Auditory Peripersonal Space in Humans: a Case of Auditory Tactile Extinction

Transcription

1 Neurocase (2001) Vol. 7, pp Oxford University Press 2001 Auditory Peripersonal Space in Humans: a Case of Auditory Tactile Extinction Elisabetta Làdavas 1, Francesco Pavani 1,2 and Alessandro Farnè 1,2 1 Dipartimento di Psicologia, Università degli Studi di Bologna and 2 Ospedale I.N.R.C.A. I Fraticini, Firenze, Italy Abstract Animal experiments have shown that the spatial correspondence between auditory and tactile receptive fields of ventral pre-motor neurons provides a map of auditory peripersonal space around the head. This allows neurons to localize a near sound with respect to the head. In the present study, we demonstrated the existence of an auditory peripersonal space around the head in humans. In a right-brain damaged patient with tactile extinction, a sound delivered near the ipsilesional side of the head extinguished a tactile stimulus delivered to the contralesional side of the head (crossmodal auditory tactile extinction). In contrast, when an auditory stimulus was presented far from the head, cross-modal extinction was dramatically reduced. This spatially specific cross-modal extinction was found only when a complex sound like a white noise burst was presented; pure tones did not produce spatially specific cross-modal extinction. These results show a high degree of functional similarity between the characteristics of the auditory peripersonal space representation in humans and monkeys. This similarity suggests that analogous physiological substrates might be responsible for coding this multisensory integrated representation of peripersonal space in human and non-human primates. Introduction Neurophysiological and neuropsychological studies have shown the existence of an integrated visuo-tactile representation of peripersonal space in non-human primates as well as in humans. In particular, neurophysiological evidence demonstrated the existence of aligned visual and tactile maps of peripersonal space coded by bimodal neurons located in the putamen (Graziano and Gross, 1993), the ventral premotor cortex (area F4) (Rizzolatti et al., 1981) and in parietal areas 7b (Graziano and Gross, 1995) and Ventral Intra Parietal (Duhamel et al., 1991). These bimodal neurons have tactile receptive fields (RFs) on the hand or on the face and correspondent visual RFs in the space immediately adjacent to the tactile fields. These neurons typically respond best, i.e. with the highest firing rate, to visual stimuli presented near the skin surface. In contrast, weak responses are evoked when visual stimuli are presented far from the skin (Gentilucci et al., 1983, 1988; Fogassi et al., 1996; Graziano et al., 1997; Duhamel et al., 1998). Due to these functional properties, these neurons appear ideally suited to code an integrated visuo-tactile representation of the peripersonal space (Rizzolatti et al., 1998; Graziano and Gross, 1998). Recent studies have also documented the existence in humans of an integrated system encoding both visual and tactile inputs within the peripersonal space around the hand (di Pellegrino et al., 1997a; Làdavas et al., 1998a, 2000) and the face (Làdavas et al., 1998b). Recently, Graziano et al. (1999) studied neurons in monkey ventral pre-motor cortex with tactile RFs that covered the contralateral side of the head, including the back of the animal s head. They found that 53% of these neurons were multimodal, that is they responded to tactile stimuli, but also to visual and acoustic stimuli presented near the head. Auditory RFs matched the location of tactile RFs. Importantly, neuron responses were almost extinguished when the auditory stimulus was presented 30 or 50 cm away from the head. Another interesting characteristic of these neurons was that pure tones of various frequencies ( Hz) did not elicit any response, whereas more complex sounds, such as a burst of white noise, jingling keys, claps or crinkling paper, produced strong responses. These multimodal neurons thus seem to be the neurophysiological substrate of an integrated auditory and tactile representation of the peripersonal space near the head. Here we studied whether a similar integrated system exists in humans, by investigating cross-modal extinction between a sound and a touch, in a patient with tactile extinction. Patients with extinction fail to report a contralesional stimulus when a competing stimulus is presented simultaneously on Correspondence to: Elisabetta Làdavas, Department of Psychology, University of Bologna, Viale Berti Pichat 5, Bologna, Italy. Tel: 39 (0) ; Fax: 39 (0) ; ladavas@psibo.unibo.it

2 98 E. Làdavas, F. Pavani and A. Farnè the ipsilesional side, even though they can report either stimulus when it is presented alone (Bender, 1952; De Renzi, 1982). The extinction phenomenon has been conceived as a competition between concurrent targets for access to limited attentional resources (Ward et al., 1994; di Pellegrino et al., 1997b). Unilateral damage of a brain area with a contralateral field representation results in a reduction in competitive weights in the affected field. As a consequence, stimuli presented in the contralesional field evoke a weak activation of that portion of space and, therefore, they are extinguished when competing stimuli are simultaneously presented in the intact ipsilesional field. The existence of multimodal neurons with tactile RFs on the head and corresponding auditory RFs that extend outwards from the tactile field into the space near the head, predicts that in patients with tactile extinction, an ipsilesional auditory stimulus will extinguish a contralesional tactile stimulus (cross-modal auditory tactile extinction). In particular, this phenomenon should be found when an auditory stimulus is presented near the ipsilesional side of the head and a tactile stimulus is delivered to the contralesional side of the head. This is because, due to the activity of these neurons, an acoustic stimulus delivered near the head also activates the corresponding somatosensory representation of that side of the head. The simultaneous activation of the somatosensory representation of the left side of the head (by a tactile stimulus) and of the right side of the head (by an acoustic stimulus) should produce an extinction of those stimuli presented in the weaker representation, i.e. that corresponding to the contralesional side of the head. In contrast, a substantial reduction in cross-modal extinction should be expected when the auditory stimulus is presented on the ipsilesional side, but far from the head. This is because an auditory stimulus presented in the far space should activate the system encoding peripersonal space to a lesser degree. Moreover, because Graziano et al. (1999) have shown that multimodal neurons are not activated by pure tones but only by complex sounds, spatially specific cross-modal extinction should only be expected when the auditory stimulus is a white noise. These hypotheses were tested in a right-brain damaged patient with tactile extinction. Case report Patient CA is an 83-year-old lady who gave her informed consent to participate in the study. She suffered an ischaemic stroke in the right hemisphere 19 months before the test. A computed tomography (CT) scan at the time of testing revealed a right unilateral lesion involving frontal, parietal and temporal lobes (see Fig. 1). On clinical examination she was alert and oriented in time and space; the Mini Mental State Examination was within normal limits. No visual field defects were evident when assessed by Goldman perimetry. She did not present any sign of visual neglect, as assessed by a variety of tests taken from the Behavioural Inattention Test (Wilson et al., 1987) neglect battery, including letter Fig. 1. Lesion site of patient CA reconstructed according to Damasio and Damasio (1989). cancellation [left side (L) 100%, right side (R) 85% of correct responses], picture scanning and menu reading (L 100%, R 100% on both tasks). Bell s cancellation test was also performed correctly (L 100%, R 100%). No sign of personal neglect was evident, as measured by means of the Fluff test, in which the patient was asked to search for six adhesive stickers located on the contralesional side of her body. She did not show visual extinction, as assessed by the confrontation test. Twenty unilateral left and right visual stimuli and 20 bilateral simultaneous visual stimuli were presented. CA s performance was correct in both unilateral and bilateral trials. She did not report any sensory loss in hearing, as confirmed by an audiometry test. Moreover, the patient did not show any sign of auditory neglect when tested with the same auditory stimuli employed in the experimental investigation (white noise and 1.5 khz pure tone). When 20 unilateral left and right acoustic stimuli were presented, her performance was errorless. CA presented a left-arm hemiparesis; her grip force, as measured with a dynamometer, revealed an important force reduction in the contralesional hand (left 3.6 kg) with respect to the ipsilesional hand (right 13 kg). In addition, she showed left tactile extinction on the hand and on the neck. Tactile stimuli delivered to the hands were presented on the dorsal aspect of the index finger; tactile stimuli delivered to the neck were presented behind the ears, just below the mastoid process of the temporal bone. In either case, 20 unilateral left and right tactile stimuli and 20 bilateral simultaneous tactile stimuli were delivered. CA s performance was normal in detecting a single left or right tactile stimulation (100% correct detection), showing a normal tactile sensitivity in relation to both body parts. In contrast, she showed a severe left tactile extinction when two stimuli were simultaneously applied to the hands (15% correct detection) and to the neck (20% correct detection). Materials and methods The apparatus for stimuli presentation consisted of a touchsensitive sensor (1 cm 2 ), a loudspeaker (0.3 W, 8 Ω) and a remote switch. These three devices were connected in an optically isolated electric circuit. The switch could be in two positions (pressed on; released off) in order to control sound emission through the loudspeaker. Contact of the

3 Auditory peripersonal space in humans 99 Fig. 2. Schematic drawing of the experimental set up illustrating the four possible loudspeaker positions relative to the patient s head. The arrow on the left side of the head indicates the tactile stimulus location. Percentages of correct detection of the left tactile stimuli in bilateral stimulation (bold) and in left unilateral stimulation (in brackets) are reported for each loudspeaker position as a function of the type of sound (white noise and pure tone). Both single and bilateral trials refer to cross-modal conditions. (white noise burst or pure tone). Each condition was run in a separate block. In the unimodal tactile condition, an identical touch- sensitive sensor (1 cm 2 ) was attached to the index finger of the experimenter s right hand. In this condition, sensors were inactive, i.e. they did not trigger any sound, and they were used to obtain identical tactile stimulation on both sides of the neck. Four types of trials were presented: (1) a single touch on the left; (2) a single touch on the right; (3) two simultaneous touches, one on the left and the other on the right; (4) no tactile stimulation (catch trials). The patient was informed that she would feel either a light touch on the left side of the neck, or a light touch on the right side of the neck, or two simultaneous touches on both sides of the neck. The patient was also informed that occasionally she would feel nothing at all. In cross-modal conditions, the experimenter held the remote switch in his right hand to control sound emission. Four types of trials were presented: (1) a single touch on the left; (2) a single sound on the right (for this stimulus condition the experimenter closed the circuit by contact between the sensor and his own thumb); (3) a touch on the left and a simultaneous sound on the right; (4) no tactile or acoustic stimulation (catch trials). The patient was informed that she would either feel a light touch on the left side of the neck, or a sound on the right side of the head, or a touch on the left side and a simultaneous sound on the right side. She was told that occasionally she would feel nothing at all. In both unimodal and cross-modal conditions the patient was instructed to respond verbally to what she perceived, with the words left, right, both or none, irrespective of the sensory modality. The patient was asked to respond with an appropriate verbal label immediately after the stimulation. However, when she failed to respond spontaneously, she was prompted by the experimenter to produce one of the four sensor with the skin triggered the simultaneous production of a sound only when the switch was pressed (on). Sound stimuli were either white noise bursts or pure tones (1.5 khz) and lasted only the duration of contact between the sensor and the skin ( 1 s). Tactile stimuli were brief ( 1 s), light contacts of the sensor, attached to the experimenter s index finger, with the patient s skin. Tactile stimuli were applied behind the ear and just below the mastoid process of the temporal bone. The patient sat blindfolded in the centre of a silent room, and was instructed to keep her head straight throughout the entire experiment. The experimenter sat behind the patient, with the sensor firmly attached to the index finger of his left hand. The loudspeaker was mounted at ear level on an adjustable metal stand, and was positioned on the right of the patient s head, either in front or behind it. Specifically, the loudspeaker was placed in four locations along a line 45 between the sagittal and coronal planes: (1) in the front right quadrant 20 cm from the patient s head; (2) in the front right quadrant 70 cm from the patient s head; (3) in the back right quadrant 20 cm from the patient s head; (4) in the back right quadrant 70 cm from the patient s head (see Fig. 2). The sound pressure level measured at the patient s ear was always 70 db. Sound intensity was held constant, regardless of the loudspeaker position and the type of sound, by adjusting the sound intensity before each experimental condition by means of a sound meter. Gaze direction was not monitored because a differential bias of gaze orientation should not be expected with respect to the type of stimulation (tactile versus auditory), the type of sound (white noise versus pure tone) and spatial location (near versus far). CA was tested under nine different conditions: a unimodal tactile condition and eight cross-modal auditory tactile conditions, one for each combination of loudspeaker location (right front or right back space), distance of the loudspeaker from the patient s head (near and far) and type of sound

4 100 E. Làdavas, F. Pavani and A. Farnè possible responses. A second experimenter recorded her response after each trial. CA was tested in three separate sessions, run on three consecutive days. In each session she was tested in the unimodal tactile condition and in the eight cross-modal conditions. Each condition was composed of 40 randomized trials: 10 single left stimulation, 10 single right stimulation, 10 bilateral stimulation and 10 catch trials. The conditions were blocked, and their order was randomized between sessions. Results of the neck, an acoustic stimulus presented near the right ear strongly interfered with the processing of a tactile stimulus simultaneously presented on the left side of the neck (crossmodal auditory tactile extinction). In contrast, only a weak interference of audition on touch perception was observed when the same acoustic stimulus was presented far from the patient s right ear, despite the fact that the auditory stimulus intensity at the patient s ear remained constant. This finding clearly shows that cross-modal auditory tactile extinction was mainly segregated into the peripersonal space surrounding the head. This spatially specific pattern of results is what should be expected if an ipsilesional acoustic stimulus was processed in an integrated auditory tactile system coding head peri- personal space, like the one described in monkeys (Graziano et al., 1999). Because of this integrated system, a right acoustic stimulus presented near the head strongly activates the right somatosensory representation of the head, which then competes with the left somatosensory representation of the head, activated by a tactile stimulus. In patients with left tactile extinction, competition is biased in favour of the ipsilesional stimulus that, as a result, extinguishes the contra- lesional stimulus whose representation is assigned with weaker competitive weights (Ward et al., 1994; di Pellegrino et al., 1997b). In contrast, an acoustic stimulus that falls outside the peripersonal space weakly activates the somatosensory representation of the ipsilesional side. As a consequence, competition is much reduced, as documented by the reduction in cross-modal auditory tactile extinction found far from the patient s head. Interestingly, this pattern of results emerged only when a white noise burst was presented. Although pure tones produced a mild cross-modal extinction, this effect did not appear to be segregated in the peripersonal space. This different pattern of results found with white noise and pure tone is in accordance with the neurophysiological observation reported by Graziano et al. (1999) showing that a pure tone does not activate multimodal neurons responsible for coding auditory peripersonal space near the head. One way to interpret the different effects of pure tone and white noise on the peripersonal space is by considering the functional role of ventral premotor (PMv) neurons which might directly encode the location of the sound with respect to the body (Graziano et al., 1999). Because these neurons are not activated by pure tones, the distance between near and far tones cannot be estimated, and, as a consequence, a segregated cross-modal extinction should not occur. Although this selective activation of multimodal neurons in the PMv can appear surprising, it might be the result of a normal adaptive process. Because the responses of monkey multimodal neurons do not change after repeated series of stimulation (Graziano et al., 1997), it has been suggested that their functional properties are hardwired, and the spatial correspondence between RFs can be calibrated through experience, perhaps within a critical period early in life (Salinas and Abbot, 1995; Graziano et al., 1997). In the case Patient CA never responded to catch trials and she showed a strong unimodal tactile extinction. Left tactile stimuli were missed in most bilateral trials (17% correct responses), whereas they were correctly detected when presented alone (100%; P by Fisher s exact test). A substantial extinction phenomenon also emerged in cross-modal condi- tions. Overall, the left tactile stimulus was detected in 59% of trials when presented together with a simultaneous sound on the right, and was detected in 97% of trials when presented alone (P by Fisher s exact test). Most important, cross-modal extinction varied according to the type of sound and according to the sound position with respect to the patient s head (see Fig. 2). White noise When the sound was a burst of white noise a significant decrement in tactile detection was always observed in bilateral trials compared with unilateral trials (see Fig. 2; P 0.05 by Fisher s exact test in all comparisons), revealing a consistent cross-modal extinction in all conditions. However, crossmodal extinction was more pronounced when the sound source was near the patient s head (33%) than when it was far from it (75%; P by Fisher s exact test). As can be seen from Fig. 2, the larger cross-modal extinction for near space with respect to far space was found in both back (33 versus 77%, P 0.001) and front space (33 versus 73%, P 0.002). Pure tone When the sound was a pure tone, a significant cross-modal extinction was again observed in all conditions (see Fig. 2, all P 0.05 by Fisher s exact test). However, cross-modal extinction was no longer modulated by the proximity of the sound source to the patient s head (near 62%; far 67%; P 0.7 by Fisher s exact test), in either front (near 63%; far 63%) or back space (near 60%; far 70%). Discussion The results of the present study demonstrate the existence in humans of a peripersonal auditory space around the head coded by an integrated auditory and tactile system. In a rightbrain damaged patient with left tactile extinction at the level

5 Auditory peripersonal space in humans 101 of auditory tactile neurons, the obvious crucial experience representation of the peripersonal space, however, does not through which the spatial calibration is achieved consists of necessarily predict this outcome. To find this type of repeated exposition to ecological sounds which are more extinction, all or most of the brain structures containing similar to a white noise burst than to a pure tone. Indeed, auditory tactile neurons should be damaged by the lesion. CA spontaneously referred to a white noise as an air puff. This would produce a misrepresentation of the auditory Thus, the lack of effect operated by a pure tone might reflect contralesional peripersonal space, and, as a consequence, a sort of impenetrability of the integrated auditory tactile patients should extinguish auditory stimuli presented near system to a sound that has little chance of occurring in nature. the contralesional side of the body. In contrast, the functional Another way to interpret the differential effects produced integrity of these areas is a necessary condition for coding by the two acoustic stimuli is by conceiving that distance in peripersonal auditory space and for inducing the modulatory depth is not directly coded by multimodal neurons, but by effects of audition on touch perception found in the present different cerebral structures coding different distant cues. In study. the present study, the main auditory cue for sound localization Finally, in addition to a cross-modal effect spatially in depth (i.e. sound intensity) was minimized by keeping the segregated in the peripersonal space, we found evidence of sound intensity at the patient s ears constant. In the absence a more generalized cross-modal competition between audition of intensity cues, other auditory cues like reverberation of and touch. Although auditory events presented in the the sound from the wall, familiarity with the sound source extrapersonal space did not compete with left tactile stimuli and interaural time differences [for a review see Moore and as did the auditory events presented in the peripersonal space, King (1999)] become relevant. However, these auditory cues they produced a mild cross-modal extinction. The attenuated, depend upon the waveform of the stimulus, and are more but still present cross-modal extinction found in the extra- informative for broadband noise than pure tones (Moore, personal space, can be explained by referring to the integrated 1995). As a consequence, white noise can be localized competition model proposed by Duncan (1996). As predicted with a higher spatial resolution than pure tones. Thus, the by this model, the imbalance produced by a unilateral lesion differential effect of white noise and pure tone in activating within a given modality (e.g. touch) would affect other auditory peripersonal space might be due to the fact that modalities (e.g. audition), although to a lesser degree, because pure tone stimuli cannot be localized effectively in depth. It the competitive advantage spreads to intact areas that are would be interesting to test this hypothesis, by asking patients interconnected to the damaged one. to judge the relative location of near and far space, and by An alternative way to explain the weak cross-modal effect analysing the relative influence of different distance cues found in the extrapersonal space is by referring to the fact in determining a possible spatially segregated effect for that, as for visuo-tactile neurons coding visual peripersonal pure tones. space around the face (Duhamel et al., 1991, 1998; Graziano A further important aspect of the head peripersonal space and Gross, 1995), neurons coding auditory peripersonal space found in monkeys, is that tactile RFs of multimodal neurons around the head also have a gradient of firing that varies as are rarely restricted to the back of the head and instead a function of stimulus distance. That is, the activation of include part of the cheek, eyebrows, snout and jaw, thus some auditory tactile cells is greater at closer distances, forming a complete representation of the space around being reduced at longer distances. It is therefore possible the head (Graziano et al., 1999). In agreement with these that the mild cross-modal extinction found in the extrapersonal neurophysiological findings, the amount of cross-modal space reflects the reduced response of the neurons extinction found with white noise bursts presented near the to stimuli presented far from the head. patient s head was comparable in the front and back space. A series of neuropsychological and neurophysiological The cross-modal effect described in the present study was evidence converge in showing that human and non-human found despite the fact that CA was blindfolded and, thus, primate brains share similar multisensory systems devoted did not have any visual cue about the spatial source of to the integrated coding of somatosensory and visual auditory stimuli. This suggests that the auditory peripersonal information in the peripersonal space. Indeed, recent neuro- space can be coded by an integrated system that directly psychological investigations found, in patients with tactile controls auditory and tactile stimuli, and that visual inputs extinction, the existence of a cross-modal visuo-tactile are not necessary for achieving bimodal auditory tactile extinction. In particular, the severity of cross-modal visuotactile integration. This view is further supported by the fact that extinction was found to vary as a function of the bimodal auditory tactile integration has been found in the distance between the visual stimulus and a given body part. back space of monkeys, whereby the animal does not have Extinction of a left tactile stimulus delivered to a patient s any visual experience. contralesional hand or cheek was more severe when a If the auditory tactile extinction is driven by an auditory concurrent visual stimulus was presented in the peripersonal tactile representation, one interesting question to address is space than in the extrapersonal space (Làdavas et al., whether, in patients with tactile extinction, one could expect 1998a,b, 2000). contralesional auditory extinction following ipsilesional tact- The results of the present study considerably extend our ile stimulation. The concept of an integrated auditory tactile knowledge of the multisensory integration occurring in the

6 102 E. Làdavas, F. Pavani and A. Farnè peripersonal space, by showing the existence in humans of References an auditory space representation surrounding the head. The finding that multimodal integration involves not only Bender MB. Disorders in perception. Springfield, IL: Charles C. Thomas, Damasio H, Damasio AR. Lesion analysis in neuropsychology. Oxford: Oxford somatosensory and visual inputs, but also somatosensory and University Press, auditory inputs has important implications relative to the De Renzi E. Disorders of spatial exploration and cognition. New York: functional role played by peripersonal space representations. Wiley, As far as the visual tactile input is concerned, both animal di Pellegrino G, Basso G, Frassinetti F. Spatial extinction to double asynchronous stimulation. Neuropsychologia 1997a; 35: and human studies suggest that the integrated processing of di Pellegrino G, Làdavas E, Farnè A. Seeing where your hands are. Nature this information in the peripersonal space would allow the 1997b; 338: 730. detection and localization of an object approaching the head Duhamel JR, Colby CL, Goldberg ME. Congruent representation of visual and somatosensory space in single neurons of monkey ventral intra-parietal (or the hand), before contact with the skin occurs, and allow area (area VIP). In: Paillard J, editor. Brain and space. Oxford: Oxford preparation of an adequate response to it (Graziano and University Press, 1991: Gross, 1998; Rizzolatti et al., 1998). Depending upon its Duhamel JR, Colby CL, Goldberg ME. Ventral intraparietal area of the nature, the object could either be avoided or reached and macaque: congruent visual and somatic response properties. Journal of Neurophysiology 1998; 79: grasped. Duncan J. Coordinated brain systems in selective perception and action. In: The functional role of visual peripersonal space representa- Inui T, McLelland JL, editors. Attention and performance. Cambridge, MA, tion is better understood if we consider the functional USA: MIT Press, 1996: organization of bimodal visual tactile neurons. Cells in the Fogassi L, Gallese V, Fadiga L, Luppino G, Matelli M, Rizzolatti G. Coding of peripersonal space in inferior premotor cortex (area F4). Journal of putamen, ventral pre-motor and in parietal areas have motor Neurophysiology 1996; 76: as well as sensory functions. Thus, these areas can be best Gentilucci M, Scandolara C, Pigarev IN, Rizzolatti G. Visual responses in the conceived as sensory motor interfaces, which help to encode postarcuate cortex (area 6) of the monkey that are independent of eye the location of sensory stimuli and to generate the motor position. Experimental Brain Research 1983; 50: Gentilucci M, Fogassi L, Luppino G, Matelli M, Camarda R, Rizzolatti G. responses to those stimuli. The same neurons often discharge Functional organization of inferior area 6 in the macaque monkey. I. during movements of the head and movements of the arm Somatotopy and the control of proximal movements. Experimental Brain directed towards and away from the body. Because of the Research 1988; 71: Graziano MSA, Gross CG. A bimodal map of space: tactile receptive spatial overlapping between visual and tactile RFs, they can fields in the macaque putamen with corresponding visual receptive fields. control head and arm movements not only on the basis Experimental Brain Research 1993; 97: of cutaneous information, but also on the basis of visual Graziano MS, Gross CG. The representation of extrapersonal space: a possible information about the object s position relative to the head role for bimodal, visual-tactile neurons. In: Gazzaniga MS, editor. The cognitive neuroscience. Cambridge, MA: MIT Press, 1995: (or hand). Graziano MS, Gross CG. Spatial maps for the control of movement. Current In some circumstances, however, objects moving towards Opinion in Neurobiology 1998; 8: the body are not visible, as in the dark or when they come Graziano MSA, Hu XT, Gross CG. Visuospatial properties of ventral premotor from the back. In these cases, the object s presence can be cortex. Journal of Neurophysiology 1997; 77: Graziano MSA, Reis LAJ, Gross CG. A neuronal representation of the location revealed only by the noise it may produce in approaching of nearby sounds. Nature 1999; 397: the body. On the basis of the results of the present study, we Làdavas E, di Pellegrino G, Farnè A, Zeloni G. Neuropsychological evidence suggest that the auditory peripersonal space representation of an integrated visuotactile representation of peripersonal space in humans. could allow the detection and localization of non-visible Journal of Cognitive Neuroscience 1998a; 10: Làdavas E, Zeloni G, Farnè A. Visual peripersonal space centred on the face objects approaching the head through the noise they produce, in humans. Brain 1998b; 121: before they reach the body. This in turn would help in Làdavas E, Farnè A, Zeloni G, di Pellegrino G. Seeing or not seeing where preparing an appropriate orienting or avoiding reaction in your hands are. Experimental Brain Research 2000; 131: Moore BCJ, editor. Hearing. San Diego: Academic Press, response to these objects. This function would be especially Moore DR, King AJ. Auditory perception: The near and far of sound relevant in the case of objects approaching the head from localization. Current Biology 1999; 9: R the back, as they are always outside visual control. Rizzolatti G, Scandolara C, Matelli M, Gentilucci M. Afferent properties of In conclusion, the results of the present study constitute periarquate neurons in macaque monkeys. II. Visual responses. Behavioral Brain Research 1981; 2: an important piece of evidence showing that monkeys and Rizzolatti G, Luppino G, Matelli M. The organization of the cortical motor humans peripersonal space representations are similarly system: new concepts [Review]. Electroencephalography and Clinical coded through the integrated processing of multimodal Neurophysiology 1998; 106: sensory information involving, at least, touch, vision and Salinas E, Abbot LF. Transfer of coded information from sensory to motor networks. Journal of Neuroscience 1995; 15: audition. Ward R, Goodrich S, Driver J. Grouping reduces visual extinction: neuropsychological evidence for weight linkage in visual selection. Visual Cognition 1994; 1: Acknowledgements We are most grateful to Giacomo Rizzolatti for valuable comments on an earlier version of the manuscript. We wish to thank Daniele Maurizzi for his technical support. This work was supported by grants from MURST and CNR. Wilson B, Cockburn J, Halligan P. Development of a behavioural test of visuospatial neglect. Archives of Physical Medicine and Rehabilitation 1987; 68: Received on 6 March, 2000; resubmitted on 19 April, 2000; accepted on 31 July, 2000

7 Auditory peripersonal space in humans 103 Auditory peripersonal space in humans: a case of auditory tactile extinction E. Ladàvas, F. Pavani and A. Farnè Abstract Animal experiments have shown that the spatial correspondence between auditory and tactile receptive fields of ventral pre-motor neurons provides a map of auditory peripersonal space around the head. This allows neurons to localize a near sound with respect to the head. In the present study, we demonstrated the existence of an auditory peripersonal space around the head in humans. In a right-brain damaged patient with tactile extinction, a sound delivered near the ipsilesional side of the head extinguished a tactile stimulus delivered to the contralesional side of the head (cross-modal auditory tactile extinction). In contrast, when an auditory stimulus was presented far from the head, cross-modal extinction was dramatically reduced. This spatially specific cross-modal extinction was found only when a complex sound like a white noise burst was presented; pure tones did not produce spatially specific crossmodal extinction. These results show a high degree of functional similarity between the characteristics of the auditory peripersonal space representation in humans and monkeys. This similarity suggests that analogous physiological substrates might be responsible for coding this multisensory integrated representation of peripersonal space in human and non-human primates. Journal Neurocase 2001; 7: Neurocase Reference Number: O217 Primary diagnosis of interest Segregated auditory tactile extinction in the peripersonal space Author s designation of case CA Key theoretical issue d Auditory peripersonal space in humans revealed by segregated auditory tactile extinction Key words: tactile extinction; auditory tactile extinction; cross-modal interactions; head centred space; spatial cognition Scan, EEG and related measures CT brain scan Standardized assessment Mini Mental State Examination, letter cancellation, bell test, picture scanning, menu reading Other assessment Tactile extinction, auditory tactile extinction Lesion location d Right unilateral lesion involving frontal, parietal and temporal lobes Lesion type Ischaemic disease Language English