Page 19 - Summer 2010

as the unique skill sets required for the three, make reading, SIN perception and musical experience an interesting tri- umvirate to pursue in the investigation of processing of complex sounds in the auditory brainstem. Brainstem activ- ity reflects the acoustical characteristics of signals very well—while plastic to language49-51 and music52-56 experience and short-term auditory training,57,58—making it well suited to provide objective physiological information about com- plex-sound encoding in populations with a range of reading, music and SIN perception skills (see Tzounopoulos and Kraus59 and Skoe and Krause145 for reviews). Encoding of pitch, timing, and harmonics is selectively diminished in certain clinical populations and selectively enhanced by expertise allowing us to examine specific, separable aspects of signal encoding. We are not observing simple gain effects; that is, overall response disruption or enhancement.
The approach: Music and speech evoked brainstem responses
In the past decade, the Kraus Lab has been a pioneer in the use of speech- and music-evoked auditory brainstem response (ABR) as an innovative objective marker of audito-
obligatory system of afferent fibers carrying sensory infor- mation from the cochlea to the cortex, but there is also an extensive system of descending efferent fibers that synapse all the way down to the outer hair cells of the cochlea,124 making plasticity in the brainstem not so implausible.
Now, back to the auditory signal. Many key perceptual ingredients of speech and music are driven by particular properties of the signal and have direct brainstem-response correlates. In response to syllables or musical notes, chords or melodies, the timing of the response provides information, on the order of fractions of milliseconds, about the onset and offset of the sounds (i.e., temporal envelope cues), and spec- trotemporal patterns in the evoking signal are revealed in
57,76,125
musical note.82,111,126
To tie it all together, the components of the brainstem
response that are measurably disrupted in poor readers and individuals with poor speech-in-noise perception are the very components that are enhanced in musicians. The con- nections among the three (music, reading, and speech-in- noise perception), as revealed by the auditory brainstem response, are covered in the next section.
What the brainstem has revealed about the musician’s subcortical processing
In this mini-review of recent Kraus-lab investigations, a theme emerges. The subcortical processing augmentations in musicians are selective. Not every aspect of the brainstem response is enhanced, and enhancement does not occur to every stimulus. First, we start with a straight-forward investi- gation of subcortical processing differences between musi- cians and non-musicians to musical sounds. Then we look at the degree to which these processing advantages extend to speech, and then to non-speech vocalizations.
55
ry function in a variety of populations.
49-58;60-113,131
Here, we will
present a synopsis of this approach and some advantages over
the more widely-studied cortical response and behavioral
b
widely studied in the auditory cortex of humans and experi- mental animals for consonants,114-116 vowels,117 and pitch.118 Structural and functional reorganization of auditory and sen- sorimotor cortex occur with musical training,18;27;29;119-121,137 and
122,123
paradigms.
Neural transcription of the acoustics of sound has been
Our focus, and the focus of this review, is on subcortical (auditory brain- stem) processing of complex sounds such as speech syllables, musical notes, chords and melodies. Unlike the more abstract representation of sound in the cortex, the brainstem response resembles and sounds like the evoking sound itself (Fig. 3). Moreover, responses are reliably stable, interpretable, and meaningful in individuals. The brainstem response paradigm is passive, and its objectivity represents a significant advance over typical measures of complex-sound processing. Most such measures are behavioral in nature, with the person repeating the words that they heard, or making judgments about melodic or rhythmic properties of a musical snippet. Active engagement of processes such as attention, memory, and motor coordination is required to perform the task. Likewise, cortical physiological measures are susceptible to non-sensory factors such as state, motivation, etc. Therefore, our objective brainstem measure is a particularly effective
non-sensory structures also appear to benefit.
tool at probing unadulterated auditory processing.
While objective, there is another property of the audito- ry brainstem that is crucial to its value as a window into audi- tory processing. As mentioned above, it is experience- dependent. On the surface, experience-dependency might seem a counterintuitive property if the principal purpose of the subcortical auditory system is the passive conveyance of acoustic information from receptor to auditory cortex for final and more complex processing. But, not only is there an
Enhancement to music I: Piano chords, Lee et al., 2009. Two-note sampled-piano chords, G-E and F#-E, were used to elicit brainstem responses in a group of adult musicians and an otherwise-matched group of non-musicians. Some inter- esting aspects of brainstem responses are that they represent the pitch of the evoking signal in their spectra, and that they reveal nonlinear processes by exhibiting frequency compo- nents that are not present in the stimuli—namely distortion products or combination tones. These stimuli provided a wealth of response properties for the investigation of musi- cal-signal processing in the rostral brainstem. Interesting findings arose from this investigation. First, of the two pri- mary notes that compose the chord, only representation of the harmonics of the higher note—in both cases E—differed between groups (Fig. 4). Because the musician enhancement was selective, we interpreted the fact that the higher note revealed the group difference as indicative of the relatively greater importance that the upper note typically plays in music. A performing musician is tuned into the melody which is often the highest note of a score. This finding also
18 Acoustics Today, July 2010
Analysis of the spectral con- tent of the response provides information about the funda- mental frequency, a major contributor to the perceived pitch of the signal as well as its harmonics, including the temporal fine structure of speech formants and the overtones of a
response timing and phase.