Neural Basis of Sensorimotor Adaptation in the Speech ... · Figure 5. a) Topographical...

Introduction Results

Method

ReferencesReferencesReferences

Figure 5. a) Topographical distribution maps of correlation between absolute ERP amplitudes and the mag-nitudes of vocal motor adaptation responses to pitch-shifted auditory feedback in 100 ms time bins within a window from 100 ms before to 500 ms after the onset of vocalization. b) The trend line plots of the signi�cant correlation results in representative contacts before and after the onset of vocalization.

Background

Speech production is mediated by a series of motor movements developed for communicative intent. Vocal production is an im-portant subcomponent of speech that enables individuals to con-trol the pitch and loudness of their voice through controlling the movement of respiratory and laryngeal muscles. Recent models of the vocal system have emphasized the role of auditory and soma-tosensory feedback mechanisms in motor control of vocalization [e.g., 1, 2, 3]. Evidence from previous studies has supported the notion that motor behaviors can be shaped temporarily with cor-rective intent (adaptive behaviors) and new behaviors can be de-veloped with lasting e�ects (learned behaviors) following changes in sensory feedback stimuli [4, 5].

Objective

The present study investigated the underlying neural mechanisms of behavioral responses involved in sensorimotor adaptation to pitch-shifted auditory feedback.

Behavioral vocal responses

Vocal pitch output during baseline condition was relatively stable. Downward pitch shifts in auditory feedback were compensated for with a progressive change in voice pitch in the upward direction during adaptation that was maintained during washout.

On average, subjects compensated for 60.9% of pitch shifts in the auditory feedback during adaptation onset and 73.5% during adaptation o�set. Vocal pitch output re-mained at 57.2% above baseline level during washout.

Figure 1. Results of the behavioral vocal response analysis: a) trial-by-trial pro�le of the grand-average (n=13) vocal responses during baseline, adaptation onset, adaptation o�set and washout conditions. b) Bar plot representation of the statistical analysis for the di�erences between the mean of the grand-average vocal responses during baseline, ad-aptation onset, adaptation o�set and washout conditions (** p<0.01).

References[1] Guenther, F. H., Ghosh, S. S., & Tourville, J. a. (2006). Neural modeling and imaging of the cortical interactions underlying syllable production. Brain and Language, 96(3), 280–301. http://doi.org/10.1016/j.bandl.2005.06.001[2] Hickok, G., Houde, J., & Rong, F. (2011). Sensorimotor integration in speech processing: computational basis and neural organiza-tion. Neuron, 69(3), 407–22. http://doi.org/10.1016/j.neuron.2011.01.019[3] Houde, J. F., & Chang, E. F. (2015). The cortical computations underlying feedback control in vocal production. Current Opinion in Neurobiology, 33, 174–181. http://doi.org/10.1016/j.conb.2015.04.006[C4] Nasir, S. M., Darainy, M., & Ostry, D. J. (2013). Sensorimotor adaptation changes the neural coding of somatosensory stimuli. J Neurophysiol, 109, 2077–2085. http://doi.org/10.1152/jn.00719.2012[C5] Ostry, D. J., & Gribble, P. L. (2016). Sensory Plasticity in Human Motor Learning. Trends in Neurosciences, 39(2), 114–123. http://doi.org/10.1016/j.tins.2015.12.006[6] Houde, J. F. (1998). Sensorimotor Adaptation in Speech Production. Science, 279(5354), 1213–1216. http://doi.org/10.1126/science.279.5354.1213[7] Houde, J. F., & Jordan, M. I. (2002). Sensorimotor adaptation of speech I: Compensation and adaptation. Journal of Speech, Lan-guage, and Hearing Research: JSLHR, 45(2), 295–310. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/12003512[8] Jones, J. A., Munhall, K. G., & Kl, O. (2005). Remapping Auditory-Motor Representations in Voice Production. Current Biology, 15, 1768–1772. http://doi.org/10.1016/j.cub.2005.08.063[9] Keough, D., Hawco, C., & Jones, J. A. (2013). Auditory-motor adaptation to frequency-altered auditory feedback occurs when par-ticipants ignore feedback. BMC Neuroscience, 14(1), 1. http://doi.org/10.1186/1471-2202-14-25[10] Sengupta, R., & Nasir, S. M. (2015). Redistribution of neural phase coherence re�ects establishment of feedforward map in speech motor adaptation. J Neurophysiol, 113, 2471–2479. http://doi.org/10.1152/jn.00731.2014

Experimental task

12 healthy subjects with no reported history of voice or musical train-ing (2 males, age range: 22–27 years, mean age: 24.8 years) repeatedly produced steady vocalizations of the vowel sound /a/ while receiving voice auditory feedback across four vocalization phases:

1) Baseline: Voice auditory feedback not altered2) Adaptation (onset): Auditory feedback pitch shifted by -100 cents 3) Adaptation (o�set): Continuation of the previous adaptation phase 4) Washout: Auditory feedback returned to baseline (no alteration)

EEG recording

EEG signals were sampled at 1 kHz and recorded by 64 electrodes.

Behavioral vocal responseVocal responses to pitch shift stimuli were calculated by extracting and averaging the pitch frequency contours (in cents) across all trials.

Figure 3. a) Pro�le of the grand-average (n=12) ERP responses in CPz (centro-parietal) electrode from -200 ms before to 500 ms after the onset vocalization overlaid across baseline, adaptation onset, adaptation o�set and washout conditions. b and c) Bar plot representation of the post-hoc statistical analysis for the di�erences between grand-average ERP activities across all conditions in 100 ms time bins within a window from 100 ms before to 500 ms after the onset of vocalization. (* p<0.05)

Figure 2. Topographical distribution maps of the ERP activity during baseline, adaptation onset, adaptation o�set and washout conditions in 100 ms time bins within a window from 100 ms before to 500 ms after the onset of vocalization.

DiscussionWe propose that our �ndings support the following:

ERP responses

During baseline trials, a left-lateralized ERP component over the temporal area was elicited in a time window from -100 ms to 200 ms after vocalization onset. This activ-ity was diminished during adaptation and washout.

Pre-vocalization ERP activity was elicited from -100 to 0 ms prior to the onset of vocal production with centro-parietal distribution.

Post-vocalization ERP activity was elicited within 0-100 ms with parietal distribution and within 100-200 ms with central distribution after voice onset. ERP activities were of larger amplitude during adaptation onset.

Post-vocalization ERP activity was also elicited with fronto-central distribution from 200-500 ms with no signi�cant di�erence in amplitude among all conditions.

Speech Neuroscience LabSpeech Neuroscience LabSpeech Neuroscience Lab

Neural Basis of Sensorimotor Adaptation in the Speech Production System

Roozbeh Behroozmand and Stacey SangtianSpeech Neuroscence Lab, Department of Communication Sciences and Disorders, University of South Carolina

Behavioral vocal motor adaptation is indicative of sensorimotor remapping of feedforward motor commands in response to pitch-shifted auditory feedback.

Pitch Shift Stimulus (PSS)

AuditoryFeedback

Vowel Sound Vocalization (~2 sec)a)

/a/

0

-50

-100

Pitch Shift[cents]

Baseline Adaptation Washout

100 100 10010 10Trials

b)

100

onset

Baseline Adaptation Washout

OnsetPitch Shift: 0 cents Pitch Shift: 0 cents

Pitch Shift: -100 cents

Trial

Voca

l Res

pons

e [c

ents

]

0

20

40

60

80

100

Baselin

e

Adaptation

Washout

Voca

l Res

pons

e [c

ents

] ****

a) b)

Vocal Compensation (Adaptation onset)

Vocal Compensation

**

Adaptation

Ons

et

a)

b)

CPz

Vocalization Onset

-200 -100 0 100 200 300 400 500Time [ms]

Pre-Vocalization

−2

−1.5

−1

−0.5

0

0.5

ERP

Ampl

itude

[µV]

Adaptation onsetBaseline

Washout

BaselineAdaptation onset

Washout

*

Pre-vocalization: -100-0 ms

CPz−1

−0.8

−0.6

−0.4

−0.2

0

ERP

Ampl

itude

[µV]

−1.6

−1.4

−1.2

−1

−0.8

−0.6

−0.4

−0.2

0

* CPz

Post-vocalization: 0-100 ms

−1.2

−1

−0.8

−0.6

−0.4

−0.2

0

CPz

Post-vocalization: 100-200 ms

*

−1.2

−1

−0.8

−0.6

−0.4

−0.2

0

ERP

Ampl

itude

[µV]

Post-vocalization: 200-300 ms

CPz−1

−0.8

−0.6

−0.4

−0.2

0Post-vocalization: 300-400 ms

CPz −0.8

−0.7

−0.6

−0.5

−0.4

−0.3

−0.2

−0.1

0

CPz

Post-vocalization: 400-500 msc)

Baseline

AdaptationOnset

Adaptation

Washout

+5.0 µV

-3.0 µV

Time Window Relative to Vocalization Onset

-100-0 ms 0-100 ms 100-200 ms 200-300 ms 300-400 ms 400-500 ms

Pre-Vocalization Post-Vocalization

ERP Responses during Vocal Motor Adaptation

CPz

TP7

Cz

FCzTP8

-200 -100 0 100 200 300 400−2

−1.5

−1

−0.5

0

0.5

1

1.5

2TP7

500

Vocalization Onset

ERP

Ampl

itude

[µV]

Time [ms]

a)

−1−0.5

0

0. 5

1

1. 5

2

2. 5

3

-200 -100 0 100 200 300 400 500Time [ms]

TP8

Adaptation (onset)Baseline

Washout

Post-Vocalization

Post-Vocalization

ERP

Ampl

itude

[µV]

b)

−1.5

−1

−0.5

0

0. 5

1

1. 5

**

*

*

**

**

Post-Vocalization (0-100 ms)

ERP

Ampl

itude

[µV]

BaselineAdaptation onset

Washout

Right Hem

isphere (TP8)Left H

emisphere (TP7)

−1.5

−1

−0.5

0

0. 5

1

1. 5

**

**

**

*

Post-Vocalization (100-200 ms)

ERP

Ampl

itude

[µV]

Right Hem

isphere (TP8)Left H

emisphere (TP7)

*

*

Time Window Relative to Vocalization Onset

-100-0 ms 0-100 ms 100-200 ms 200-300 ms 300-400 ms 400-500 ms

Correlation between ERP and Vocal Motor Adaptation Magnitude

0.5

-0.5

r

a)Pre-vocalization Post-vocalization

−2 −1 0 1 2 3−100

−50

0

50

100

150

200

Absolute ERP Amplitude [µV](TP7: Left Temporo-Parietal)

Voca

l Ada

ptat

ion

[cen

ts]

r = -0.44, p<0.05

−2 −1 0 1−100

−50

0

50

100

150

200r = +0.49, p<0.05

Absolute ERP Amplitude [µV](CP2: Right Centro-Parietal)

−3 −2 −1 0 1−100

−50

0

50

100

150

200r = +0.41, p<0.05

Absolute ERP Amplitude [µV](C3: Left Central)

TP7CP2

C3 C3

Pre-vocalization (-100-0 ms) Pre-vocalization (-100-0 ms)

−3 −2 −1 0 1 2−100

−50

0

50

100

150

200r = +0.46, p<0.05

Absolute ERP Amplitude [µV](C3: Left Central)

Post-vocalization (100-200 ms) Post-vocalization (200-300 ms)b)

Figure 4. a) Pro�le of the grand-average (n=12) ERP responses in TP7 (left temporo-parietal) and TP8 (right temporo-parietal) electrodes from -200 ms before to 500 ms after the onset vocalization overlaid across baseline, adaptation onset, adaptation o�set and washout conditions. b) Bar plot representation of the post-hoc statistical analysis for the di�erences between grand-average ERP activities in the left and right hemispheres and across all conditions in time windows from 0-100 ms and 100-200 ms after the onset of vocalization. (* p<0.05)

Results (cont.)

[text]

Correlation analysisDistinct ERP response patterns that signi�cantl correlated with the magn-tiude of vocal motor adaptation pre- and post-vocalization.

Early stages of sensorimotor adaptation to pitch-shifted feedback in-volves increased contribution of the parietal cortex to incorporate audi-tory feedback for error detection and remapping of feedforward motor commands for error correction. However, as learning proceeds the error correction is internalized within the feedforward motor mechanisms of the frontal cortex and the contribution of the auditory feedback, and subsequently the interfacing parietal cortical mechanisms, are declined during vocal motor adaptation. Furthermore, we suggest that the proc-ess of error internalization is mediated by establishing an updated map of feedforward motor commands in the cortical motor areas of the left hemisphere.

Distinct neural substrates in the auditory (temporal), motor (central), and sensorimotor (parietal) cortical areas are involved in motor adaptation during vocal production under pitch-shifted auditory feedback.

This is consistent with previous studies on altered auditory feedback [e.g., 4 - 10]. Moreover, modulation of ERP activity supports that:

Figure 5. a) Topographical distribution maps of correlation between absolute ERP amplitudes and the mag-nitudes of vocal motor adaptation responses to pitch-shifted auditory feedback in 100 ms time bins within a window from 100 ms before to 500 ms after the onset of vocalization. b) The trend line plots of the signi�cant correlation results in representative contacts before and after the onset of vocalization.

Background

Speech production is mediated by a series of motor movements developed for communicative intent. Vocal production is an im-portant subcomponent of speech that enables individuals to con-trol the pitch and loudness of their voice through controlling the movement of respiratory and laryngeal muscles. Recent models of the vocal system have emphasized the role of auditory and soma-tosensory feedback mechanisms in motor control of vocalization [e.g., 1, 2, 3]. Evidence from previous studies has supported the notion that motor behaviors can be shaped temporarily with cor-rective intent (adaptive behaviors) and new behaviors can be de-veloped with lasting e�ects (learned behaviors) following changes in sensory feedback stimuli [4, 5].

Objective

The present study investigated the underlying neural mechanisms of behavioral responses involved in sensorimotor adaptation to pitch-shifted auditory feedback.

Behavioral vocal responses

Vocal pitch output during baseline condition was relatively stable. Downward pitch shifts in auditory feedback were compensated for with a progressive change in voice pitch in the upward direction during adaptation that was maintained during washout.

On average, subjects compensated for 60.9% of pitch shifts in the auditory feedback during adaptation onset and 73.5% during adaptation o�set. Vocal pitch output re-mained at 57.2% above baseline level during washout.

Figure 1. Results of the behavioral vocal response analysis: a) trial-by-trial pro�le of the grand-average (n=13) vocal responses during baseline, adaptation onset, adaptation o�set and washout conditions. b) Bar plot representation of the statistical analysis for the di�erences between the mean of the grand-average vocal responses during baseline, ad-aptation onset, adaptation o�set and washout conditions (** p<0.01).

References[1] Guenther, F. H., Ghosh, S. S., & Tourville, J. a. (2006). Neural modeling and imaging of the cortical interactions underlying syllable production. Brain and Language, 96(3), 280–301. http://doi.org/10.1016/j.bandl.2005.06.001[2] Hickok, G., Houde, J., & Rong, F. (2011). Sensorimotor integration in speech processing: computational basis and neural organiza-tion. Neuron, 69(3), 407–22. http://doi.org/10.1016/j.neuron.2011.01.019[3] Houde, J. F., & Chang, E. F. (2015). The cortical computations underlying feedback control in vocal production. Current Opinion in Neurobiology, 33, 174–181. http://doi.org/10.1016/j.conb.2015.04.006[C4] Nasir, S. M., Darainy, M., & Ostry, D. J. (2013). Sensorimotor adaptation changes the neural coding of somatosensory stimuli. J Neurophysiol, 109, 2077–2085. http://doi.org/10.1152/jn.00719.2012[C5] Ostry, D. J., & Gribble, P. L. (2016). Sensory Plasticity in Human Motor Learning. Trends in Neurosciences, 39(2), 114–123. http://doi.org/10.1016/j.tins.2015.12.006[6] Houde, J. F. (1998). Sensorimotor Adaptation in Speech Production. Science, 279(5354), 1213–1216. http://doi.org/10.1126/science.279.5354.1213[7] Houde, J. F., & Jordan, M. I. (2002). Sensorimotor adaptation of speech I: Compensation and adaptation. Journal of Speech, Lan-guage, and Hearing Research: JSLHR, 45(2), 295–310. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/12003512[8] Jones, J. A., Munhall, K. G., & Kl, O. (2005). Remapping Auditory-Motor Representations in Voice Production. Current Biology, 15, 1768–1772. http://doi.org/10.1016/j.cub.2005.08.063[9] Keough, D., Hawco, C., & Jones, J. A. (2013). Auditory-motor adaptation to frequency-altered auditory feedback occurs when par-ticipants ignore feedback. BMC Neuroscience, 14(1), 1. http://doi.org/10.1186/1471-2202-14-25[10] Sengupta, R., & Nasir, S. M. (2015). Redistribution of neural phase coherence re�ects establishment of feedforward map in speech motor adaptation. J Neurophysiol, 113, 2471–2479. http://doi.org/10.1152/jn.00731.2014

Experimental task

12 healthy subjects with no reported history of voice or musical train-ing (2 males, age range: 22–27 years, mean age: 24.8 years) repeatedly produced steady vocalizations of the vowel sound /a/ while receiving voice auditory feedback across four vocalization phases:

1) Baseline: Voice auditory feedback not altered2) Adaptation (onset): Auditory feedback pitch shifted by -100 cents 3) Adaptation (o�set): Continuation of the previous adaptation phase 4) Washout: Auditory feedback returned to baseline (no alteration)

EEG recording

EEG signals were sampled at 1 kHz and recorded by 64 electrodes.

Behavioral vocal responseVocal responses to pitch shift stimuli were calculated by extracting and averaging the pitch frequency contours (in cents) across all trials.

Figure 3. a) Pro�le of the grand-average (n=12) ERP responses in CPz (centro-parietal) electrode from -200 ms before to 500 ms after the onset vocalization overlaid across baseline, adaptation onset, adaptation o�set and washout conditions. b and c) Bar plot representation of the post-hoc statistical analysis for the di�erences between grand-average ERP activities across all conditions in 100 ms time bins within a window from 100 ms before to 500 ms after the onset of vocalization. (* p<0.05)

Figure 2. Topographical distribution maps of the ERP activity during baseline, adaptation onset, adaptation o�set and washout conditions in 100 ms time bins within a window from 100 ms before to 500 ms after the onset of vocalization.

DiscussionWe propose that our �ndings support the following:

ERP responses

During baseline trials, a left-lateralized ERP component over the temporal area was elicited in a time window from -100 ms to 200 ms after vocalization onset. This activ-ity was diminished during adaptation and washout.

Pre-vocalization ERP activity was elicited from -100 to 0 ms prior to the onset of vocal production with centro-parietal distribution.

Post-vocalization ERP activity was elicited within 0-100 ms with parietal distribution and within 100-200 ms with central distribution after voice onset. ERP activities were of larger amplitude during adaptation onset.

Post-vocalization ERP activity was also elicited with fronto-central distribution from 200-500 ms with no signi�cant di�erence in amplitude among all conditions.

Speech Neuroscience Lab

Neural Basis of Sensorimotor Adaptation in the Speech Production System

Roozbeh Behroozmand and Stacey SangtianSpeech Neuroscence Lab, Department of Communication Sciences and Disorders, University of South Carolina

Behavioral vocal motor adaptation is indicative of sensorimotor remapping of feedforward motor commands in response to pitch-shifted auditory feedback.

Pitch Shift Stimulus (PSS)

AuditoryFeedback

Vowel Sound Vocalization (~2 sec)a)

/a/

0

-50

-100

Pitch Shift[cents]

Baseline Adaptation Washout

100 100 10010 10Trials

b)

100

onset

Baseline Adaptation Washout

OnsetPitch Shift: 0 cents Pitch Shift: 0 cents

Pitch Shift: -100 cents

Trial

Voca

l Res

pons

e [c

ents

]

0

20

40

60

80

100

Baselin

e

Adaptation

Washout

Voca

l Res

pons

e [c

ents

] ****

a) b)

Vocal Compensation (Adaptation onset)

Vocal Compensation

**

Adaptation

Ons

et

a)

b)

CPz

Vocalization Onset

-200 -100 0 100 200 300 400 500Time [ms]

Pre-Vocalization

−2

−1.5

−1

−0.5

0

0.5

ERP

Ampl

itude

[µV]

Adaptation onsetBaseline

Washout

BaselineAdaptation onset

Washout

*

Pre-vocalization: -100-0 ms

CPz−1

−0.8

−0.6

−0.4

−0.2

0

ERP

Ampl

itude

[µV]

−1.6

−1.4

−1.2

−1

−0.8

−0.6

−0.4

−0.2

0

* CPz

Post-vocalization: 0-100 ms

−1.2

−1

−0.8

−0.6

−0.4

−0.2

0

CPz

Post-vocalization: 100-200 ms

*

−1.2

−1

−0.8

−0.6

−0.4

−0.2

0

ERP

Ampl

itude

[µV]

Post-vocalization: 200-300 ms

CPz−1

−0.8

−0.6

−0.4

−0.2

0Post-vocalization: 300-400 ms

CPz −0.8

−0.7

−0.6

−0.5

−0.4

−0.3

−0.2

−0.1

0

CPz

Post-vocalization: 400-500 msc)

Baseline

AdaptationOnset

Adaptation

Washout

+5.0 µV

-3.0 µV

Time Window Relative to Vocalization Onset

-100-0 ms 0-100 ms 100-200 ms 200-300 ms 300-400 ms 400-500 ms

Pre-Vocalization Post-Vocalization

ERP Responses during Vocal Motor Adaptation

CPz

TP7

Cz

FCzTP8

-200 -100 0 100 200 300 400−2

−1.5

−1

−0.5

0

0.5

1

1.5

2TP7

500

Vocalization Onset

ERP

Ampl

itude

[µV]

Time [ms]

a)

−1−0.5

0

0. 5

1

1. 5

2

2. 5

3

-200 -100 0 100 200 300 400 500Time [ms]

TP8

Adaptation (onset)Baseline

Washout

Post-Vocalization

Post-Vocalization

ERP

Ampl

itude

[µV]

b)

−1.5

−1

−0.5

0

0. 5

1

1. 5

**

*

*

**

**

Post-Vocalization (0-100 ms)

ERP

Ampl

itude

[µV]

BaselineAdaptation onset

Washout

Right Hem

isphere (TP8)Left H

emisphere (TP7)

−1.5

−1

−0.5

0

0. 5

1

1. 5

**

**

**

*

Post-Vocalization (100-200 ms)

ERP

Ampl

itude

[µV]

Right Hem

isphere (TP8)Left H

emisphere (TP7)

*

*

Time Window Relative to Vocalization Onset

-100-0 ms 0-100 ms 100-200 ms 200-300 ms 300-400 ms 400-500 ms

Correlation between ERP and Vocal Motor Adaptation Magnitude

0.5

-0.5

r

a)Pre-vocalization Post-vocalization

−2 −1 0 1 2 3−100

−50

0

50

100

150

200

Absolute ERP Amplitude [µV](TP7: Left Temporo-Parietal)

Voca

l Ada

ptat

ion

[cen

ts]

r = -0.44, p<0.05

−2 −1 0 1−100

−50

0

50

100

150

200r = +0.49, p<0.05

Absolute ERP Amplitude [µV](CP2: Right Centro-Parietal)

−3 −2 −1 0 1−100

−50

0

50

100

150

200r = +0.41, p<0.05

Absolute ERP Amplitude [µV](C3: Left Central)

TP7CP2

C3 C3

Pre-vocalization (-100-0 ms) Pre-vocalization (-100-0 ms)

−3 −2 −1 0 1 2−100

−50

0

50

100

150

200r = +0.46, p<0.05

Absolute ERP Amplitude [µV](C3: Left Central)

Post-vocalization (100-200 ms) Post-vocalization (200-300 ms)b)

Figure 4. a) Pro�le of the grand-average (n=12) ERP responses in TP7 (left temporo-parietal) and TP8 (right temporo-parietal) electrodes from -200 ms before to 500 ms after the onset vocalization overlaid across baseline, adaptation onset, adaptation o�set and washout conditions. b) Bar plot representation of the post-hoc statistical analysis for the di�erences between grand-average ERP activities in the left and right hemispheres and across all conditions in time windows from 0-100 ms and 100-200 ms after the onset of vocalization. (* p<0.05)

Results (cont.)

[text]

Correlation analysisDistinct ERP response patterns that signi�cantl correlated with the magn-tiude of vocal motor adaptation pre- and post-vocalization.

Early stages of sensorimotor adaptation to pitch-shifted feedback in-volves increased contribution of the parietal cortex to incorporate audi-tory feedback for error detection and remapping of feedforward motor commands for error correction. However, as learning proceeds the error correction is internalized within the feedforward motor mechanisms of the frontal cortex and the contribution of the auditory feedback, and subsequently the interfacing parietal cortical mechanisms, are declined during vocal motor adaptation. Furthermore, we suggest that the proc-ess of error internalization is mediated by establishing an updated map of feedforward motor commands in the cortical motor areas of the left hemisphere.

Distinct neural substrates in the auditory (temporal), motor (central), and sensorimotor (parietal) cortical areas are involved in motor adaptation during vocal production under pitch-shifted auditory feedback.

This is consistent with previous studies on altered auditory feedback [e.g., 4 - 10]. Moreover, modulation of ERP activity supports that: