Page 18 - Summer 2010

Fig. 3. Timing, pitch and harmonics describe complex acoustic signals: the acoustic waveform of “da” (blue) and its evoked brainstem response (red) on different time scales. A. Prominent timing landmarks of the stimulus, e.g., the onset, offset, and events during time-varying portions (arrows), evoke precisely synchronous and replicable electrical deflections in the auditory brainstem. For illustration purpos- es, the stimulus waveform in this figure has been delayed in time by 9 ms, to approx- imate the neural propagation time. This permits better visual coherence between stimulus and response. B. Several repeating periods of 10 ms each are shown. This imparts a pitch percept of 100 Hz, and this periodicity is mirrored in the response. C. Stimulus and response spectra in the frequency domain. Here, the stimulus has been filtered to mimic the response’s low-pass characteristic. Spectrum peaks for the stimulus and its evoked auditory brainstem response are exactly aligned, represent- ing their similarity in harmonics. (Artwork by Judy Song.)
in trained children, and seems to accelerate their develop-
26
opment between groups differed.
Additional evidence that music training is causing brain differences, rather than brain differ- ences leading to musical proclivity comes from a study in which very young children in matched musically-trained and untrained cohorts were followed for a year and brain devel-
27
ment by about three years.
talists who play different instruments reveals specialized acti-
vations. Gamma-band oscillatory activity in the brain is
strongest when induced by the sound of a musician’s own
Common and separate mechanisms
Turning back to reading and speech-in-noise perception, there are some noteworthy similarities in the skills required for these tasks and for playing a musical instrument. Two such skills, more on the cognitive end of the spectrum, are attention and working memory.
Another crucial need for all three endeavors is accurate processing of incoming auditory signals. The spoken word and music can be thought to consist of three fundamental components: pitch, harmonics, and timing. These three com- ponents of any acoustic signal can be differentiated by their time scales and carry different informational content. In speech, timing and harmonics convey the phonetic content— specific consonants and vowels—of non-tonal languages such as English, and thus are mainly responsible for the ver- bal message. Pitch conveys intent (e.g., question versus state- ment) and plays a large role in distinguishing one talker from another. In tonal languages such as Mandarin, pitch also car- ries linguistic information. In music, one could argue for similar divisions and classifications, but it is all to easy to be trapped into stretching a metaphor. What is information in music? What is intent? Suffice to say, music, even an individ- ual note played by a single instrument, has a rich acoustical structure and is, by any definition, “complex.” Thus, brain- stem evoked responses to music and speech alike are rich sources in the investigation of music training’s role in shap-
a
comprising words is a crucial building block of reading. A consonant that has a particular voicing onset (timing) and a spectrum of a particular shape (harmonics) is eventually associated by a young reader with, for example, the letter T. The pitch of this combination of sounds—was it spoken by mom, dad, or the funny-looking purple creature on televi- sion?—does not affect its phonetic identity. Timing and har- monic features in speech are especially vulnerable in poor readers and pose particular perceptual challenges30-32 while pitch perception is generally intact.
Speech-in-noise perception, on the other hand, presents a different set of problems and a corresponding set of skills to accomplish it. Among these are keying in on location cues,
33-44
28
ly tuned to one’s own instrument.28,29
instrument.
Cortical evoked responses also are preferential-
ing the nervous system.
Forming phonological representations of the sounds
stream segregation, and grouping of the acoustic scene. Together, these are used to tag and follow the speaker’s voice, and rely on pitch43-48 as well as the timing and harmonic prop- erties of the signal.
Reading, SIN perception and music share a core set of skills—working memory, attention, perception of pitch, tim- ing and harmonics—and each also requires some unique skills. Reading requires the use of phonology and the devel- opment of a vocabulary corpus. SIN perception relies on object formation and grouping, stream segregation, and voice tagging. Music involves knowledge of melody, harmony, and rhythm. The skill sets are a mix of low-level sensory process- ing and high-level cognitive proficiency. A strong sensory- cognitive link seems to be a factor in proficiency across domains, and each requires the formation of sound-to-mean- ing connections. The intersection of common skills, as well
Research on instrumen-
Musician’s Auditory World 17