TY - JOUR
T1 - On the context-dependent nature of the contribution of the ventral premotor cortex to speech perception
AU - Tremblay, Pascale
AU - Small, Steven L.
N1 - Funding Information:
We thank Margaret Flynn, Blythe Buchholz, Anthony S. Dick and Michael Andric for their help collecting the data, and Matthew Schiel and Priya Santhanam for their help pre-processing the fMRI data. Thanks also to all participants. This study was supported by the National Institutes of Health under NIDCD grants R33 DC008638 and R01 DC003378 to S.L. Small, and by a postdoctoral fellowship from the Canadian Institute for Health research (CIHR) to P. Tremblay. Their support is gratefully acknowledged.
PY - 2011/8/15
Y1 - 2011/8/15
N2 - What is the nature of the interface between speech perception and production, where auditory and motor representations converge? One set of explanations suggests that during perception, the motor circuits involved in producing a perceived action are in some way enacting the action without actually causing movement (covert simulation) or sending along the motor information to be used to predict its sensory consequences (i.e., efference copy). Other accounts either reject entirely the involvement of motor representations in perception, or explain their role as being more supportive than integral, and not employing the identical circuits used in production. Using fMRI, we investigated whether there are brain regions that are conjointly active for both speech perception and production, and whether these regions are sensitive to articulatory (syllabic) complexity during both processes, which is predicted by a covert simulation account. A group of healthy young adults (1) observed a female speaker produce a set of familiar words (perception), and (2) observed and then repeated the words (production). There were two types of words, varying in articulatory complexity, as measured by the presence or absence of consonant clusters. The simple words contained no consonant cluster (e.g. " palace"), while the complex words contained one to three consonant clusters (e.g. "planet"). Results indicate that the left ventral premotor cortex (PMv) was significantly active during speech perception and speech production but that activation in this region was scaled to articulatory complexity only during speech production, revealing an incompletely specified efferent motor signal during speech perception. The right planum temporal (PT) was also active during speech perception and speech production, and activation in this region was scaled to articulatory complexity during both production and perception. These findings are discussed in the context of current theories of speech perception, with particular attention to accounts that include an explanatory role for mirror neurons.
AB - What is the nature of the interface between speech perception and production, where auditory and motor representations converge? One set of explanations suggests that during perception, the motor circuits involved in producing a perceived action are in some way enacting the action without actually causing movement (covert simulation) or sending along the motor information to be used to predict its sensory consequences (i.e., efference copy). Other accounts either reject entirely the involvement of motor representations in perception, or explain their role as being more supportive than integral, and not employing the identical circuits used in production. Using fMRI, we investigated whether there are brain regions that are conjointly active for both speech perception and production, and whether these regions are sensitive to articulatory (syllabic) complexity during both processes, which is predicted by a covert simulation account. A group of healthy young adults (1) observed a female speaker produce a set of familiar words (perception), and (2) observed and then repeated the words (production). There were two types of words, varying in articulatory complexity, as measured by the presence or absence of consonant clusters. The simple words contained no consonant cluster (e.g. " palace"), while the complex words contained one to three consonant clusters (e.g. "planet"). Results indicate that the left ventral premotor cortex (PMv) was significantly active during speech perception and speech production but that activation in this region was scaled to articulatory complexity only during speech production, revealing an incompletely specified efferent motor signal during speech perception. The right planum temporal (PT) was also active during speech perception and speech production, and activation in this region was scaled to articulatory complexity during both production and perception. These findings are discussed in the context of current theories of speech perception, with particular attention to accounts that include an explanatory role for mirror neurons.
UR - http://www.scopus.com/inward/record.url?scp=79960206343&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79960206343&partnerID=8YFLogxK
U2 - 10.1016/j.neuroimage.2011.05.067
DO - 10.1016/j.neuroimage.2011.05.067
M3 - Article
C2 - 21664275
AN - SCOPUS:79960206343
SN - 1053-8119
VL - 57
SP - 1561
EP - 1571
JO - NeuroImage
JF - NeuroImage
IS - 4
ER -