Report. Changing Reference Frames during the Encoding of Tactile Events. Elena Azañón 1,2 and Salvador Soto-Faraco 1,2,3, * 1

Transcription

1 Current Biology 18, , July 22, 2008 ª2008 Elsevier Ltd All rights reserved DOI /j.cub Changing Reference Frames during the Encoding of Tactile Events Report Elena Azañón 1,2 and Salvador Soto-Faraco 1,2,3, * 1 Parc Científic de Barcelona Barcelona Spain 2 Departament de Psicologia Bàsica Universitat de Barcelona Barcelona Spain 3 Institució Catalana de Recerca i Estudis Avançats (ICREA) Barcelona Spain Summary The mindless act of swatting a mosquito on the hand poses a remarkable challenge for the brain. Given that the primary somatosensory cortex maps skin location independently of arm posture [1, 2], the brain must realign tactile coordinates in order to locate the origin of the stimuli in extrapersonal space. Previous studies have highlighted the behavioral relevance of such an external mapping of touch, which results from combining somatosensory input with proprioceptive and visual cues about body posture [3 7]. However, despite the widely held assumption about the existence of this remapping process from somatotopic to external space and various findings indirectly suggesting its consequences [8 11], a demonstration of its changing time course and nature was lacking. We examined the temporal course of this multisensory interaction and its implications for tactile awareness in humans using a crossmodal cueing paradigm [12, 13]. What we show is that before tactile events are referred to external locations [12 15], a fleeting, unconscious image of the tactile sensation abiding to a somatotopic frame of reference rules performance. We propose that this early somatotopic glimpse arises from the initial feed-forward sweep of neural activity to the primary somatosensory cortex, whereas the later externally-based, conscious experience reflects the activity of a somatosensory network involving recurrent connections from association areas. Results and Discussion Participants held their arms crossed over the body midline and were presented with a lateralised visual target at one of four unpredictable positions (top or bottom of the left or right hand) [12, 13] (see Figure 1A and Experimental Procedures). Target presentation was preceded by a spatially nonpredictive tactile tap (at the ring finger, on the same or opposite side as the flash) at a variable interval (30, 60, 180, and 360 ms; see Figure 1B). In the cueing task, participants were asked to judge as quickly as possible the position of the light flash in the vertical dimension (top-bottom), irrespective of the side of presentation and the location of the irrelevant tactile cue. The measure of interest was the spatial-cueing effect; that is, the difference in response *Correspondence: salvador.soto@icrea.es latencies to visual targets presented at the opposite minus same side as the tactile cue (according to external space; see Table 1). This measure provides an index of the influence of touch on the processing of the visual event, thus allowing us to infer its spatial reference frame. The question is whether early somatotopic representations, characteristic of the primary somatosensory cortex (SI), are in fact inconsequential for overt behavior or whether they are momentarily available but rapidly obliterated by an externally based representation. If the latter were true, we would expect to see the typical cueing effects abiding to an external reference frame at long cue target onset asynchronies (CTOA, hereafter) but, in contrast, at sufficiently short CTOAs, the cueing effect would correspond to anatomical space. Thus, given that in the crossed-hands posture the somatosensory and the external maps are misaligned, we predicted a reversal in terms of the spatial-congruency effect from short to long intervals. The results showed that reaction times (RTs) to the visual targets, presented during the first hundred milliseconds after the tactile cue, were faster in opposite-side (anatomically congruent but spatially incongruent) trials than in same-side trials (210 ms cueing effects at both 30 ms [t(15) = 2.36; p = 0.032] and 60 ms [t(15) = 2.85; p = 0.012] intervals; see Figure 2A and Table 1) thus, an inverse cueing effect in which touches to the left hand, now placed on the right side, facilitated processing of left hemispace visual events and vice versa. This pattern completely reversed after about two hundred milliseconds, so that tactile cues produced a facilitation of targets presented at the same external location (8 ms marginally significant cueing effect at 180 ms CTOA [t(15) = 22.06; p = 0.057] and 13 ms cueing effect at 360 ms CTOA [t(15) = 22.39; p = 0.031]). These effects were verified by an ANOVA on the RTs, with spatial cue (same side versus opposite side) and CTOA (30, 60, 180, and 360 ms) as independent variables. There was a main effect of CTOA (F[3,45] = 31.39, p < 0.001), with quicker responses as the CTOA increased (see Table 1), a trend that accords to the typical foreperiod effect in RT experiments [16]. Critically, the interaction between spatial cue and CTOA was significant (F[3,45] = 6.99, p = 0.001), reflecting the fact that shorter CTOAs produced negative cueing effects whereas longer CTOAs produced positive cueing effects. This critical interaction was consistent throughout the experiment and across response effectors (see Supplemental Data, available online). An analogous ANOVA on the error data did not reveal significant main effects or interactions; accuracy was high overall (>93%; see Table 1). In a control experiment, 16 participants were tested, with their hands placed at the same intermanual distance but in an uncrossed posture. In this case, cueing effects for both short (8 ms at 30 ms CTOA [t(15) = 22.19; p = 0.044] and 14 ms at 60 ms CTOA [t(15) = 23.81; p = 0.002]) and long (19 ms at both 180 ms [t(15) = 26.18; p < 0.001] and 360 ms CTOA [t(15) = 24.17; p < 0.001]) intervals were, as expected, positive (see Figure 2A and Table 1). This is so because, with participants hands uncrossed, both somatotopic and external representations are aligned with each other. An ANOVA on the RT data confirmed a main effect of spatial cue (F[1,15] = 48.36, p < 0.001; same side 612 ms versus opposite side 627 ms) and a main effect of CTOA (F[3,45] = 30.40, p < 0.001), with

2 Changing Reference Frames during Tactile Encoding 1045 Figure 1. Experimental Setup and Trial Sequence (A) Frontal view of the setup used in the main experiment (see Experimental Procedures for details). The participant s arms were crossed over the body midline (30 cm separation between ring fingers) and occluded beneath a black cardboard screen (rendered transparent in this picture for illustrative purposes). The LEDs producing the visual targets were mounted just above each hand (top versus bottom LED, 8 cm separation). The solenoid tappers producing the tactile stimuli (driven by a 9 V square wave) were attached to the dorsal surface of the middle phalanx of each ring finger. A central LED, 40 cm from the participant s eyes, served as a fixation point. Two loudspeakers produced white noise throughout the experiment in order to mask potential noises from the stimulators. (B) Trial sequence. In a typical trial, after 1 s fixation, a 9 ms tap was delivered to one of the hands and followed (at four different intervals: 30, 60, 180, or 360 ms) by the brief (9 ms) visual flash from one of the two LEDs (top or bottom) placed above one of the hands (same or different from the one receiving the tap, equiprobably). In the cueing task, participants were asked to respond as quickly as possible to the elevation of the light flash in the vertical dimension, with the tactile cue being irrelevant. In the crossmodal comparison task (order of the tasks counterbalanced), participants were asked to judge, as accurately as possible and without time pressure, whether the light flash occurred in the same or different side of the tactile event. decreasing latencies as the CTOA increased (see Table 1). Yet, the interaction between spatial cue and CTOA was, in contrast with the crossed-hands condition, not significant (F[3,45] = 1.98, p = 0.130). An analogous ANOVA on the error data did not reveal significant main effects or interactions. In order to address the spatial specificity of the interaction found in the crossed-hands posture, we tested a new group of participants (n = 20) in the crossed posture, with the target lights now placed 28 cm above the hands/solenoid tappers. In this case, the cueing effects were neutralized at all CTOAs (2, 22, 22, and 1 ms cueing effects for the 30, 60, 180, and 360 ms; all p > 0.1; see Figure 2A), suggesting that the cueing effects previously reported were spatially specific and not the result of a general orienting bias toward the whole hemispace (see analysis in the Supplemental Data). Our data are in accord with previous spatial-cueing studies showing that crossmodal links between vision and touch can remap across postural changes in order to take current body posture into account [12 15]. The results also show that these crossmodal links are spatially specific and are thus related with previous findings in which visuotactile interactions appear to be stronger in peripersonal as compared to extrapersonal space [17, 18]. The present study, however, extends and qualifies these previous results, showing that at early stages, which had not been previously explored (that is, CTOAs below 100 ms), the remapping process has not started or is still not complete. This explains the early negative cueing effects and why, at intermediate intervals (i.e., 180 ms), a time when the remapping process is presumably ongoing, only a marginally significant cueing effect (consistent with external coordinates) is found. An additional experiment (see Supplemental Data), tracking this time period in more detail, confirmed the early negative cueing at 30 and 60 ms, the positive cueing at 360 ms, and the nonsignificant cueing effects at intermediate intervals (90 to 180 ms). In turn, this indefinite pattern at intermediate intervals could help account for the mixed results found in previous studies testing tactile visual cueing around these intervals (i.e., 200 ms CTOA [12, 13]). In order to explore the subjective sensory experience associated with the somatotopic code operating at early stages of tactile processing, we introduced a crossmodal comparison task in which participants (the same ones tested in the cueing experiment, task order counterbalanced) were presented with the same sequence of events but asked to judge, as accurately as possible, whether the spatial location of the light flash was the same or different as that of the tactile event (see Figure 2B). If participants perceived touch location in terms of the early somatotopic code, systematic errors would be expected at short intervals in this task. However, the results revealed that subjects performed well above chance at all intervals (R 83% [t(15) = 10.29; 10.07; 10.84; 16.25, respectively; all p < 0.001]). That is, performance did not switch from below- to above-chance levels from short to long touch-vision intervals, as would have been expected if perception was based at first on the initial somatotopic glimpse. This latter finding is relevant as to whether people are aware of the changing spatial representations resulting from the remapping process that was revealed in the cueing task. This is a nontrivial question, given that some authors have precisely claimed that the conscious sensation of touch is initially linked to external space rather than to skin [19], but direct empirical evidence was so far scarce. Here, we saw that when asked explicitly to localize touch, participants did not base their reports on an anatomical reference frame, even for very short touch-vision intervals. In fact, what we found is that explicit localization

3 Current Biology Vol 18 No Table 1. Interparticipant Mean RTs, Accuracy, and Cuing Effects in the Crossed and Uncrossed Cueing-Task Experiments CTOA 30 ms 60 ms 180 ms 360 ms Target location Same Opp. Diff. Same Opp. Diff. Same Opp. Diff. Same Opp. Diff. Crossed hands RT (ms) a a c a Errors (%) Uncrossed hands RT (ms) a a b b Errors (%) Same and Opp. (opposite) refer to the position of the target with respect to the cue in external space. Diff. (difference) indicates the subtraction of opposite-side minus same-side trials in RTs (cueing effect) and accuracy (error percentages). Footnotes denote the significance level (two-tailed) for the opposite- minus same-side differences. a p <.05. b p <.001. c marginally significant. performance of tactile events is spatially stable and accurate throughout all intervals, a result that is in stark contrast with the pattern of effects found with the cueing paradigm and consistent with the lack of awareness of the initial somatotopic representations. The present study exposes the temporal course in the encoding of tactile space, from a somatotopic frame of reference, reflecting the neural organization in SI, to the external representations prevailing in orienting behaviors. Many of the accounts regarding how tactile spatial information is represented highlight the importance of a putative remapping process between these representations, and some authors have even proposed concrete estimates of its timing. For instance, from the results of crossed-hands experiments, Yamamoto and Kitazawa [10] suggested that remapping takes about 300 ms to completion, on the basis of the frequent confusions registered in ordering two tactile events at shorter intervals [9 11] (but see also [8]). In addition, there is evidence that when the hands are crossed, quick saccades to tactile stimuli presented at one hand tend to be initially executed in the wrong direction and then corrected online after about 265 ms of initiation ([20], as inferred from Groh et al. s Figure 6, though numerical analysis is lacking). However, to our knowledge, this is the first study that has explicitly measured the contrasting behavioral consequences of the changing spatial representations of touch at different points in its time course. According to our results, the process of remapping spatial coordinates would not start before the 60 ms after stimulus application (point until which the initial somatotopic frame of reference would prevail) and would be completed in an interval ranging from 180 to 360 ms after stimulus onset (a time from which an external reference frame would prevail). Given that tactile experience often relies on externally based representations [4, 6, 7, 21], the present results provide empirical evidence that primary somatotopic representations exist implicitly but are behaviorally relevant as a first step toward external representations. Previous findings have shown that neither the early electrical potentials evoked on SI neurons by a stimulus on the skin [22, 23] nor the isolated activity of these SI neurons, per se, are sufficient conditions for a subjective sensory experience [24]. These findings are consistent with the idea that conscious perception of tactile events may be referred to locations in external space, given that in neither case is the activity in SI linked with information about external spatial location [19]. The remapping to an external frame of reference, possibly bringing the subjective experience of touch location [19], is dependant on convergence of somatosensory inputs with information from visual and proprioceptive systems. Previous physiological data allows one to speculate about the neural underpinnings of this tactile remapping process. Some areas within the parietal cortex, such as the ventral intraparietal area (VIP) and the superior parietal cortex, seem to play a key role in transformations between spatial representations in different sensory modalities [25 29] and are thus good candidates to form part of the tactile remapping system. In particular, the role of the VIP in mediating multisensory representation of limb position in humans has been suggested by fmri [30] and TMS [31] evidence, and monkey electrophysiology shows that this area contains neurons representing visual, vestibular, and tactile information [32]. In addition, the superior parietal cortex codes for spatial positions by combining visual and somatosensory input, as revealed by fmri and human lesion studies [33, 34]. Finally, additional nonparietal areas, such as the premotor cortex [30, 35] and the putamen [36], both containing bimodal neurons with spatially overlapping somatosensory and visual receptive fields, and the superior colliculus, where some neurons present somatosensory activation dependent on eye position [37], have also been suggested to play a role in tactile remapping. Interestingly, it has been hypothesized that perception of tactile location might be embodied by the feedback from these multisensory integration structures in secondary- and higher-association cortices back to primary somatosensory areas [19] (see [38] for a review of feedback from multisensory regions to sensoryspecific brain areas and [39 41] for a similar idea of the crucial role of feedback to primary areas for awareness in the visual domain). The time course suggested in the present study is consistent with the proposed neural architecture at the functional level. Following this tentative argument, tactile influence observed ms after stimulus onset would reflect mainly the initial feed-forward sweep of somatosensory information activating the primary somatosensory area (starting about ms after stimulus onset and lasting until around 100 ms after stimulus onset, according to some estimates [42 44]). The ensuing representation of touch organized in terms of external spatial coordinates, as inferred by participants performance in our task, would be available at about 180 ms after stimulus application, possibly reflecting the processing of a somatosensory network encompassing the somatosensory cortex (SI and SII) as well as structures beyond SII, including parietal areas (superior parietal cortex and VIP) and possibly frontal premotor areas and the superior colliculus.

4 Changing Reference Frames during Tactile Encoding 1047 Figure 2. Experimental Results (A) Mean spatial-cueing effects (opposite- minus same-side trial RTs) as a function of CTOA in the cueing task. A positive cueing effect indicates faster performance for visual events in the same-side trials as compared to the opposite-side trials, as coded with respect to positions in external space. The black line represents trials in which the subjects performed the task with the hands crossed. The thick gray line corresponds to trials in which participants hands were uncrossed (uncrossed control experiment), and the thin gray line corresponds to trials in which participants hands were crossed but a physical distance (28 cm) between the location of the tactile cue and the location of the visual target was introduced (far-crossed experiment). The error bars represent the SEM. (B) Proportion of same-side responses as a function of touch-vision interval in the crossmodal comparison task. The solid line represents trials in which the tap and the visual event were presented on the same side, and the dotted line represents trials in which they were presented on different sides. The error bars represent the SEM. In conclusion, the present study provides direct evidence of the behavioral consequences of the changing spatial-reference frames during the encoding of tactile events throughout time. To our knowledge, this is the first empirical demonstration in humans that the encoding of tactile information starts in the form of a fleeting, though behaviorally consequential, somatotopic representation that is later replaced by an external representation that governs the perception of tactile events in space. Although this temporal sequence had been advanced previously, its behavioral consequences had never been directly measured. A corollary conclusion from our study is that the initial, somatotopically based, stages of tactile coding occur below the threshold of consciousness. This would explain why we are seldom confused about where to direct our attention to sudden tactile events on our skin, despite potential conflicts between frames of reference across frequent updates in body posture. Instead, our initial reaction to a tactile event is usually accurately referred to its origin in space. Experimental Procedures Participants Fifty-two undergraduate students from the University of Barcelona participated in the study (16 in the main crossed-hands experiment, 16 in the uncrossed control experiment, and 20 in the far-crossed experiment) in exchange for course credit (mean age 21 years; SD = 1.87). All had normal or corrected-to-normal vision and reported normal tactile sensitivity (by selfreport). Given previous findings that right-handed people and professional pianists present better tactile temporal resolution abilities under some conditions [9, 45], we ensured that none of our participants were right handed or had previous experience playing any musical instrument. The experiment was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. All of the participants gave their informed consent prior to their inclusion in the study and were naive as to the purpose of the experiment. Apparatus The experimental setup is shown in Figure 1A. Participants sat at a table with their head resting on a chin-rest in a dimly-lit, sound-attenuated room. A central, yellow light emitting diode (LED) placed in front of the subjects midline, 40 cm from their eyes, served as a fixation point. Tactile stimulation was delivered to the dorsal surface of the middle phalanx of each ring finger. It consisted of a 9 ms tap with suprathreshold intensity, delivered through tappers 8 mm in diameter (Solenoid Tactile Tapper, M&E Solve, UK), driven by a 9 V square wave. The distance between the ring fingers of each hand was kept constant at 30 cm (15 cm to left and to the right of central fixation). Visual events consisted of the brief illumination of one of the four red LEDs (5 mm diameter) for 9 ms. These target LEDs were set out in two vertically arranged dyads (8 cm top-bottom separation), one placed in each hemispace (15 cm left and right from the fixation point), and attached to the cardboard screen just above the hands. Responses were made through two foot pedals placed beneath the right (dominant) foot, one positioned under the heel and the other under the toes. White noise was presented continuously throughout the experiment from two loudspeakers placed near each hand to mask any subtle sound made by the tactile stimulation. E-prime software was used to control stimulus presentation and register pedal responses through a custom-made relay box connected to the parallel port. Design and Procedure Participants were asked to sit and rest their hands (palms down) on the table, adopting a crossed-hands posture, with the ring fingers placed at marked positions. The left arm was placed over the right arm in half of the participants, and vice versa in the other half of the participants. The arms were then covered by a rigid black cardboard screen, which housed the LEDs. Each participant performed 12 blocks of 96 trials, with task (cueing versus crossmodal comparison) being varied after six blocks (task order counterbalanced across participants). In all cases, a trial started with the illumination of the fixation LED (participants were instructed to look straight ahead at the fixation point throughout the experiment). After an interval of 1000 ms, the 9 ms tactile event on either the right or the left ring finger was presented. After a variable interval (30, 60, 180, or 360 ms, selected randomly and equiprobably at each trial), a single visual event (9 ms flash) in one of the two positions (top or bottom) of one side (left or right hemispace) was presented (see Figure 1B). Therefore, in half of the trials, the tactile event was presented on the same side as the visual light whereas in the rest of the trials, it was presented on the opposite side as the visual light (randomly selected on a trial-by-trial basis). Note that same side and different side always refer, in this study, to the position of the events in external (extrapersonal) space. Thus, in same-side trials, the tactile event on the right hand, now placed on the left, was followed by a light appearing on the left hemispace, whereas in opposite-side trials, tactile events in the right hand were followed by a light appearing on the right hemispace. Participants

5 Current Biology Vol 18 No indicated their responses by lifting their toes or heel off the pedals that they had to otherwise press throughout the experimental block. In the cueing task, they lifted their toes or heel as quickly as possible after upper or lower visual targets (respectively). In the crossmodal comparison task, they lifted their toes or heel as accurately as possible after same- or different-side trials (respectively) between touch and light. Participants were also informed about the nonpredictive nature of tactile events with respect to the location or elevation of the upcoming visual targets in the cueing task. The trial was terminated after a response was made, after which an intertrial interval of 1000 ms led to the next trial (starting with the illumination of the fixation LED). Prior to the beginning of each task, a block of 32 practice trials was run, with feedback provided on erroneous responses (flicker of the central fixation LED). These were not included in the analyses. Supplemental Data Supplemental data include Supplemental Results and Discussion and one figure and can be found with this article online at current-biology.com/cgi/content/full/18/14/1044/dc1/. Acknowledgments This research was supported by grants from the Spanish Ministerio de Educación y Ciencia (Spain; SEJ /PSIC and CDS00012) and by a fellowship Beca de Formación de Profesorado Universitario from the Spanish Ministerio de Educación y Ciencia to E.A. The authors would like to thank James T. Enns for helpful comments to an earlier version of this manuscript. Received: March 25, 2008 Revised: May 22, 2008 Accepted: June 9, 2008 Published online: July 10, 2008 References 1. Penfield, W., and Rasmussen, T. (1950). The cerebral cortex of man: a clinical study of localization of function (New York: Mc-Millan). 2. Sutherland, M.T. (2006). The hand and the ipsilateral primary somatosensory cortex. J. Neurosci. 32, Aglioti, S., Smania, N., and Peru, A. (1999). Frames of reference for mapping tactile stimuli in brain-damaged patients. J. Cogn. Neurosci. 1, Azañón, E., and Soto-Faraco, S. (2007). Alleviating the crossed-hands deficit by seeing uncrossed rubber hands. Exp. Brain Res. 4, Driver, J., and Grossenbacher, P.G. (1996). Multimodal spatial constraints on tactile selective attention. In Attention and performance XVI: Information integration in perception and communication, T. Inui and J.L. McClelland, eds. (Cambridge, MA: MIT Press), pp Shore, D.I., Gray, K., Spry, E., and Spence, C. (2005). Spatial modulation of tactile temporal-order judgments. Perception 10, Soto-Faraco, S., Ronald, A., and Spence, C. (2004). Tactile selective attention and body posture: assessing the multisensory contributions of vision and proprioception. Percept. Psychophys. 7, Shore, D.I., Spry, E., and Spence, C. (2002). Confusing the mind by crossing the hands. Brain Res. Cogn. Brain Res. 1, Wada, M., Yamamoto, S., and Kitazawa, S. (2004). Effects of handedness on tactile temporal order judgment. Neuropsychologia 14, Yamamoto, S., and Kitazawa, S. (2001). Reversal of subjective temporal order due to arm crossing. Nat. Neurosci. 7, Yamamoto, S., Moizumi, S., and Kitazawa, S. (2005). Referral of tactile sensation to the tips of L-shaped sticks. J. Neurophysiol. 5, Kennett, S., Eimer, M., Spence, C., and Driver, J. (2001). Tactile-visual links in exogenous spatial attention under different postures: convergent evidence from psychophysics and ERPs. J. Cogn. Neurosci. 4, Kennett, S., Spence, C., and Driver, J. (2002). Visuo-tactile links in covert exogenous spatial attention remap across changes in unseen hand posture. Percept. Psychophys. 7, Driver, J., and Spence, C. (1998). Cross-modal links in spatial attention. Philos. Trans. R. Soc. Lond. B Biol. Sci. 1373, Eimer, M., Cockburn, D., Smedley, B., and Driver, J. (2001). Cross-modal links in endogenous spatial attention are mediated by common external locations: evidence from event-related brain potentials. Exp. Brain Res. 4, Mowrer, O.H. (1940). Preparatory Set (Expectancy): Some Methods of Measurement. Psychol. Monogr. 52, di Pellegrino, G., Làdavas, E., and Farné, A. (1997). Seeing where your hands are. Nature 388, Làdavas, E., di Pellegrino, G., Farnè, A., and Zeloni, G. (1998). Neuropsychological evidence of an integrated visuotactile representation of peripersonal space in humans. J. Cogn. Neurosci. 10, Kitazawa, S. (2002). Where conscious sensation takes place. Conscious. Cogn. 3, Groh, J.M., and Sparks, D.L. (1996). Saccades to somatosensory targets. I. behavioral characteristics. J. Neurophysiol. 1, Schicke, T., and Roder, B. (2006). Spatial remapping of touch: confusion of perceived stimulus order across hand and foot. Proc. Natl. Acad. Sci. USA 31, Libet, B., Alberts, W.W., Wright, E.W., Jr., and Feinstein, B. (1967). Responses of human somatosensory cortex to stimuli below threshold for conscious sensation. Science 808, Schubert, R., Blankenburg, F., Lemm, S., Villringer, A., and Curio, G. (2006). Now you feel it now you don t: ERP correlates of somatosensory awareness. Psychophysiology 1, de Lafuente, V., and Romo, R. (2005). Neuronal correlates of subjective sensory experience. Nat. Neurosci. 12, Andersen, R.A., Essick, G.K., and Siegel, R.M. (1985). Encoding of spatial location by posterior parietal neurons. Science 4724, Bremmer, F., Schlack, A., Duhamel, J.R., Graf, W., and Fink, G.R. (2001). Space coding in primate posterior parietal cortex. Neuroimage 1, S46 S Buneo, C.A., and Andersen, R.A. (2006). The posterior parietal cortex: sensorimotor interface for the planning and online control of visually guided movements. Neuropsychologia 13, Grefkes, C., and Fink, G.R. (2005). The functional organization of the intraparietal sulcus in humans and monkeys. J. Anat. 1, Sakata, H., Shibutani, H., Ito, Y., and Tsurugai, K. (1986). Parietal cortical neurons responding to rotary movement of visual stimulus in space. Exp. Brain Res. 3, Lloyd, D.M., Shore, D.I., Spence, C., and Calvert, G.A. (2003). Multisensory representation of limb position in human premotor cortex. Nat. Neurosci. 1, Bolognini, N., and Maravita, A. (2007). Proprioceptive alignment of visual and somatosensory maps in the posterior parietal cortex. Curr. Biol. 21, Schlack, A., Sterbing-D Angelo, S.J., Hartung, K., Hoffmann, K.P., and Bremmer, F. (2005). Multisensory space representations in the macaque ventral intraparietal area. J. Neurosci. 18, Felician, O., Romaiguere, P., Anton, J.L., Nazarian, B., Roth, M., Poncet, M., and Roll, J.P. (2004). The role of human left superior parietal lobule in body part localization. Ann. Neurol. 5, Wolpert, D.M., Goodbody, S.J., and Husain, M. (1998). Maintaining internal representations: the role of the human superior parietal lobe. Nat. Neurosci. 6, Graziano, M.S. (1999). Where is my arm? The relative role of vision and proprioception in the neuronal representation of limb position. Proc. Natl. Acad. Sci. USA 18, Graziano, M.S., and Gross, C.G. (1993). A bimodal map of space: somatosensory receptive fields in the macaque putamen with corresponding visual receptive fields. Exp. Brain Res. 1, Groh, J.M., and Sparks, D.L. (1996). Saccades to somatosensory targets. III. eye-position-dependent somatosensory activity in primate superior colliculus. J. Neurophysiol. 1, Driver, J., and Noesselt, T. (2008). Multisensory interplay reveals crossmodal influences on sensory-specific brain regions, neural responses, and judgments. Neuron 57, Ahissar, M., and Hochstein, S. (2004). The reverse hierarchy theory of visual perceptual learning. Trends Cogn. Sci. 10, Hochstein, S., and Ahissar, M. (2002). View from the top: hierarchies and reverse hierarchies in the visual system. Neuron 5, Juan, C.H., and Walsh, V. (2003). Feedback to V1: a reverse hierarchy in vision. Exp. Brain Res. 2, Allison, T., McCarthy, G., and Wood, C.C. (1992). The relationship between human long-latency somatosensory evoked potentials recorded from the cortical surface and from the scalp. Electroencephalogr. Clin. Neurophysiol. 4,

6 Changing Reference Frames during Tactile Encoding Allison, T., McCarthy, G., Wood, C.C., Darcey, T.M., Spencer, D.D., and Williamson, P.D. (1989). Human cortical potentials evoked by stimulation of the median nerve. I. Cytoarchitectonic areas generating shortlatency activity. J. Neurophysiol. 3, Mauguiere, F., Merlet, I., Forss, N., Vanni, S., Jousmaki, V., Adeleine, P., and Hari, R. (1997). Activation of a distributed somatosensory cortical network in the human brain. A dipole modelling study of magnetic fields evoked by median nerve stimulation. Part I: Location and activation timing of SEF sources. Electroencephalogr. Clin. Neurophysiol. 4, Kobor, I., Furedi, L., Kovacs, G., Spence, C., and Vidnyanszky, Z. (2006). Back-to-front: improved tactile discrimination performance in the space you cannot see. Neurosci. Lett. 1 2,