Early-blind human subjects localize sound sources better than sighted subjects

N. LESSARD*, M. PARɆ, F. LEPORE*† & M. LASSONDE*†

* Groupe de Recherche en Neuropsychologie Expérimentale, Département de Psychologie, and † Centre de Recherche en Sciences Neurologiques, Université de Montréal, C.P. 6128, Succ.Centre-ville, Montréal, Québec H3C 3J7, Canada

Do blind persons develop capacities of their remaining senses that exceed those of sighted individuals? Besides anecdotal suggestions, two views based on experimental studies have been advanced¹. The first proposes that blind individuals should be severely impaired, given that vision is essential to develop spatial concepts². The second suggests that compensation occurs through the remaining senses, allowing them to develop an accurate concept of space³. Here we investigate how an ecologically critical function, namely three-dimensional spatial mapping, is carried out by early-blind individuals with or without residual vision. Subjects were tested under monaural and binaural listening conditions. We find that early-blind subjects can map the auditory environment with equal or better accuracy than sighted subjects. Furthermore, unlike sighted subjects, they can correctly localize sounds monaurally. Surprisingly, blind individuals with residual peripheral vision localized sounds less precisely than sighted or totally blind subjects, confirming that compensation varies according to the aetiology and extent of blindness⁴. Our results resolve a long-standing controversy in that they provide behavioural evidence that totally blind individuals have better auditory ability than sighted subjects, enabling them to compensate for their loss of vision.

We examined how subjects with congenital deficits affecting the peripheral visual system (retina and optic nerve) localized sounds in space. Four groups were tested: totally blind subjects (n = 8); blind subjects with residual vision in the peripheral field (n = 3); normally sighted but blindfolded controls (n = 7); and sighted controls (n = 29). Subjects were asked to localize a sound source presented on the horizontal plane. The sounds were delivered randomly through 16 loudspeakers mounted on a semicircular perimeter. Peri-central field was defined arbitrarily as the space covered by the four centrally located loudspeakers (up to 16° on either side of the midline), whereas the lateral fields extended to 78°. Subjects were tested under monaural and binaural conditions, each providing a specific set of cues to sound localization. Binaural cues refer to the discrepancies of inputs between the ears in terms of timing and intensity, whereas monaural cues arise from the spectral filtering of sounds by the circonvolution of the pinna.

The sound-localization performance in the binaural condition is plotted in Fig. 1. Because a preliminary analysis showed that the mean error scores in each field were not different for blindfolded and sighted controls (P > 0.05), their results were pooled (Fig. 1a). The principal results indicate that totally blind subjects were at least as accurate as sighted controls (Fig. 1c), although because of a ceiling effect it was not possible to show that they were actually better than the latter. To bring out these differences, it may be necessary to render the task more difficult by using, for example, narrow-band stimuli. The results nonetheless demonstrate that blind individuals can develop a three-dimensional map of space using auditory information.

This raises questions regarding the role of vision in calibrating auditory space. Experience-dependent plasticity in the visual calibration of the auditory space map has been demonstrated in the barn owl's optic tectum⁵, where neurons form a map of auditory space based on interaural time and level differences. When visual and auditory worlds are experimentally misaligned by rearing owls with prismatic spectacles that displace the visual field in azimuth, tectal neurons acquire responses to interaural time differences that correspond to the displaced visual fields and the neurons abandon responses to normal interaural time differences. This seems to support the notion that vision is essential for calibrating space, including auditory space. Our results, however, indicate clearly that this is not the case with humans who totally lack the sense of vision, as our early-blind subjects develop excellent spatial representation in the absence of this modality.

Blind subjects with residual vision were less accurate than all other subjects, particularly in the pericentral field (Fig. 1b). This result was unexpected as it had been predicted that these subjects would show normal localization behaviour in peripheral fields (where vision was present), and a performance similar to that of the early-blind subjects in central visual field (where vision was lacking). Their subnormal performance across the entire auditory field has three possible explanations. First, these subjects would need to develop an auditory map of space in part supported by vision (peripheral field) and in part independent of vision (central field), which might cause some confusion. Second, if auditory compensation in blind subjects depends on the recruitment of the deafferented sensory areas (see below), the latter would not show a similar amount of plasticity if they were still stimulated, albeit at a reduced rate, by their normal afferences. Third, these partially blind subjects demonstrated abnormal orienting behaviours, which might interfere with correct localization, as they often would fixate the source of a sound (the voice of the experimenter, the test sound during pretest, when head movement was allowed) by turning their head so that the source would be visible by their residual visual field.

In the monaural condition, where the stimuli were presented while one ear was blocked, sighted and blindfolded controls were once again indistinguishable from each other (P > 0.05) and their results were pooled. Figure 2a shows that these subjects localized the sound on the side of the unobstructed ear, approximately at mid-field, independently of the side from which it originated. As in the binaural condition, the performance of blind subjects with residual vision was worse than that of the others (Fig. 2b). They also showed the positional bias in favour of the unobstructed ear when localizing a sound presented on the side of the obstructed ear. The performance of the totally blind subjects, however, was quite exceptional. Half (Fig. 2c) appeared to respond like controls, that is, with a positional bias favouring the side of the unobstructed ear. However, contrary to controls, their performance showed more variability when the stimulus was presented on the side of the obstructed ear. Moreover, they often reported that the sound presented on this side seemed qualitatively different. For example, one subject reported that the sound seemed to show a narrower bandpass, as if some frequencies were cut off. The most interesting aspect of this set of results was the fact that the other four totally blind subjects localized the sound on the appropriate side--that is, even when it was presented on the side of the obstructed ear.

The ability to localize the sound in the correct hemifield, irrespective of accuracy, in both the monaural and binaural testing conditions is illustrated in Fig. 3. The most striking observation concerns the totally blind subjects: half of them localized correctly the sound presented on the side of the obstructed ear. This behaviour, moreover, was neither marginal, as performance was nearly perfect, nor was it observed in any of the large number of sighted subjects (29 normally stimulated, 7 tested while blindfolded). This finding, coupled with the observation that the other four totally blind subjects, although unable to localize on the correct side, did report qualitative differences in the sound presented to the blocked ear--again a behaviour not seen in any of the other groups--suggests that blind individuals utilize monaural cues more efficiently to auditorily explore their environment.

The finding that early-blind subjects can perform better than sighted subjects can be attributed to reorganizations in neuronal populations involved in processing localization cues and/or to improved learning. Thus, compensation may result from the increased use of spectral information within or between the structures normally involved in spatial analysis, including the superior^6,7,8 and inferior colliculus⁹, as well as primary auditory cortex^10,11. Alternately, compensation may occur through the recruitment of brain structures left unused by the lack of visual input. In cats, visual deprivation beginning shortly after birth results in compensatory effects at the collicular level¹², improved auditory responses in neurons involved in visual processing¹³, and superior localization performance¹⁴. Similarly, in early-blind humans, Kujala and colleagues^15,16 showed that, using event-related potentials or magnetoencephalography, the response to auditory stimuli was distributed more posteriorly compared with sighted controls, suggesting increased use of parietal or perhaps even occipital brain areas in sound localization. They propose that this cross-modal reorganization might rest on the unmasking¹⁷ of latent, pre-existing auditory connections to the visual cortex. In addition, better learning strategies could explain their superior performance, as psychophysical studies have shown that normal subjects fitted with an earplug, who demonstrate a prominent displacement in their localization judgement towards the side of the open ear, can increase their precision with learning¹⁸. Similarly, Slattery and Middlebrooks¹⁹ studied sound localization in five unilaterally deaf subjects and showed that three of them had little or no localization bias towards the functional ear. This implies the development of strategies for localizing sounds monaurally, which might likewise be adopted by blind persons.

Methods
The apparatus used to test sound localization, described in detail elsewhere²⁰, consisted of 16 loudspeakers mounted on a graduated semicircular perimeter (radius 50 cm) positioned at 5°, 16°, 26°, 37°, 47°, 58°, 68° and 78° on either side of the mid-sagittal plane. The subject was seated in the centre of the perimeter, the head placed on a head-rest attached to the chair, and the speakers were positioned at ear level. All testing was done in an anechoic chamber. The stimuli were two broad-band noise bursts that lasted 30 ms (10-ms rise and 10-ms fall time) with an interburst interval of 30 ms (total stimulus duration 90 ms). The sound pressure level (SPL) was maintained at 40 dB SPL (20 Pa). The stimulus was delivered through a randomly selected loudspeaker and repeated five times for each position. The head was not restrained other than being apposed on the head-rest, but a warning buzzer told the subjects that a stimulus was about to be presented and that they should maintain stable head position and fixate straight ahead. Only a subgroup of control subjects were blindfolded; the others, including the partly blind subjects, were free to have their eyes open. Compliance to all instructions was verified by an experimenter who stayed in the chamber and remained behind the subject. The speaker array was covered with black cloth so that sighted subjects could not see the actual speakers and response consisted of pointing with the dominant hand to the apparent source of stimulation. Lines graduated in 1° steps were drawn on the perimeter and the experimenter recorded the response of the subject. For monaural testing, either the right or the left ear was plugged with a soft foam earplug (mean attenuation = 37.5 dB SPL) and covered by a hearing protector muff (mean attenuation, 29 dB SPL). To ensure that no sounds could be perceived with the earplug and hearing protector muff in place, a preliminary study was done to test the efficacy of the earplugs during which both ears were blocked. Broad-band noise bursts ranging from 25 to 60 dB SPL were delivered randomly from four different positions (16° and 68°) for a total of 20 presentations. No subject reported hearing a stimulus presented within these ranges.

Received 23 February;accepted 19 June 1998.

References

Rauschecker, J. P. Compensatory plasticity and sensory substitution in the cerebral cortex. Trends Neurosci. 18, 36-43 (1995). Links

Save Citation

Axelrod, S. Effects of Early Blindness (Am. Found. Blind, New York, 1959). Links

Save Citation

Rice, C. E. Early blindness, early experience and perceptual enhancement. Res. Bull. Am. Found. Blind 22, 1-22 (1970). Links

Save Citation

Kujala, T. et al. Electrophysiological evidence for cross-modal plasticity in humans with early- and late-onset blindness. Psychophysiology 34, 213-216 (1997). Links

Save Citation

Feldman, D. E. & Knudsen, E. I. An anatomical basis of visual calibration of the auditory space map in the barn owl's midbrain. J. Neursci. 17, 6820-6837 (1997). Links

Save Citation

Chalupa, L. M. & Rhoades, R. W. Responses of visual, somatosensory, and auditory neurones in the golden hamster's superior colliculus. J. Physiol. 270, 595-626 (1977). Links

Save Citation

King, A. J. A map of auditory space in the mammalian brain: neural computation and development. Exp. Physiol. 78, 559-590 (1993). Links

Save Citation

Middlebrooks, J. C. & Knudsen, E. I. A neural code for auditory space in the cat's superior colliculus. J. Neurosci. 4, 2621-2634 (1984). Links

Save Citation

Spitzer, M. W. & Semple, M. N. Interaural phase coding in auditory midbrain: influence of dynamic stimulus features. Science 254, 721-723 (1991). Links

Save Citation

Stumpf, E., Toronchuk, J. M. & Cynader, M. S. Neurons in cat primary auditory cortex sensitive to correlates of auditory motion in three dimensional space. Exp. Brain Res. 88, 158-168 (1992). Links

Save Citation

Toronchuk, J. M., Stumpf, E. & Cynader, M. S. Auditory cortex neurons sensitive to correlates of auditory motion--underlying mechanisms. Exp. Brain Res. 88, 169-180 (1992). Links

Save Citation

Rauschecker, J. P. & Harris, L. R. Auditory compensation of the effects of visual deprivation in the cat's superior colliculus. Exp. Brain Res. 50, 69-83 (1983). Links

Save Citation

Rauschecker, J. P. & Korte, M. J. Auditory compensation for early blindness in cat cerebral cortex. J. Neurosci. 13, 4538-4548 (1993). Links

Save Citation

Rauschecker, J. P. & Kniepert, U. Auditory localization behaviour in visually deprived cats. Eur. J. Neurosci. 6, 149-160 (1994). Links

Save Citation

Kujala, T., Alho, K., Paavilainen, P., Summala, H. & Näätänen, R. Neural plasticity in processing of sound localization by the early blind: an event-related potential study. Electroencephalogr. Clin. Neurophysiol. 84, 469-472 (1992). Links

Save Citation

Kujala, T. et al. Auditory an somatosensory event-related potentials in early blind humans. Exp. Brain Res. 104, 519-526 (1995). Links

Save Citation

Jacobs, K. M. & Donoghue, J. P. Reshaping the cortical motor map by unmasking latent intracortical connections. Science 251, 944-947 (1991). Links

Save Citation

Butler, R. A. An analysis of the monaural displacement of sound in space. Percept. Psychophys. 41, 1-7 (1987). Links

Save Citation

Slattery, W. H. & Middlebrooks, J. C. Monaural sound localization: acute versus chronic unilateral impairement. Hearing Res. 75, 38-46 (1994). Links

Save Citation

Poirier, P., Miljours, S., Lassonde, M. & Lepore, F. Sound localization in acallosal human listeners. Brain 116, 53-69 (1993). Links

Save Citation

Acknowledgements. We thank the Regroupement pour les Aveugles et Amblyopes de Montréal (RAAM), especially S. Poulin and F. Boulet, for their assistance in recruiting participants. This work was supported by grants from NSERC and FCAR.

Correspondence should be addressed to F.L. (e-mail: leporef@ere.umontreal.ca).

---

Nature © Macmillan Publishers Ltd 1999 Registered No. 785998 England.
Back to Science and Society