1
news and views Discriminating native sounds: language-specific brain responses in infants Most adults find it difficult to hear the differences among sounds that do not occur in their own language. How do children learn to perceive the sounds of their native language more easily than the sounds of a foreign one? A study by Marie Cheour and colleagues (on page 351) reports the first neural correlate of this learning in infants. The authors examined mismatch negativity (measured by scalp electrodes, see photo), which is an electrical signal produced in response to an unexpected auditory stimulus in a series of repeating stimuli (in this case, the vowel /e/). Six-month-old Finnish infants had a slightly smaller mismatch negativity response to the Finnish vowel /Ö/ (which also occurs in Estonian) than to the uniquely Estonian vowel /Õ/. By one year, the same Finnish infants responded much more strongly to the vowel from their native language, whereas Estonian children of the same age perceived both sounds equally well. This study shows that language-specific memory traces in the human brain emerge between six months and one year of age. Kalyani Narasimhan © 1998 Nature America Inc. • http://neurosci.nature.com c. • http://neurosci.nature.com

document

  • Upload
    kalyani

  • View
    213

  • Download
    1

Embed Size (px)

Citation preview

news and views

nature neuroscience • volume 1 no 5 • september 1998 343

As a consequence of its physical nature,sound unfolds over time. This presentsunique problems for the nervous system,which is faced with the task of extractingand encoding information contained invibrating air molecules over short timeperiods, as well as integrating successivesonic events over longer time frames. Theauditory system is therefore highly special-ized for processing temporal events, whichconstitute the lowest common denomina-tor of everything from the bark of a dog toa Bach cantata. The brain imaging study

by Griffiths and colleagues in this issue ofNature Neuroscience (pp 422–427) exam-ines how the human brain may process thetemporal structure contained in a musicalsound. They identify dif-ferent cortical regionsinvolved in the processingof short-term (pitch com-putation) and longer-term (melodic pattern)temporal information.

These investigators took advantage of aphenomenon described some three hun-dred years ago by Huygens1, who notedthat the periodic reflections of the noisemade by a fountain from the stone steps ofa staircase resulted in an audible pitch (Fig.1). The same phenomenon can be repro-duced and studied in the laboratory by tak-

How do our brains analyzetemporal structure in sound?Robert Zatorre

How does the human brain process the temporal structure of amusical sound? A PET imaging study identifies cortical regionsinvolved in pitch computation and melodic pattern recognition.

Robert Zatorre is at the Department ofNeuropsychology, Montreal NeurologicalHospital, 3901 University,Montreal PQ H3A 2B4 Canadaemail: [email protected]

ing a sample of random noise and passingit through a cascade of delay-and-add net-works. That is, the noise is displaced by abrief time delay and added to itself; thenthe output of that procedure is again addedto itself with the same delay, and so on forsome number of iterations. This processresults in an audible pitch correspondingto the reciprocal of the time delay constantused (ref. 1 and Fourcin, A., Fifth Intl. Con-gress Acoust., B42, 1965). This pitch sensa-tion is of particular interest because it isperceived despite the absence of any audi-tory spectral cues that could activate a par-ticular frequency representation in thecochlea, in contrast to a typical complextone. Instead, the perception of pitch mustresult from the central nervous system’s

capacity to encode temporal regularity inthe neural firing pattern.

Since the time of Helmholtz, it has beencontroversial whether temporal or spectralinformation is the basis for the sensationof pitch. This duality arises because tem-poral regularity in a signal, or periodicity,gives rise both to temporally regular neur-

Discriminating native sounds: language-specific brain responsesin infantsMost adults find it difficult to hear thedifferences among sounds that do not occur intheir own language. How do children learn toperceive the sounds of their native languagemore easily than the sounds of a foreign one? Astudy by Marie Cheour and colleagues (on page351) reports the first neural correlate of thislearning in infants. The authors examinedmismatch negativity (measured by scalpelectrodes, see photo), which is an electricalsignal produced in response to an unexpectedauditory stimulus in a series of repeating stimuli(in this case, the vowel /e/). Six-month-oldFinnish infants had a slightly smaller mismatchnegativity response to the Finnish vowel /Ö/(which also occurs in Estonian) than to theuniquely Estonian vowel /Õ/. By one year, thesame Finnish infants responded much morestrongly to the vowel from their native language, whereas Estonian children of the same age perceived both sounds equally well.This study shows that language-specific memory traces in the human brain emerge between six months and one year of age.

Kalyani Narasimhan

...seeking areas whose activitycorrelated... with increasing

salience of the melody.

© 1998 Nature America Inc. • http://neurosci.nature.com©

199

8 N

atu

re A

mer

ica

Inc.

• h

ttp

://n

euro

sci.n

atu

re.c

om