, , October 7, 2010 ª2010 Elsevier Inc. 141

5 Figure 4. Brain Areas that Show a Preference for Spatial over Nonspatial Processing in Blind Subjects Task-related brain activation in 12 early-blind subjects (EB). Activation maps were projected onto a 3D representation of the right and left hemispheres (RH and LH) of the brain of a representative subject. The statistical maps were obtained using RFX with an uncorrected threshold of p < The top of the figure shows the brain activation related to identification (in blue) and localization (in yellow), as obtained from the contrasts of Identification minus Localization and of Localization minus Identification. The right middle occipital gyrus (MOG, BA19) and several subregions of the parietal cortex were more activated during localization than identification conditions. The main parietal activation foci were located in the precuneus in both hemispheres (see Table S9). No activation focus in the occipital cortex of EB subjects was found to be more activated during identification than localization conditions. The color-gradient scale codes the t values. The group comparison for the contrast of Localization minus Identification (interaction) confirms that the right (and the left) MOG was more activated in EB than in SC subjects and during spatial than nonspatial processing (see Figure 5). The bottom of Figure 4 shows the activity profiles (i.e., the percentages of signal change as a function of task and modality) in the left and right precuneus and the right MOG in EB (first row) and SC subjects (second row). In EB subjects, the activity profiles for the right MOG and the precuneus were very similar; localization conditions in both modalities activated these regions more strongly than identification or detection conditions. In sighted subjects, the activity profile was similar to that in EB subjects in the precuneus but not in the right MOG, which was mainly deactivated in both modalities. In the precuneus, the signal change appears weaker in SC than in EB subjects, but the precuneus areas with the biggest differences between identification and localization in SC do not exactly correspond to the ones in EB subjects (see Table S9). *p < 0.01; **p < 0.005; ***p < Standard error bars (SEM) are displayed in the graphs. See also Tables S6 S9 and Figures 3 and 5. Table S9). A significant correlation between performance accuracy and percentage of signal change was observed during the auditory localization task in the right MOG of EB subjects (r = 0.61; p = 0.034, uncorrected for multiple comparisons). No other significant correlations were found (all r < 0.50 and all p values > 0.05). 142 Neuron 68, , October 7, 2010 ª2010 Elsevier Inc.

6 Figure 5. Between-Group Comparison for the Contrast of Localization minus Identification Activation maps resulting from a group comparison for the contrast of Localization minus Identification using RFX with a threshold of q < 0.05 corrected for multiple comparisons using a FDR in combination with a cluster size threshold of p < 0.05, superimposed onto the averaged brain of all subjects. Results show bilateral activation in the MOG. The color-gradient scale codes the t values. SAG, sagittal; TRA, transverse; COR, coronal; R, right; L, left. See also Table S8. DISCUSSION Functional Specialization for Spatial Processing in the MOG In the present study, we found that the anterior part of the right MOG, which is considered a part of the visual dorsal stream in sighted subjects (Dumoulin et al., 2000; Wandell et al., 2007), showed a functional preference for processing spatial properties of both auditory and tactile stimuli in EB subjects (Figure 4). The region was most strongly activated during localization conditions in both modalities, and its activity was specifically correlated with the accuracy of individual performance in auditory localization. Results from a control experiment confirmed that the same region was preferentially involved in visual spatial processing in sighted subjects (Figure 6). It is unlikely that the observed functional preference was due to differences in arousal or attention, since task-related activity differences were obtained in the auditory modality even though the localization and identification tasks were equally difficult (Figure 1B). Furthermore, within the tactile modality, the localization task was actually somewhat easier than the identification task (Figure 1C), yet the right MOG clearly showed a preference for spatial over nonspatial processing (Figure 4). These findings suggest that even in the absence of visual stimulation, the anterior part of the right MOG attains a functional role in spatial processing and is overlaid with modal sensory responses. Since the anterior right MOG is also involved in visual spatial processing in sighted subjects, we hypothesize that functional specialization in the occipital cortex (OC) develops independently of visual experience. While visual deprivation leads to a replacement of visual by nonvisual input ( compensatory plasticity ; Rauschecker, 1995), each structure retains its designated functional role. Task Specificity in Cross-modal Plasticity The involvement of dorsal-stream occipital areas in either auditory or tactile spatial processing by blind subjects has been reported previously (Weeks et al., 2000; Gougoux et al., 2005; Poirier et al., 2006; Ricciardi et al., 2007; Bonino et al., 2008; Saenz et al., 2008; Ptito et al., 2009). However, the present study demonstrates for the first time that a specific part of the OC, the right MOG, shows a preference for spatial over nonspatial processing that is independent of modality and relates this preference to individual sound localization performance. The resulting concept of task specificity in cross-modal plasticity is also supported by other studies: Sathian and coworkers have identified a region in the intraoccipital sulcus (IOS, probable area V3A) of blind subjects that shows specific activation during a microspatial tactile discrimination task in EB subjects (Zangaladze et al., 1999; Weisser et al., 2005; Stilla et al., 2008). More anterior-lateral areas in the OC, including the lateral occipital complex (LOC), seem to play a functional role in shape processing regardless of sensory experience (Arno et al., 2001; Amedi et al., 2007). The present results further advance the notion of preserved functional specialization: by testing what and where functions simultaneously with different modalities in the same blind subjects as well as in sighted controls, the principle of task-specificity in cross-modal plasticity can now be generalized across various functions. Future studies could test whether the concept holds even beyond OC and applies to the development of all cortical areas. Activation versus Deactivation of OC by Multisensory Input The MOG focus found in the present study is in the vicinity of what is known as hmt+, the human analog of area MT, first described as a visual motion area in monkeys (Allman and Neuron 68, , October 7, 2010 ª2010 Elsevier Inc. 143

7 Figure 6. Task-Specific Brain Areas in Vision Functional brain activity in six sighted subjects (SC) during identification and localization of visual stimuli was projected onto a 3D representation of the right and left hemispheres (RH and LH) in one representative brain. The activation maps resulting from the contrasts between identification and localization conditions were obtained using fixed-effects analyses (FFX) with a threshold of q < 0.02, corrected for multiple comparisons using FDR, in combination with a cluster size threshold of p < Brain activation related to visual identification (in blue) and visual localization (in yellow) is shown. The right (and the left) MOG, including the MOG focus found in blind subjects (x: 51, y: 64, z: 5; see Figure 4), was more strongly activated during the localization than the identification task. Additional activation foci were found in the precuneus and the superior parietal lobule. The inferior occipital gyrus (IOG) in both hemispheres was more activated during the identification task (shape processing) than during localization. Kaas, 1971). Area hmt+ seems to be specialized for motion processing in both sighted and blind humans (Poirier et al., 2005, 2006; Matteau et al., 2010). However, according to these studies, hmt+ has to be considered a multisensory (supra- or metamodal) brain area that is also activated by moving auditory and tactile stimuli in both sighted and blind subjects. In the present study, such multisensory OC activation was only observed in a part of the LG/cuneus region (BA17/18) (Figure 2). By contrast, the MOG focus was activated by nonvisual (auditory and tactile) stimulation in EB but deactivated bilaterally during all auditory and tactile conditions in SC subjects (Figure 4). This finding is in general agreement with other studies demonstrating deactivation of OC by nonvisual stimuli in sighted subjects (Weeks et al., 2000; Sadato et al., 1996; Burton et al., 2004; Haxby et al., 1994; Kawashima et al., 1995; Shulman et al., 1997). Deactivation of the OC during auditory and tactile conditions may minimize potential interference from visual stimuli during nonvisual tasks (Gougoux et al., 2005; Laurienti et al., 2002). Both activation and deactivation by nonvisual tasks indicate the presence of nonvisual input to OC even in sighted individuals. This adds to the growing evidence for the multisensory nature of relatively early processing stages in sensory cortices (Pascual-Leone et al., 2005; Ghazanfar and Schroeder, 2006; Kayser et al., 2008). In EB subjects, multisensory processing became even more pronounced, and nonvisual deactivation (suppression) turned into activation (excitation): both auditory and tactile inputs converged upon the same regions of the OC, including the LG/cuneus and the middle and inferior occipital gyri (MOG and IOG, BA19). The absence of dissociation between auditory and tactile modalities in the OC of blind subjects further supports the metamodal nature of perceptual and cognitive processes in a wide range of tasks (Röder et al., 1996; Weaver and Stevens, 2007). The preordained existence of nonvisual inputs to OC in sighted subjects may provide a basis for their expansion in individuals with early blindness. Occipitoparietal System for Spatial Localization In addition to the right MOG activation observed in EB subjects, medial parietal areas showed a similar preference for spatial processing in both EB and SC subjects. The precuneus in particular has previously been described as a multisensory operator that is specialized for spatial processing in sighted subjects (Renier et al., 2009). Studies in visually deprived cats and monkeys have suggested that cross-modal plasticity could in fact originate in multisensory parietal regions, where the opportunity for cross-modal competition may be the greatest (Hyvärinen et al., 1981; Rauschecker and Korte, 1993). Prior imaging data from blind humans using interregional correlation techniques have been compatible with this proposition (Weeks et al., 2000). Other possible mechanisms include a strengthening of corticocortical input from other sensory modalities (Falchier et al., 2002; Rockland and Ojima, 2003) or of nonvisual thalamocortical connections (Hackett and Schroeder, 2009). One may further hypothesize that the altered dorsal occipitoparietal regions contribute to the superior abilities of EB subjects 144 Neuron 68, , October 7, 2010 ª2010 Elsevier Inc.

8 commonly observed for spatial tasks (Lessard et al., 1998; Röder et al., 1999; Fortin et al., 2008). Lending support to this idea, a previous study showed that repetitive transcranial magnetic stimulation (rtms), which is thought to create a transient virtual lesion, applied to the dorsal extrastriate cortex selectively impaired sound localization performance in EB but not SC subjects (Collignon et al., 2007). Conclusions The present results suggest a hierarchical model of visual cortical development in which modal sensory responses overlay dorsal-stream spatial selectivity. Thus our study demonstrates that although the occipital cortex of early-blind subjects is largely multisensory, specific regions retain the functional role they assume in sighted individuals. Additional work with an even greater variety of stimuli and paradigms is needed to elucidate how strengthened auditory and tactile inputs reach the occipital cortex of early-blind subjects and whether additional functional specialization is present in the occipital cortex of the blind. EXPERIMENTAL PROCEDURES Subjects Twelve EB subjects (4 males, 10 right-handed, mean age = 49 years; standard deviation [SD] = 9.2) and 12 sighted control subjects (SC) matched for age, sex, and handedness participated in this study (mean age = 45 years; SD = 10.9; F 1,23 = 1.03, p = 0.32). In addition, a separate group of 6 sighted subjects participated in another control experiment in the visual domain (4 males, all right-handed, mean age = 29; SD = 2.4). All participants were healthy. Blind participants were totally blind from birth or by the second year of life. They had no history of normal vision and no memories of visual experience. Table 1 provides details about gender, age, handedness, etiology, and onset of blindness. The protocol was approved by Georgetown University s Institutional Review Board (IRB); written informed consent was obtained from all participants prior to the experiment. MRI Acquisition All fmri data were acquired at Georgetown University s Center for Functional and Molecular Imaging with an echo-planar imaging (EPI) sequence on a 3-Tesla Siemens Tim Trio scanner with a 12-channel head coil (flip angle = 90, TR = 3 s, TE = 60 ms, matrix). The FOV was mm 2 with at matrix size; slice thickness was 2.8 mm with a 0.2 mm gap, resulting in an effective voxel resolution of 3 mm 3. For each run, 184 volumes of 50 slices (slice thickness: 2.8 mm and gap thickness: mm) were acquired in a continuous sampling. At the end, a three-dimensional T1-weighted MPRAGE image (resolution mm 3 ) was acquired for each subject. Equipment and Stimuli Auditory Stimulation We used a set of 16 stimuli that consisted of four distinct piano chords originating from four different locations in the virtual auditory space in front of the listener (see Figure S1). The stimuli were presented via electrostatic MRIcompatible headphones (STAX, frequency transfer function 20 Hz 20 KHz). The chords were created with Fleximusic Composer and consisted of four different notes. The chords were then further modified with Amphiotik Synthesis from Holistiks to generate an authentic virtual auditory environment based on the head related transfer functions (HRTF) measured from a mean head size. The four sound sources were located within a half-circle in front of the listener. The first source was located at the extreme left of the subject corresponding to 90 azimuth. The other sources were placed at 30, +30, and +90, respectively, progressing toward the right of the listener. The stimulus duration was 1000 ms (including 10 ms rise/fall times); the interstimulus interval was 1000 ms. Pretests performed on a separate group of ten subjects showed that the sounds could be localized accurately in a forcedchoice paradigm with a mean accuracy > 95% in all subjects. Tactile Stimulation Vibrotactile stimuli were produced by nonmagnetic, ceramic piezoelectric bending elements (benders Q220-A4-303YB, Piezo Systems, Quick Mount Bender) placed directly under the four fingers (index, middle, ring, and little finger) of the subject s right hand (see Figure S1). The benders were driven by sinusoidal tones of 40, 80, 160, and 320 Hz and produced vibrations corresponding to that driving frequency. The stimulus intensity was amplified with Alesis RA150 Amplifiers. The stimulus duration (1000 ms including 10 ms rise/fall times) and interstimulus interval (1000 ms) were identical to those used for the auditory stimulation. The stimuli were created with Adobe Audition software (Adobe Systems) and generated by an echo audiofire8 soundcard (AudioAmigo). The four piezoelectric elements were attached to the fingers via hook-and-loop fasteners (Velcro). The stimulated hand was fixed within an antispasticity ball splint (Sammons Preston) to enable maximal comfort and prevent hand movements during the experiment. Experimental Paradigms We used a block design paradigm in which six conditions (duration of one condition = 18 s) were separated by off conditions. The experimental conditions comprised three tasks stimulus detection, identification, and localization for each of the sensory modalities auditory and vibrotactile. Before scanning, subjects underwent a brief practice session during which the intensity of auditory and tactile stimulation was adjusted per subject to ensure that stimuli were clearly distinguishable yet comfortable. Special care was taken to equalize the subjective intensity of stimulation in the two modalities. A predetermined pseudorandom order of runs was counterbalanced across subjects. SC subjects were blindfolded during the experiments. During the localization and identification conditions, we used a one-back comparison task in which we asked subjects to compare each stimulus with the previous one and determine whether it was the same or different regarding either identity (stimulus frequency) or spatial location (stimulated finger or sound source). The subjects delivered their responses by pressing one of two buttons on a response pad held in the left hand. During the detection tasks, subjects were required to press the left button of the response pad at stimulus presentation. The onset of each condition was announced through the headphones upon termination of the preceding resting period (12 s). Stimulus presentation and collection of behavioral data were controlled by Superlab software 4.0 (Cedrus corporation). Visual Control Experiment In this experiment, subjects were presented with four geometrical figures that appeared at four different locations on a projection screen (see Figure S2). Subjects had to determine in a one-back comparison task whether each stimulus was the same as or different than the preceding one regarding either its identity (shape; nonspatial) or its location (spatial). During the experiment, subjects gazed at a fixation cross in the center of the screen and provided their responses for each comparison by pressing one of two buttons on a response pad held in the left hand. Each stimulus was displayed for 500 ms (interstimulus interval = 1500 ms). Subjects brain activity was monitored in two runs of about 5min. Data Analysis The BOLD responses for the six experimental conditions were compared with BrainVoyager QX (Version 1.10, BrainInnovation) by applying a regression analysis. Prior to analysis, preprocessing consisted of slice timing correction, temporal high-pass filtering (removing frequencies lower than 3 cycles/run), and correction for small interscan head movements. Data were transformed into Talairach space (Talairach and Tournoux, 1988). For anatomical reference, the computed statistical maps were overlaid on the 3D T1-weighted scans. In each individual, the predictor time courses of the regression model were computed on the basis of a general linear model of the relation between neural activity and hemodynamic response function (HRF), assuming an HRF neural response during phases of sensory stimulation. A random-effects analysis (RFX) in the group was then performed at the whole-brain level, using Neuron 68, , October 7, 2010 ª2010 Elsevier Inc. 145

9 one-sample and two-sample t tests with a threshold of q < 0.05, corrected for false discovery rate (FDR), in combination with a cluster size threshold of p < Alternatively, a threshold of p < (uncorrected for multiple comparisons) was used in the condition comparisons depending on the strictness of the contrasts performed. Brain areas that were activated during both auditory and tactile conditions were identified by conjunction analyses with a threshold of p < (uncorrected for multiple comparisons), in combination with a cluster size threshold of p < Given the relatively small size of the sample used in the visual control experiment (n = 6), fixed-effects analysis (FFX) in the group was performed at the whole-brain level, using one-sample t tests with a threshold of q < 0.02, corrected for FDR, in combination with a cluster size threshold of p < The maps obtained were projected onto the inflated cortical surface of a representative subject for presentation purposes. Analysis of the behavioral data was performed with a three-way mixeddesign ANOVA with group (EB versus SC) as a between-subjects factor and modality (Auditory versus Tactile) and task (Identification and Localization) as within-subject factors. An additional two-way ANOVA with group as a between-subjects factor and modality as a within-subjects factor was performed to analyze separately the detection scores. When significant main effects were observed (p < 0.05), post-hoc analyses were performed with the Newman-Keuls test. SUPPLEMENTAL INFORMATION Supplemental Information includes two figures and nine tables and can be found with this article online at doi: /j.neuron ACKNOWLEDGMENTS The authors are grateful to the blind volunteers for their participation in our study. The study was supported by grants from the National Institutes of Health (R01 NS and R01EY018923), the Cognitive Neuroscience Initiative of the National Science Foundation (BCS ), the NSF PIRE program (OISE ), and the Tinnitus Research Initiative to J. P. Rauschecker, and by grants from the Academy of Finland to S.C. (a National Center of Excellence grant and Neuro-Program) and to I.A. (grant number ). A.G.D.V. is a senior research associate and L. Renier is a postdoctoral researcher at the National Fund for Scientific Research (Belgium). L.R. was also supported by the Belgian American Educational Foundation and the Special Fund for Research from the Université Catholique de Louvain (Belgium). We also wish to thank Jennifer Sasaki and Valérie Wuyts for their help in the subject recruitment and Priyanka Chablani for her editing work. Accepted: September 19, 2010 Published: October 6, 2010 REFERENCES Alain, C., Arnott, S.R., Hevenor, S., Graham, S., and Grady, C.L. (2001). What and where in the human auditory system. Proc. Natl. Acad. Sci. USA 98, Alary, F., Duquette, M., Goldstein, R., Elaine Chapman, C., Voss, P., La Buissonnière-Ariza, V., and Lepore, F. (2009). Tactile acuity in the blind: a closer look reveals superiority over the sighted in some but not all cutaneous tasks. Neuropsychologia 47, Allman, J.M., and Kaas, J.H. (1971). A representation of the visual field in the caudal third of the middle tempral gyrus of the owl monkey (Aotus trivirgatus). Brain Res. 31, Amedi, A., Raz, N., Pianka, P., Malach, R., and Zohary, E. (2003). Early visual cortex activation correlates with superior verbal memory performance in the blind. Nat. Neurosci. 6, Amedi, A., Stern, W.M., Camprodon, J.A., Bermpohl, F., Merabet, L., Rotman, S., Hemond, C., Meijer, P., and Pascual-Leone, A. (2007). Shape conveyed by visual-to-auditory sensory substitution activates the lateral occipital complex. Nat. Neurosci. 10, Arno, P., De Volder, A.G., Vanlierde, A., Wanet-Defalque, M.C., Streel, E., Robert, A., Sanabria-Bohórquez, S., and Veraart, C. (2001). Occipital activation by pattern recognition in the early blind using auditory substitution for vision. Neuroimage 13, Bavelier, D., and Neville, H.J. (2002). Cross-modal plasticity: where and how? Nat. Rev. Neurosci. 3, Bonino, D., Ricciardi, E., Sani, L., Gentili, C., Vanello, N., Guazzelli, M., Vecchi, T., and Pietrini, P. (2008). Tactile spatial working memory activates the dorsal extrastriate cortical pathway in congenitally blind individuals. Arch. Ital. Biol. 146, Büchel, C., Price, C., Frackowiak, R.S., and Friston, K. (1998). Different activation patterns in the visual cortex of late and congenitally blind subjects. Brain 121, Burton, H., Snyder, A.Z., Conturo, T.E., Akbudak, E., Ollinger, J.M., and Raichle, M.E. (2002). Adaptive changes in early and late blind: a fmri study of Braille reading. J. Neurophysiol. 87, Burton, H., Diamond, J.B., and McDermott, K.B. (2003). Dissociating cortical regions activated by semantic and phonological tasks: a FMRI study in blind and sighted people. J. Neurophysiol. 90, Burton, H., Sinclair, R.J., and McLaren, D.G. (2004). Cortical activity to vibrotactile stimulation: an fmri study in blind and sighted individuals. Hum. Brain Mapp. 23, Cohen, L.G., Celnik, P., Pascual-Leone, A., Corwell, B., Falz, L., Dambrosia, J., Honda, M., Sadato, N., Gerloff, C., Catalá, M.D., and Hallett, M. (1997). Functional relevance of cross-modal plasticity in blind humans. Nature 389, Collignon, O., Renier, L., Bruyer, R., Tranduy, D., and Veraart, C. (2006). Improved selective and divided spatial attention in early blind subjects. Brain Res. 1075, Collignon, O., Lassonde, M., Lepore, F., Bastien, D., and Veraart, C. (2007). Functional cerebral reorganization for auditory spatial processing and auditory substitution of vision in early blind subjects. Cereb. Cortex 17, De Volder, A.G., Toyama, H., Kimura, Y., Kiyosawa, M., Nakano, H., Vanlierde, A., Wanet-Defalque, M.C., Mishina, M., Oda, K., Ishiwata, K., and Senda, M. (2001). Auditory triggered mental imagery of shape involves visual association areas in early blind humans. Neuroimage 14, Deibert, E., Kraut, M., Kremen, S., and Hart, J., Jr. (1999). Neural pathways in tactile object recognition. Neurology 52, Dumoulin, S.O., Bittar, R.G., Kabani, N.J., Baker, C.L., Jr., Le Goualher, G., Bruce Pike, G., and Evans, A.C. (2000). A new anatomical landmark for reliable identification of human area V5/MT: a quantitative analysis of sulcal patterning. Cereb. Cortex 10, Falchier, A., Clavagnier, S., Barone, P., and Kennedy, H. (2002). Anatomical evidence of multimodal integration in primate striate cortex. J. Neurosci. 22, Fortin, M., Voss, P., Lord, C., Lassonde, M., Pruessner, J., Saint-Amour, D., Rainville, C., and Lepore, F. (2008). Wayfinding in the blind: larger hippocampal volume and supranormal spatial navigation. Brain 131, Ghazanfar, A.A., and Schroeder, C.E. (2006). Is neocortex essentially multisensory? Trends Cogn. Sci. 10, Goldreich, D., and Kanics, I.M. (2003). Tactile acuity is enhanced in blindness. J. Neurosci. 23, Goodale, M.A., and Milner, A.D. (1992). Separate visual pathways for perception and action. Trends Neurosci. 15, Gougoux, F., Lepore, F., Lassonde, M., Voss, P., Zatorre, R.J., and Belin, P. (2004). Neuropsychology: Pitch discrimination in the early blind. Nature 430, 309. Gougoux, F., Zatorre, R.J., Lassonde, M., Voss, P., and Lepore, F. (2005). A functional neuroimaging study of sound localization: visual cortex activity predicts performance in early-blind individuals. PLoS Biol. 3, e27. Hackett, T.A., and Schroeder, C.E. (2009). Multisensory integration in auditory and auditory-related areas of cortex. Hear. Res. 258, Neuron 68, , October 7, 2010 ª2010 Elsevier Inc.

10 Hamilton, R.H., Pascual-Leone, A., and Schlaug, G. (2004). Absolute pitch in blind musicians. Neuroreport 15, Haxby, J.V., Grady, C.L., Horwitz, B., Ungerleider, L.G., Mishkin, M., Carson, R.E., Herscovitch, P., Schapiro, M.B., and Rapoport, S.I. (1991). Dissociation of object and spatial visual processing pathways in human extrastriate cortex. Proc. Natl. Acad. Sci. USA 88, Haxby, J.V., Horwitz, B., Ungerleider, L.G., Maisog, J.M., Pietrini, P., and Grady, C.L. (1994). The functional organization of human extrastriate cortex: a PET-rCBF study of selective attention to faces and locations. J. Neurosci. 14, Hyvärinen, J., Carlson, S., and Hyvärinen, L. (1981). Early visual deprivation alters modality of neuronal responses in area 19 of monkey cortex. Neurosci. Lett. 26, Kaas, J.H., and Hackett, T.A. (2000). Subdivisions of auditory cortex and processing streams in primates. Proc. Natl. Acad. Sci. USA 97, Kawashima, R., O Sullivan, B.T., and Roland, P.E. (1995). Positron-emission tomography studies of cross-modality inhibition in selective attentional tasks: closing the mind s eye. Proc. Natl. Acad. Sci. USA 92, Kayser, C., Petkov, C.I., and Logothetis, N.K. (2008). Visual modulation of neurons in auditory cortex. Cereb. Cortex 18, Kujala, T., Alho, K., Paavilainen, P., Summala, H., and Näätänen, R. (1992). Neural plasticity in processing of sound location by the early blind: an eventrelated potential study. Electroencephalogr. Clin. Neurophysiol. 84, Kujala, T., Alho, K., Kekoni, J., Hämäläinen, H., Reinikainen, K., Salonen, O., Standertskjöld-Nordenstam, C.G., and Näätänen, R. (1995). Auditory and somatosensory event-related brain potentials in early blind humans. Exp. Brain Res. 104, Laurienti, P.J., Burdette, J.H., Wallace, M.T., Yen, Y.F., Field, A.S., and Stein, B.E. (2002). Deactivation of sensory-specific cortex by cross-modal stimuli. J. Cogn. Neurosci. 14, Lessard, N., Paré, M., Lepore, F., and Lassonde, M. (1998). Early-blind human subjects localize sound sources better than sighted subjects. Nature 395, Lewald, J. (2002). Vertical sound localization in blind humans. Neuropsychologia 40, Maeder, P.P., Meuli, R.A., Adriani, M., Bellmann, A., Fornari, E., Thiran, J.P., Pittet, A., and Clarke, S. (2001). Distinct pathways involved in sound recognition and localization: a human fmri study. Neuroimage 14, Matteau, I., Kupers, R., Ricciardi, E., Pietrini, P., and Ptito, M. (2010). Beyond visual, aural and haptic movement perception: hmt+ is activated by electrotactile motion stimulation of the tongue in sighted and in congenitally blind individuals. Brain Res. Bull. 82, Merabet, L.B., and Pascual-Leone, A. (2010). Neural reorganization following sensory loss: the opportunity of change. Nat. Rev. Neurosci. 11, Pascual-Leone, A., and Torres, F. (1993). Plasticity of the sensorimotor cortex representation of the reading finger in Braille readers. Brain 116, Pascual-Leone, A., Amedi, A., Fregni, F., and Merabet, L.B. (2005). The plastic human brain cortex. Annu. Rev. Neurosci. 28, Poirier, C., Collignon, O., De Volder, A.G., Renier, L., Vanlierde, A., Tranduy, D., and Scheiber, C. (2005). Specific activation of the V5 brain area by auditory motion processing: An fmri study. Brain Res. Cogn. Brain Res. 25, Poirier, C., Collignon, O., Scheiber, C., Renier, L., Vanlierde, A., Tranduy, D., Veraart, C., and De Volder, A.G. (2006). Auditory motion perception activates visual motion areas in early blind subjects. Neuroimage 31, Prather, S.C., Votaw, J.R., and Sathian, K. (2004). Task-specific recruitment of dorsal and ventral visual areas during tactile perception. Neuropsychologia 42, Ptito, M., Moesgaard, S.M., Gjedde, A., and Kupers, R. (2005). Cross-modal plasticity revealed by electrotactile stimulation of the tongue in the congenitally blind. Brain 128, Ptito, M., Matteau, I., Gjedde, A., and Kupers, R. (2009). Recruitment of the middle temporal area by tactile motion in congenital blindness. Neuroreport 20, Rauschecker, J.P. (1995). Compensatory plasticity and sensory substitution in the cerebral cortex. Trends Neurosci. 18, Rauschecker, J.P. (1998). Cortical processing of complex sounds. Curr. Opin. Neurobiol. 8, Rauschecker, J.P., and Kniepert, U. (1994). Auditory localization behaviour in visually deprived cats. Eur. J. Neurosci. 6, Rauschecker, J.P., and Korte, M. (1993). Auditory compensation for early blindness in cat cerebral cortex. J. Neurosci. 13, Rauschecker, J.P., and Tian, B. (2000). Mechanisms and streams for processing of what and where in auditory cortex. Proc. Natl. Acad. Sci. USA 97, Raz, N., Amedi, A., and Zohary, E. (2005). V1 activation in congenitally blind humans is associated with episodic retrieval. Cereb. Cortex 15, Renier, L.A., Anurova, I., De Volder, A.G., Carlson, S., VanMeter, J., and Rauschecker, J.P. (2009). Multisensory integration of sounds and vibrotactile stimuli in processing streams for what and where. J. Neurosci. 29, Ricciardi, E., Vanello, N., Sani, L., Gentili, C., Scilingo, E.P., Landini, L., Guazzelli, M., Bicchi, A., Haxby, J.V., and Pietrini, P. (2007). The effect of visual experience on the development of functional architecture in hmt+. Cereb. Cortex 17, Rockland, K.S., and Ojima, H. (2003). Multisensory convergence in calcarine visual areas in macaque monkey. Int. J. Psychophysiol. 50, Röder, B., and Rösler, F. (2003). Memory for environmental sounds in sighted, congenitally blind and late blind adults: Evidence for cross-modal compensation. Int. J. Psychophysiol. 50, Röder, B., Rösler, F., Hennighausen, E., and Näcker, F. (1996). Event-related potentials during auditory and somatosensory discrimination in sighted and blind human subjects. Brain Res. Cogn. Brain Res. 4, Röder, B., Teder-Sälejärvi, W., Sterr, A., Rösler, F., Hillyard, S.A., and Neville, H.J. (1999). Improved auditory spatial tuning in blind humans. Nature 400, Röder, B., Rösler, F., and Neville, H.J. (2001). Auditory memory in congenitally blind adults: A behavioral-electrophysiological investigation. Brain Res. Cogn. Brain Res. 11, Röder, B., Stock, O., Bien, S., Neville, H., and Rösler, F. (2002). Speech processing activates visual cortex in congenitally blind humans. Eur. J. Neurosci. 16, Romanski, L.M., Tian, B., Fritz, J., Mishkin, M., Goldman-Rakic, P.S., and Rauschecker, J.P. (1999). Dual streams of auditory afferents target multiple domains in the primate prefrontal cortex. Nat. Neurosci. 2, Sadato, N., Pascual-Leone, A., Grafman, J., Ibañez, V., Deiber, M.P., Dold, G., and Hallett, M. (1996). Activation of the primary visual cortex by Braille reading in blind subjects. Nature 380, Sadato, N., Pascual-Leone, A., Grafman, J., Deiber, M.P., Ibañez, V., and Hallett, M. (1998). Neural networks for Braille reading by the blind. Brain 121, Saenz, M., Lewis, L.B., Huth, A.G., Fine, I., and Koch, C. (2008). Visual motion area MT+/V5 responds to auditory motion in human sight-recovery subjects. J. Neurosci. 28, Shulman, G.L., Corbetta, M., Buckner, R.L., Raichle, M.E., Fiez, J.A., Miezin, F.M., Petersen, S.E., and Petersen, S.E. (1997). Top-down modulation of early sensory cortex. Cereb. Cortex 7, Stilla, R., Hanna, R., Hu, X., Mariola, E., Deshpande, G., and Sathian, K. (2008). Neural processing underlying tactile microspatial discrimination in the blind: A functional magnetic resonance imaging study. J. Vis. 8, Stoesz, M.R., Zhang, M., Weisser, V.D., Prather, S.C., Mao, H., and Sathian, K. (2003). Neural networks active during tactile form perception: Common and Neuron 68, , October 7, 2010 ª2010 Elsevier Inc. 147