Cerebral Cortex November 2007;17:2725--2732
doi:10.1093/cercor/bhl182
Advance Access publication February 8, 2007
Persistent Responsiveness of Long-Latency
Auditory Cortical Activities in Response to
Repeated Stimuli of Musical Timbre and
Vowel Sounds
Shinya Kuriki, Keisuke Ohta and Sachiko Koyama
Long-latency auditory-evoked magnetic field and potential show
strong attenuation of N1m/N1 responses when an identical stimulus
is presented repeatedly due to adaptation of auditory cortical
neurons. This adaptation is weak in subsequently occurring P2m/P2
responses, being weaker for piano chords than single piano notes.
The adaptation of P2m is more suppressed in musicians having
long-term musical training than in nonmusicians, whereas the amplitude of P2 is enhanced preferentially in musicians as the spectral
complexity of musical tones increases. To address the key issues of
whether such high responsiveness of P2m/P2 responses to complex sounds is intrinsic and common to nonmusical sounds, we conducted a magnetoencephalographic study on participants who had
no experience of musical training, using consecutive trains of piano
and vowel sounds. The dipole moment of the P2m sources located
in the auditory cortex indicated significantly suppressed adaptation
in the right hemisphere both to piano and vowel sounds. Thus, the
persistent responsiveness of the P2m activity may be inherent, not
induced by intensive training, and common to spectrally complex
sounds. The right hemisphere dominance of the responsiveness to
musical and speech sounds suggests analysis of acoustic features
of object sounds to be a significant function of P2m activity.
mented in musically well-experienced people. When an acoustic feature is manipulated, the spectral complexity of piano
sounds increases the amplitude of the P2 response, but not N1,
preferentially in musically experienced people (Shahin et al.
2005). These observations of the response amplitude suggest
high susceptibility to modification of auditory cortical neurons
of the P2/P2m responses by acoustic/musical training. Slow
development of long-latency auditory responses in humans
(Paetau et al. 1995; Ponton et al. 2000) and plasticity of the
neurons in the auditory cortex revealed in animal studies on
learning effects (Edeline 1999; Weinberger 2004) may be the
basis of the training effects of P2/P2m responses.
The robustness of P2m activity to repetitive stimuli might be
related to object analysis in the ‘‘what’’ pathway of auditory information, which contrasts with spatial analysis in the ‘‘where’’
pathway (Romanski et al. 1999; Rauscheker and Tian 2000). For
N1/N1m activities, strong adaptation of feature-specific neurons in the posterior auditory cortex has been suggested to
serve as a novelty detection mechanism in the ‘‘where’’ pathway,
filtering out monotonous sounds in the environment but being
recovered quickly by a novel (deviant) sound to allocate attention to the sound source (May et al. 1999; Jääskeläinen et al.
2004). Assuming that the P2m neurons with robust activity are
involved in the ‘‘what’’ pathway of object analysis, we expect
that high responsiveness of P2m is observed commonly in ecologically relevant sounds, such as vowel speech, and is not
caused by intensive or long-term training but is intrinsic.
The results of various imaging studies indicate the existence
of a hemispheric difference in auditory responses to speech and
speech-like sounds. Left hemisphere dominance of cortical
activity appears when phonetic cues characterized by rapid
frequency change, that is, voice onset in consonant-vowels, are
processed (Poeppel et al. 1996; Belin et al. 1998; Kayser et al.
2001; Zattore and Belin 2001; Jäncke et al. 2002). This hemispheric specialization is supported by the anatomical asymmetry of the auditory cortical structure, which facilitates fast
temporal processing in the left hemisphere (Penhune et al. 1996;
Zatorre et al. 2002). However, for vowels with no rapid spectral
or temporal change, diverse results for N1m responses that do
not converge with respect to the hemispheric effect have been
reported in magnetoencephalographic (MEG) studies (Gootjes
et al. 1999; Mäkelä et al. 2003; Shestakova et al. 2004; Tiitinen
et al. 2005). Little is known about the hemispheric laterality of
the P2m response to the acoustic or phonetic feature of speech
sounds.
Here, we report the results of a study on the responsiveness
of N1m and P2m activities to repetitive stimulation of piano
tones and vowel sounds. Pure tones were used as control
stimuli. We employed a dichotic listening scheme to detect the
Keywords: adaptation, auditory cortex, auditory-evoked response, P2m,
magnetoencephalography
Introduction
The long-latency auditory-evoked potential of N1 (Ritter et al.
1968; Budd et al. 1998) and magnetic field of N1m (Hari et al.,
1982; Tiitinen et al., 1993) show strong attenuation of the
response when auditory stimuli are presented repeatedly due to
adaptation of auditory cortical neurons. The N1m response
reflects neural activities having multiple recovery times, of the
order of seconds, to the change in repetition period (Sams et al.
1993; Loveless et al. 1996). Occurring at a longer latency subsequent to N1m, the magnetic P2m response shows different
behaviors to repeated auditory stimuli. The attenuation of the
dipole moment of the P2m response is smaller for repetition of
chords than single tones of piano timbre (Kuriki et al. 2006), in
contrast to the dipole moment of the N1m response, which is
strongly adapted equally to chord and single piano tones. As
a result, notable P2m peaks are present in the responses to a
sequence of consecutive chords. Adaptation of the P2m activity
is more suppressed in individuals with long-term musical
training than in individuals without musical experience.
The amplitude of the potential P2 response is enhanced
by training in acoustic discrimination with complex sounds
(Tremblay et al. 2001; Atienza et al. 2002; Reinke et al. 2003).
The potential P2 (Shahin et al. 2003) and magnetic P2m (Kuriki
et al. 2006) responses elicited by musical timbre are also augÓ The Author 2007. Published by Oxford University Press. All rights reserved.
For permissions, please e-mail: [email protected]
Research Institute for Electronic Science, Hokkaido
University, Sapporo 060, Japan
N1m and P2m responses from the auditory cortex exclusively of
one hemisphere based on contralateral pathway dominance
(Kimura 1961, 1967). None of the participants in the study had
received musical training. We examined whether P2m responses
exhibit weak adaptation and/or high sensitivity to vowel sounds
as well as musical sounds, in comparison with the responses to
pure tones that have a single frequency component, and
whether specific responsiveness of the P2m response to a musical sound is observed in musically nonexperienced subjects.
Different responsiveness is expected for the N1m response to
these complex sounds. We also examined hemispheric effects
on the N1m and P2m responses, focusing on the issue of
whether any difference in the responsiveness to the complex
sounds and pure tones depends on the hemisphere.
Materials and Methods
Subjects
Five male subjects (mean age ± standard deviation [SD] of 24.2 ± 2.4
years) and 5 female subjects (22.2 ± 2.7 years) participated in the
experiment. The subjects were all right handed as evaluated with
Edinburgh test. They had never played an instrument and had never
received formal musical education except for music courses in elementary school and junior high school. All the subjects gave written informed consent to participate in the study after receiving a thorough
explanation of the MEG recording. The study was approved by the
Ethics Committee of the School of Medicine, Hokkaido University.
Stimuli and Presentation
Measurements were made in 2 separate sessions using variants of a piano
tone and vowel speech. In the piano session, an A3 piano tone
(fundamental frequency of 220 Hz) and pure tones of its harmonics
(440 and 880 Hz) were used. In the vowel session, a vowel speech sound
of /a/ pronounced by a male broadcast announcer and pure tones of 2
peak frequencies (238 and 365 Hz), which are included in the F1
formant of the vowel, were used. The contours of the pure tones used in
the piano and vowel sessions were shaped to follow the envelope of the
piano tone and the vowel speech, respectively. The stimuli consisted of
repetition of 3 identical tones or speech of 312.5 ms in duration and
a silence period of the same length. Figure 1 shows the waveforms and
frequency spectra of stimulus sounds. An oddball sound of white noise
was inserted in a stimulus sequence, replacing 1 of the 3 sounds in the
sequence. In each of the piano or vowel sessions, sequences of piano
tones or vowel sounds were mixed with pure-tone sequences of 2
frequencies in random order. The interstimulus intervals between the
sequences were 1000--1050 ms. Sequences that included the oddball
sound were also mixed with the normal stimuli sequences to about 10%
probability.
MEG Recording
The stimuli were delivered monaurally to subjects at a level of sound
pressure of 40 dB above the hearing threshold with magnetic-free earpieces and ear tubes. A 1/f pink noise of 75 dB in sound pressure level
was applied to the ear contralateral to the stimulation side. Prior to MEG
recording, the hearing threshold had been measured for each stimulus
sequence for the right ear and left ear stimulation of each subject. The
frequency characteristic of the sound delivery system, including an
amplifier and a transducer, was adjusted to within 2 dB from 20 Hz to 4
kHz by means of an equalizer. MEG measurements were made using a
custom-made helmet-shaped SQUID system (Elekta-Neuromag, Helsinki),
which has 76- and 19-channel magnetometers arranged on a lower layer
and upper layer, respectively, and 6 reference channels for noise
cancellation. The 2 sensor layers formed concentric spherical surfaces
separated by 4 cm. The lower layer of 76 channels, separated from each
other by 2.5 cm, was used in this study. Subjects were instructed to
respond to oddballs by movement of the right or left index finger, which
was detected by a laser-light switch. A single session of recordings lasted
for about 30 min with short intermissions. The piano and vowel sessions
were repeated for right ear and left ear stimulation in each subject. The
order of recordings of piano/vowel sessions and right/left ear stimulation was randomized across the subjects.
Data Analysis
MEG responses to oddball sequences and those exceeding 3 pT in
amplitude were excluded. More than 200 epochs of 2075 ms in duration, which included a 200-ms prestimulus period for the baseline,
were averaged. The averaged responses were filtered to 1--20 Hz.
Figure 1. Waveforms and frequency spectra of stimulus sounds. In the piano session of measurement, the A3 piano tone and pure tones of 440 and 880 Hz harmonics were used,
whereas the vowel /a/ and pure tones of 238 and 365 Hz of the first Formant were used in the vowel session.
2726 Auditory Responses to Complex Sounds
d
Kuriki et al.
The cutoff frequencies of 1 and 20 Hz were used to reduce the
low-frequency and power line (50 Hz) noises in environment, respectively, which existed at high amplitudes in the signals detected by
the magnetometer-type sensors. Effects of this narrow bandwidth were
examined by widening the cutoff frequencies of the filter to a passband
of 0.3--40 Hz. The waveforms of N1m and P2m responses were unchanged after this process, except a slight rounding of the peaks (see
Fig. 2b). An increase of the power line noise and a small shift of the
baseline appeared. These results may assure the selected bandwidth in
the analysis of measured signals.
To represent the waveform of response, the magnetic fields of 12
adjacent channels at the field maximum in the posterior temporal region
were used to calculate
the root-mean--squared (RMS) amplitude of
1=2
: Single equivalent current dipoles (ECDs) were
BRMS = +Bn2 =12
localized at the peaks of N1m and P2m responses using the field data
in the right and left hemispheres contralateral to the stimulated ear.
ECDs having more than 80% of the goodness-of-fit value were selected.
The grand mean of the coordinates of ECDs across subjects was
obtained for N1m and P2m responses from the selected ECDs of 8--9
subjects for each condition of stimuli and hemispheres. Then, the
waveform of the dipole moment of N1m and P2m responses was
obtained by taking an inner product of the measured fields in a single
hemisphere and the template dipole fields that were calculated from the
grand mean ECD (Kuriki et al. 2006). Different sources of the grand
mean N1m and P2m obtained from the 8--9 subjects were used in the
calculation of the template dipole fields. The values of correlation
between the measured and template fields in the calculation of this
subspace projection were more than 85%, mostly above 90% in the
responses to the first-stimulus sound. The general principle of the
subspace projection was described by Hämäläinen (1995) and Tesche
et al. (1995). This method enables quantitative evaluation of the dipole
moment in the absence of any subjective steps (Hertrich et al. 2004).
The magnitudes of the N1m and P2m components were determined as
the peak values in the time series of the dipole moment of each
component.
Figure 2. (a) Superposed MEG responses at 41 channels in the right hemisphere of
one subject and (b) time series of the dipole moment (black line) calculated from the
data of (a). The responses were measured in a piano session using 3 successive A3
tones. The moment was obtained by projecting the measured fields on the grand mean
dipole fields of N1m. The dipole moment calculated using the data, in which the
frequency range of MEG responses was widened from 1--20 to 0.3--40 Hz, was also
shown in (b), as indicated by a gray line.
To examine the effects of stimulus sound, analysis of variance
(ANOVA) and post hoc Scheffe’s tests were conducted for the dipole
moments of N1m and P2m responses. Because MEG measurements
were repeated for the right ear and left ear stimulations in piano and
vowel sessions, a quantitative comparison was made between the responses to piano/vowel sounds and pure tones, across which the
stimulation level was equated above the hearing threshold at each ear.
Thus, an ANOVA test within each hemisphere was carried out with the
factors of stimulus and stimulus order. To examine the attenuation of
the dipole moment by stimulus repetition, the dipole moments of the
second and third stimuli were normalized by the moment of the first
stimulus in individual subjects. The mean of the normalized moments of
the second and third stimuli was analyzed by ANOVA and Scheffe’s tests
with stimulus as the factor.
Results
Behavioral Performance
The mean (and SD) accuracy of the detection of oddball stimuli
was 97.8% (4.18%), indicating that subjects listened attentively
to the stimuli during MEG recording. The stimulus sound
pressure levels presented to the subjects at 40 dB above the
hearing threshold, averaged across 2 ears, were 78.1 (1.10), 72.1
(0.93), 69.9 (0.62), and 63.3 (0.64) dB for pure tones of 238, 365,
440, and 880 Hz, respectively, in the piano and vowel sessions.
The frequency dependence of these sound pressures qualitatively agrees with the equal-loudness contour (ISO226) of the
human auditory system, which has a minimum threshold at
about 1 kHz and increases as the frequency decreases to be
10 dB higher at 250 Hz. The stimulus sound pressures of piano
tone and vowel speech were 67.8 (0.85) and 66.3 (1.0) dB,
respectively. These values correspond to the levels of pure-tone
stimuli at 500--600 Hz, which are within the range of
main spectral peaks of the A3 tone (220--880 Hz) and vowel
(120--1200 Hz).
Activities of N1m and P2m Responses
Figure 2 shows examples of the measured fields for successive
piano tones and the calculated dipole moment in the right
hemisphere of one subject. The superposed responses of 41
channels in (a) show clear N1m and P2m peaks, which were
reduced in amplitude to nearly constant levels for the second
and third stimuli. The time course of dipole moment in (b) was
obtained by projecting the measured fields in (a) on the fields of
the grand mean of N1m dipoles. Stable moments for the second
and third stimuli suggest invariable location and direction of
current of the N1m source. In addition, appreciable negative
peaks were obtained for P2m, but the P2m moment was evaluated from the projection of the measured fields on the fields of
the mean P2m dipole. There was no significant difference in the
mean locations of N1m and P2m dipoles between piano tone
and vowel speech. They were located in the auditory cortex at
about Heschl’s gyrus in the superior temporal plane. The grand
mean dipoles across subjects averaged over all the stimulus
sounds indicated that the P2m source was anterior and medial
to the N1m source (Fig. 3). The anteriority of the P2m source to
the N1m was observed in the left hemisphere of 15 of the 18
subjects and in the right hemisphere of 11 of the 14 subjects, in
whom the dipoles were localized across control and main
sessions.
Figure 4 shows the grand mean amplitude of RMS fields across
subjects of the right hemisphere obtained in the piano and
vowel sessions. The reduction in the peak amplitude of the N1m
Cerebral Cortex November 2007, V 17 N 11 2727
Figure 3. Grand mean location of N1m and P2m dipoles averaged over all the
stimulus sounds. The dipoles were projected on coronal and axial magnetic resonance
images of one subject who had an average skull size of all subjects.
Figure 4. Grand mean amplitude of the RMS fields over 12 largest channels at the
maximum N1m peak in the right posterior temporal area. Results are shown for piano
and vowel sessions.
and P2m responses in the second and third stimuli was similar
for all the stimulus sounds of piano, vowel, and pure tones,
exhibiting second and third stimuli responses of nearly equal
amplitudes. It appeared that the piano tone (A3) evoked larger
responses than did the pure tones (440 and 880 Hz) in the piano
session. However, fluctuation of the latency of responses
influenced the grand mean amplitude of the response peak in
the waveform of RMS fields. For quantitative analysis, the peak
values of the dipole moments of N1m and P2m responses were
determined in the time course, as shown in Figure 2b, and
averaged across subjects. The results are summarized in Figure
5. In the piano session, main effects of stimulus on N1m were
found by ANOVA in the right hemisphere (F2,18 = 9.38, P <
0.002) and left hemisphere (F2,18 = 4.37, P < 0.003). Post hoc
2728 Auditory Responses to Complex Sounds
d
Kuriki et al.
Scheffe’s test revealed that the moments of piano tones were
significantly larger than the moments of pure tones of 440 Hz
(P < 0.002) and 880 Hz (P < 0.029) in the right hemisphere and
of 880 Hz (P < 0.031) in the left hemisphere. Similarly, main
effects of stimulus on P2m were found in the right hemisphere
(F2,18 = 10.9, P < 0.001) and left hemisphere (F2,18 = 15.5, P <
0.001), indicating that the moments of piano tones were larger
than the moments of pure tones of 880 Hz (P < 0.001) in the
right hemisphere and of 440 Hz (P < 0.002) and 880 Hz (P <
0.001) in the left hemisphere. In the vowel session, there were
no significant effects of stimulus on N1m in both hemispheres.
Effects of stimulus on P2m were also nonsignificant in most
cases, except in the left hemisphere (F2,18 = 7.65, P < 0.004) for
the moment of the vowel larger than that of 238-Hz tone (P <
0.004). Effects of stimulus order on N1m and P2m were evident
in both hemispheres in the piano and vowel sessions (F2,18 >
17.8, P < 0.001 for all conditions), the moment of the first stimulus being larger that the moments of the second and third
stimuli. However, there were no significant differences in the
moments of the second and third stimuli, suggesting that the
attenuation of both N1m and P2m responses by the repetition of
stimuli was saturated from the second stimulus.
Normalized Dipole Moments
We examined the attenuation of the dipole moment by stimulus
repetition in terms of the normalized dipole moment. The
moments of the second and third stimuli were normalized by
the moment of the first stimulus and averaged. Responses to the
piano tone, 440-Hz tone, vowel, and 365-Hz tone in single subjects for each were omitted from the analysis because of the
extremely small amplitudes of N1m or P2m. The grand mean
values of the obtained normalized moments are summarized in
Table 1. It is noted that the normalized moment is 1.0 when
there is no attenuation, being smaller for larger attenuation of
the moment by stimulus repletion. As indicated in Table 1,
significant effects of stimulus were found on P2m in the right
hemisphere for the piano (F2,16 = 11.9, P < 0.001) and vowel
(F2,16 = 5.13, P < 0.019) sessions, the normalized P2m moments
being larger for the piano tone (A3) than for the pure tones of
440 (P < 0.004) and 880 Hz (P < 0.001). Furthermore, the
normalized P2m moments were larger for the vowel /a/ than
the 238-Hz tone (P < 0.020). The difference between the vowel
and 365-Hz tone was not significant, but a paired t-test (2 tailed)
showed a marginal difference (P < 0.090) with the vowel being
larger than 365 Hz. No significant differences were found in the
normalized moments of P2m for the piano and vowel sessions in
the left hemisphere or the normalized moments of N1m in
either hemisphere or session.
Discussion
We used a dichotic listening scheme in stimulus presentation
and conducted separate MEG measurements to analyze the
right hemisphere and left hemisphere responses. In the responses of the analysis side, interference from the auditory
cortex of the other side may be minimal because of the
contralateral pathway dominance in dichotic listening (Kimura
1961, 1967). We made quantitative comparison of the responses
elicited by piano/vowel sounds and pure tones for N1m and
P2m in each hemisphere. The intrahemispheric analysis may not
be influenced by the existence of difference in the suppression
of the ipsilateral pathway of the right and left hemispheres
Figure 5. Grand mean dipole moments of N1m and P2m peaks for the first, second, and third stimuli obtained in piano and vowel sessions. Vertical bars indicate standard errors.
Table 1
Grand mean (SE) of the normalized moments of N1m and P2m dipoles obtained in the piano
and vowel sessions
Left hemisphere
Piano session
A3
440 Hz
880 Hz
Vowel session
/a/
238 Hz
365 Hz
Right hemisphere
N1m (SE)
P2m (SE)
N1m (SE)
P2m (SE)
0.46 (0.05)
0.52 (0.05)
0.63 (0.08)
0.63 (0.08)
0.63 (0.05)
0.53 (0.05)
0.57 (0.05)
0.52 (0.06)
0.53 (0.05)
0.85 (0.08)*,**
0.58 (0.06)**
0.56 (0.04)*
0.53 (0.06)
0.51 (0.05)
0.57 (0.06)
0.59 (0.11)
0.42 (0.05)
0.53 (0.10)
0.53 (0.03)
0.60 (0.04)
0.55 (0.07)
0.90 (0.13)***
0.48 (0.06)***
0.61 (0.10)
Note: SE, standard error. Statistical significance was found for the pairs designated by ‘‘*’’.
*P \ 0.001; **P \ 0.004; ***P \ 0.02.
(Fujiki et al. 2002). When the hemispheric effect was analyzed,
we examined the dependence of the difference in the dipole
moment or normalized moment of the responses on the
hemisphere side.
Persistent Responsiveness
The normalized moments of the P2m response were significantly larger for both piano and vowel sounds than for pure
tones with the intensity equated in hearing level, which was
found in the right hemisphere but not in the left hemisphere.
This result suggests that the adaptation by stimulus repetition of
the neural activity of P2m is reduced for spectrally complex
sounds, irrespective of musical or nonmusical timbres. The
right-sided laterality suggests acoustic processing of the relevant function of P2m activity. Furthermore, weak adaptation to
piano tones was observed in musically nonexperienced subjects, suggesting that persistent responsiveness of P2m activity
may be innately present in auditory cortical neurons but not
induced by training effects. It can be strengthened by practice,
however, because adaptation of the P2m response to piano
tones is more suppressed in musical subjects than in nonmusical
subjects (Kuriki et al. 2006). In clear contrast to P2m, there was
no significant difference in the normalized moments of the N1m
responses to piano/vowel sounds and pure tones in the right
and left hemispheres. These observations regarding the adaptation of N1m and P2m activities are in general agreement with
the results of recent MEG studies showing distinct sensitivities
of N1m and P2m responses to piano-timbre tones (Shahin et al.
2005; Kuriki et al. 2006).
Within the auditory cortex of primates and humans, the
caudomedial and rostrolateral portions in the parabelt area that
surrounds the primary auditory cortex contribute to the spatial
and object streams, respectively (Rauschecker and Tian 2000).
In this anatomical relationship, the dipole source of P2m is
estimated to be in the parabelt area at a site anterior to Heschl’s
gyrus of the primary auditory cortex (Godey et al. 2001).
Anterior location of the P2m source to that of the N1m has
also been confirmed (Hari et al. 1987; Papanicolaou et al. 1990;
Lütkenhöner and Steinstrater 1998; Kuriki et al. 2006). Though
not statistically significant, the P2m sources were located
anteriorly to the N1m sources in the right and left hemispheres
in this study. The N1m response is generated by 2 auditory
cortical sources of an early posterior and a late anterior subcomponent dominating the ascending and descending aspects
of the response, respectively (Sams et al. 1993; Loveless et al.
1996). The anterior subcomponent of N1m shows weak
adaptation to repetitive stimuli with decrease in the frequency
interval of successive stimuli, in contrast to the strong adaptation of the posterior N1m (Jääskeläinen et al. 2004). Thus, the
anterior N1m has been suggested to be involved in attentional
analysis of object features in the ‘‘what’’ processing pathway
(Jääskeläinen et al. 2004). The weakness of adaptation, its right
hemisphere dominance, and the anteriority of the dipole source
give strong support to the notion that P2m neurons are involved
in the ‘‘what’’ pathway of acoustic feature analysis. In an extension of the previous suggestion of frequency and/or temporal processing of complex sounds (Kuriki et al. 2006), we
speculate that the analyses of multiplicity of spectral components and regularity of the temporal waveform might be
specifically involved in the function of P2m activity.
Cerebral Cortex November 2007, V 17 N 11 2729
In an MEG study using successive vowels and envelopematched tones presented at a short interstimulus interval,
Teismann et al. (2004) showed that the relative amplitude of
N1m of the fourth-stimulus response to that of the first-stimulus
response is significantly larger in the left hemisphere than in the
right hemisphere for the vowel /a/ but not for a pure tone of
single frequency. They suggested that the persistent responsiveness of the N1m response to vowels represents the left
dominant auditory speech processing, but they did not describe
the P2m response. Because their measurements were conducted separately for the right and left hemispheres as in this
study, comparison of the results (Fig. 3 in Teismann et al. 2004)
between the vowel and pure tone within each hemisphere
would give more accurate evaluation than interhemispheric
comparison. Visual inspection of their data suggests that the
relative amplitude of N1m is lower for the vowel than the pure
tone in the right hemisphere. Here, the N1m response may
substantially include the contribution of the posterior subcomponent of N1m due to a long intertrain interval of stimulation of 4--5 s (Teismann et al. 2004). Their results therefore
suggest that the posterior N1m activity in the right hemisphere
is more adapted to vowel sounds than pure tones, which may
subserve the detection of ecologically irrelevant unexpected
sounds (May et al. 1999; Jääskeläinen et al. 2004). In contrast,
observations in the present study did not show significant
difference in the normalized moment of N1m between the
vowel and pure tones, either in the right or left hemisphere.
Such behaviors may be ascribed to the short intertrain interval
of 1 s in this study, which may have reduced the contribution of
the posterior subcomponent to the observed N1m response. In
addition to this, the attention to stimuli in the discrimination
task in this study, in contrast to nonattentive listening while
video watching (Teismann et al. 2004), may alter the stimulus
dependence of adaptation of the posterior component of N1m
activity. Modification by attention of the auditory-evoked responses at N1m latency have been widely reported (Rif et al.
1991; Woldorff et al. 1993; Petkov et al. 2004; Obleser et al.
2004).
Magnitude of Activities
The results of the vowel session showed no significant differences in the dipole moments of N1m or P2m between the vowel
/a/ and pure tones of 238/365 Hz. This tendency was observed
in the 2 hemispheres, except for the P2m moment of the vowel
being larger than that of the pure tone of 238 Hz in the left
hemisphere. Considering that the normalized moment of P2m
was larger for the vowel than pure tones in the right hemisphere, weak adaptation may be an important factor, compared
with the magnitude of activity, in the neural function of P2m.
Contrary to this, the dipole moments of N1m and P2m were
significantly larger for the piano tone A3 than pure tones of 440/
880 Hz in both hemispheres. This point will be discussed later.
As for the hemisphere effect, diverse results for vowel-evoked
MEG responses under various recording conditions have been
obtained. The amplitude of N1m is larger for vowels than piano
and pure tones in the left hemisphere with binaural stimulation
under the condition of a target detection task (Gootjes et al.
1999), and the N1m amplitude is larger for vowels synthesized
from a natural periodic wave than for simulated vowels made of
tonal and noise waves in both hemispheres with binaural stimulation in passive recording (Tiitinen et al. 2005). A laterality
2730 Auditory Responses to Complex Sounds
d
Kuriki et al.
index of N1m given by LI = (L – R)/(L + R) shows left hemisphere dominance of LI > 0 in strongly right-handed subjects
measured with binaural stimulation under a matching task,
where L and R refer to the ratio of the N1m strength to the
vowel and tone responses over the left and right hemispheres,
respectively (Kirveskari et al. 2006). This index is relevant to the
data in this study because the value of L and R can be obtained
from the N1m moments of the vowel and tone responses within
the hemisphere. The calculated results showed that the LI was
positive only in 3 of 9 right-handed subjects. The mean value of
LI across subjects was close to zero but not positive to be leftside dominant.
We infer that binaural interaction and attention effects may
have contributed to the diversity of the above results on the
hemisphere effect of N1m. Our observations suggest no clear
hemispheric difference in the magnitude or the ratio of N1m of
vowel and pure-tone responses under the conditions of the
dichotic listening that excludes binaural interaction and a sound
detection task that requires acoustic processing. In the sustained field that occurs after P2m latency, the dipole moment is
larger for vowels than pure tones, being more pronounced in
the left hemisphere than in the right hemisphere with monaural
stimulation under the condition of a target detection task
(Eulitz et al. 1995). We did not observe the component of
sustained field for vowel stimulus in this study. This may be the
result of short intervals (1.0--1.05 s) between the sequential
stimuli that lasted for a long period (1.9 s), whereas long
interstimulus intervals (1.8--2.3 s) compared with stimulus periods (0.49--0.6 s) were used in the previous MEG studies
reporting prominent sustained fields (Hari et al. 1989; Eulitz
et al. 1995).
Shahin et al. (2005) observed significantly larger amplitude of
P2m, but not N1m, for piano than pure tones in both musically
experienced and nonexperienced subjects, but hemispheric
difference was not examined. In their experiment, the sound
intensity was adjusted above the hearing threshold measured
for each stimulus, and the envelope of the stimulus sounds was
matched. In a recent study, Lütkenhöner et al. (2006) found
larger amplitude (and dipole moment) of N1m for piano tones
than for envelope-matched pure tones in both musicians and
nonmusicians in the left hemisphere. They adjusted the stimulus intensity above the pure-tone threshold. It seems that
adjustment of the stimulus intensity and fine matching of the
stimulus envelope have strong effects on the amplitude of
piano-tone responses.
In the present study, the piano tones elicited significantly
larger N1m and P2m moments than did pure tones in both
hemispheres. Visual inspection of the stimulus waveforms indicated a bump structure at the beginning of the piano tone,
which may correspond to ‘‘attack’’ of key (see the waveform of
A3 in Fig. 1). This bump structure was not simulated in the
envelope of the waveform of pure tones. Because the N1m
response is sensitive to the onset parameters of stimulus (Hari
et al. 1987; Hari 1990; Biermann and Heil 2000), the amplitude
of N1m and P2m responses may reflect the maximum slope in
the initial envelope, whereas the hearing threshold may be
given by the average amplitude during the length of the sound.
We computed the ratio of the maximum sound pressure in the
initial 100 ms to the average sound pressure of the whole envelope. The results showed that the maximum-to-average
pressure ratios were 2.0 for piano tones and 1.8 for 440 and
880 Hz in the piano session and 1.1 for /a/ and 1.2 for 238 and
368 Hz in the vowel session. Thus, the initial envelope of the
stimulus waveform may be responsible for the enhanced N1m
and P2m moments for piano tones.
Notes
Part of this work was supported by grant-in-aids (15300148, 16500166)
from the Ministry of Education, Science, and Culture of Japan. Conflict
of Interest: None declared.
References
Atienza M, Cantero JL, Dominguez-Marin E. 2002. The time course of
neural changes underlying auditory perceptual learning. Learn Mem.
9:138--150.
Belin P, Zilbovicius M, Crozier S, Thivard L, Fontaine A, Masure MC,
Samson Y. 1998. Lateralization of speech and auditory temporal
processing. J Cogn Neurosci. 10:536--540.
Biermann S, Heil P. 2000. Parallels between timing of onset responses of
single neurons in cat and of evoked magnetic fields in human
auditory cortex. J Neurophysiol. 84:2426--2439.
Budd TW, Barry RJ, Gordon E, Rennie C, Michie PT. 1998. Decrement of
the N1 auditory event-related potential with stimulus repetition:
habituation vs. refractoriness. Int J Psychophysiol. 31:51--68.
Edeline JM. 1999. Learning-induced physiological plasticity in the
thalamo-cortical sensory systems: a critical evaluation of receptive
field plasticity, map changes and their potential mechanisms. Prog
Neurobiol. 57:165--224.
Eulitz C, Diesch E, Pantev C, Hampson S, Elbert T. 1995. Magnetic and
electric brain activity evoked by the processing of tone and vowel
stimuli. J Neurosci. 15:2748--2755.
Fujiki N, Jousmaki V, Hari R. 2002. Neuromagnetic responses to
frequency-tagged sounds: a new method to follow inputs from
each ear to the human auditory cortex during binaural hearing. J
Neurosci. 22(RC205):1--4.
Godey B, Schwartz D, de Graaf JB, Chauvel P, Liegeois-Chauvel C. 2001.
Neuromagnetic source localization of auditory evoked fields and
intracerebral evoked potentials: a comparison of data in the same
patients. Clin Neurophysiol. 112:1850--1859.
Gootjes L, Raij T, Salmelin R, Hari R. 1999. Left-hemisphere dominance
for processing of vowels: a whole-scalp neuromagnetic study.
Neuroreport. 10:2987--2991.
Hämäläinen MS. 1995. Functional localization based on measurements
with a whole-head magnetometer system. Brain Topogr. 7:283--289.
Hari R. 1990. The neuromagnetic method in the study of the human
auditory cortex. In: Grandori F, Hoke M, Romani GL, editors.
Auditory evoked magnetic fields and electric potentials. Basel
(Switzerland): Karger. Adv Audiol. 6:222--282.
Hari R, Hamalainen M, Kaukoranta E, Makela J, Joutsiniemi SL, Tiihonen J.
1989. Selective listening modifies activity of the human auditory
cortex. Exp Brain Res. 74:463--470.
Hari R, Kaila K, Katila T, Tuomisto T, Varpula T. 1982. Interstimulus
interval dependence of the auditory vertex response and its
magnetic counterpart: implications for their neural generation.
Electroencephalogr Clin Neurophysiol. 54:561--569.
Hari R, Pelizzone M, Makela JP, Hallstrom J, Leinonen L, Lounasmaa OV.
1987. Neuromagnetic responses of the human auditory cortex to onand offsets of noise bursts. Audiology. 26:31--43.
Hertrich I, Mathiak K, Lutzenberger W, Ackermann H. 2004. Transient
and phase-locked evoked magnetic fields in response to periodic
acoustic signals. Neuroreport. 15:1687--1690.
Jääskeläinen IP, Ahveninen J, Bonmassar G, Dale AM, Ilmoniemi RJ,
Levanen S, Lin FH, May P, Melcher J, Stufflebeam S, et al. 2004. Human
posterior auditory cortex gates novel sounds to consciousness. Proc
Natl Acad Sci USA. 101:6809--6814.
Jäncke L, Wustenberg T, Scheich H, Heinze HJ. 2002. Phonetic
perception and the temporal cortex. Neuroimage. 15:733--746.
Kayser J, Bruder GE, Tenke CE, Stuart BK, Amador XF, Gorman JM 2001.
Event-related brain potentials (ERPs) in schizophrenia for tonal and
phonetic oddball tasks. Biol Psychiatry. 49:832--847.
Kimura D. 1961. Cerebral dominance and the perception of verbal
stimuli. Can J Physiol. 15:166--171.
Kimura D. 1967. Functional asymmetry of the brain in dichotic listening.
Cortex. 3:163--177.
Kirveskari E, Salmelin R, Hari R. 2006. Neuromagnetic responses to
vowels vs. tones reveal hemispheric lateralization. Clin Neurophysiol.
117: 643--648.
Kuriki S, Kanda S, Hirata Y. 2006. Effects of musical experience on
different components of MEG responses elicited by sequential pianotones and chords. J Neurosci. 26:4046--4053.
Loveless N, Levanen S, Jousmaki V, Sams M, Hari R. 1996. Temporal
integration in auditory sensory memory: neuromagnetic evidence.
Electroencephalogr Clin Neurophysiol. 100:220--228.
Lütkenhöner B, Steinstrater O. 1998. High-precision neuromagnetic
study of the functional organization of the human auditory cortex.
Audiol Neurootol. 3:191--213.
Lütkenhöner B, Seither-Preisler A, Seither S. 2006. Piano tones evoke
stronger magnetic fields than pure tones or noise, both in musicians
and non-musicians. Neuroimage. 30:927--937.
Mäkelä AM, Alku P, Tiitinen H. 2003. The auditory N1m reveals the lefthemispheric representation of vowel identity in humans. Neurosci
Lett. 353:111--114.
May P, Tiitinen H, Ilmoniemi RJ, Nyman G, Taylor JG, Naatanen R. 1999.
Frequency change detection in human auditory cortex. J Comput
Neurosci. 6:99--120.
Obleser J, Elbert T, Eulitz C. 2004. Attentional influences on functional
mapping of speech sounds in human auditory cortex. BMC Neurosci.
5:24.
Paetau R, Ahonen A, Salonen O, Sams M. 1995. Auditory evoked magnetic
fields to tones and pseudowords in healthy children and adults. J Clin
Neurophysiol. 12:177--185.
Papanicolaou AC, Baumann S, Rogers RL, Saydjari C, Amparo EG,
Eisenberg HM. 1990. Localization of auditory response sources using
magnetoencephalography and magnetic resonance imaging. Arch
Neurol. 47:33--37.
Penhune VB, Zatorre RJ, MacDonald JD, Evans AC. 1996. Interhemispheric anatomical differences in human primary auditory cortex:
probabilistic mapping and volume measurement from magnetic
resonance scans. Cereb Cortex. 6:661--672.
Petkov CI, Kang X, Alho K, Bertrand O, Yund EW, Woods DL. 2004.
Attentional modulation of human auditory cortex. Nat Neurosci.
7:658--663.
Poeppel D, Yellin E, Phillips C, Roberts TP, Rowley HA, Wexler K,
Marantz A. 1996. Task-induced asymmetry of the auditory evoked
M100 neuromagnetic field elicited by speech sounds. Brain Res Cogn
Brain Res. 4:231--242.
Ponton CW, Eggermont JJ, Kwong B, Don M. 2000. Maturation of human
central auditory system activity: evidence from multi-channel
evoked potentials. Clin Neurophysiol. 111:220--236.
Rauschecker JP, Tian B. 2000. Mechanisms and streams for processing of
‘‘what’’ and ‘‘where’’ in auditory cortex. Proc Natl Acad Sci USA. 97:
11800--11806.
Reinke KS, He Y, Wang C, Alain C. 2003. Perceptual learning modulates
sensory evoked response during vowel segregation. Brain Res Cogn
Brain Res. 17:781--791.
Rif J, Hari R, Hamalainen MS, Sams M. 1991. Auditory attention affects
two different areas in the human supratemporal cortex. Electroencephalogr Clin Neurophysiol. 79:464--472.
Ritter W, Vaughan HG Jr, Costa LD. 1968. Orienting and habituation
to auditory stimuli: a study of short term changes in average
evoked responses. Electroencephalogr Clin Neurophysiol. 25:
550--556.
Romanski LM, Tian B, Fritz J, Mishkin M, Goldman-Rakic PS, Rauschecker
JP. 1999. Dual streams of auditory afferents target multiple domains
in the primate prefrontal cortex. Nat Neurosci. 2:1131--1136.
Sams M, Hari R, Rif J, Knuutila J. 1993. The human auditory sensory
memory trance persists about 10 sec: neuromagnetic evidence.
J Cong Neurosci. 5:363--370.
Shahin A, Bosnyak DJ, Trainor LJ, Roberts LE. 2003. Enhancement of
neuroplastic P2 and N1c auditory evoked potentials in musicians.
J Neurosci. 23: 5545--5552.
Cerebral Cortex November 2007, V 17 N 11 2731
Shahin A, Roberts LE, Pantev C, Trainor LJ, Ross B. 2005. Modulation of
P2 auditory-evoked responses by the spectral complexity of musical
sounds. Neuroreport. 16:1781--1785.
Shestakova A, Brattico E, Soloviev A, Klucharev V, Huotilainen M. 2004.
Orderly cortical representation of vowel categories presented by
multiple exemplars. Brain Res Cogn Brain Res. 21:342--350.
Teismann IK, Soros P, Manemann E, Ross B, Pantev C, Knecht S. 2004.
Responsiveness to repeated speech stimuli persists in left but not
right auditory cortex. Neuroreport. 15:1267--1270.
Tesche CD, Uusitalo MA, Ilmoniemi RJ, Huotilainen M, Kajola M, Salonen
O. 1995. Signal-space projections of MEG data characterize both
distributed and well-localized neuronal sources. Electroencephalogr
Clin Neurophysiol. 95:189--200.
Tiitinen H, Alho K, Huotilainen M, Ilmoniemi RJ, Simola J, Naatanen R.
1993. Tonotopic auditory cortex and the magnetoencephalographic
(MEG) equivalent of the mismatch negativity. Psychophysiology.
30:537--540.
2732 Auditory Responses to Complex Sounds
d
Kuriki et al.
Tiitinen H, Makela AM, Makinen V, May PJ, Alku P. 2005. Disentangling
the effects of phonation and articulation: hemispheric asymmetries
in the auditory N1m response of the human brain. BMC Neurosci.
6:62.
Tremblay K, Kraus N, McGee T, Ponton C, Otis B. 2001. Central auditory
plasticity: changes in the N1-P2 complex after speech-sound
training. Ear Hear. 22:79--90.
Weinberger NM. 2004. Specific long-term memory traces in primary
auditory cortex. Nat Rev Neurosci. 5:279--290.
Woldorff MG, Gallen CC, Hampson SA, Hillyard SA, Pantev C, Sobel D,
Bloom FE. 1993. Modulation of early sensory processing in human
auditory cortex during auditory selective attention. Proc Natl Acad
Sci USA 90:8722--8726.
Zatorre RJ, Belin P. 2001. Spectral and temporal processing in human
auditory cortex. Cereb Cortex. 11:946--953.
Zatorre RJ, Belin P, Penhune VB. 2002. Structure and function of
auditory cortex: music and speech. Trends Cogn Sci. 6:37--46.