A mediating role of the premotor cortex in phoneme segmentation
Introduction
Given the non-linear mapping between phonetic segments and their acoustic realization, distinct theoretical perspectives have been proposed to explain speech perception. A crucial distinction among these perspectives is the use of purely auditory perceptual mechanisms, or that of articulatory control processes and more generally the involvement of the motor system (for reviews, see Diehl et al., 2004, Galantucci et al., 2006, Schwartz et al., 2002). Many researchers have advocated models of speech perception that focus only on the auditory system and on the acoustic properties of speech (for a review, see Diehl et al., 2004). It is hypothesized that the speech signal is highly structured in that it contains invariant acoustic patterns for phonetic features, and that speech sounds are perceived using the same mechanisms of audition and perceptual learning. From this view, speech perception is assumed to be driven by auditory mechanisms, based on invariant properties of the acoustic signal, and not mediated by the motor system. In contrast, in the motor theory of speech perception (Liberman et al., 1967, Liberman and Mattingly, 1985, Liberman and Whalen, 2000), the objects of speech perception are the speaker’s intended articulatory gestures rather than acoustic or auditory events. According to this view, speech gestures are represented in the brain as invariant motor control structures while their manifestation in the acoustic signal or in the articulatory movements may vary contextually. Since speaker and listener share a common repertoire of motor primitives, speech perception is thought as a set of perceptuo-motor processes, used to recover the articulated gestures as the objects of perception. As such, speech perception is a sensorimotor process in that perceiving speech is perceiving speech motor (articulatory) gestures.
Consistent with this view, recent brain imaging studies demonstrate that frontal brain areas involved in the planning and execution of speech gestures (i.e., the posterior part of the left inferior frontal gyrus, namely Broca’s area, and the ventral premotor cortex) are activated during passive auditory, visual and/or auditory-visual speech perception (e.g., Ojanen et al., 2005, Pekkola et al., 2006, Pulvermuller et al., 2006, Skipper et al., 2005, Skipper et al., 2007, Wilson and Iacoboni, 2006, Wilson et al., 2004). Single-pulse TMS studies also show that lip or tongue motor-evoked potentials (MEP) are enhanced during both passive speech listening and viewing, when stimulating the corresponding area of the left primary motor cortex (Fadiga et al., 2002, Roy et al., 2008, Sundara et al., 2001, Watkins and Paus, 2004, Watkins et al., 2003). This increased excitability of the motor system during speech perception is related to an increase in activity in Broca’s area and the ventral premotor cortex (vPMC), as shown by a recent study combining the technique of positron emission tomography with TMS (Watkins & Paus, 2004). Finally, this speech motor ‘resonance’ mechanism (Fadiga et al., 2002) appears to be articulator specific, motor facilitation being stronger when the recorded muscle activity and the auditory speech stimulus reflect the same articulator (Fadiga et al., 2002, Roy et al., 2008). This is also suggested by two recent functional magnetic resonance imaging (fMRI) studies showing similar somatotopic patterns of motor activity in the vPMC during both producing and listening to or viewing lips- and tongue-related phonemes (Pulvermuller et al., 2006, Skipper et al., 2007). Altogether, these studies support the idea that both Broca’s area and the vPMC are recruited during speech processing, and suggest that speech perception involves a specific mapping from the speaker’s articulatory gestures into the listener’s motor plans.
However, despite accumulating evidence that passive speech perception induces cortical activity in both Broca’s area, the vPMC and the orofacial primary motor cortex, whether speech motor centers actually contribute to speech perception remains debated. While previous brain imaging and single-pulse TMS studies demonstrate the recruitment of motor areas during passive speech perception, these results are intrinsically correlational and cannot be used to address causality. Two powerful techniques for establishing causality are through the use of rTMS and electrocortical stimulation during awake neurosurgical operations to directly manipulate brain functioning. Both techniques can be used to temporarily disrupt neural activity of a given cortical region, thereby creating a transient ‘virtual lesion’, and to explore causal relations between the stimulated region and specific motor, sensory and/or cognitive functions (for reviews, see Boatman, 2004, Devlin and Watkins, 2007, Pascual-Leone et al., 2000, Walsh and Cowey, 2000). However, both electrocortical mapping studies and rTMS studies are inconclusive regarding a possible functional role of Broca’s area and the left svPMC in auditory speech processing under normal listening conditions. Indeed, temporarily disrupting the activity of the left inferior frontal gyrus does not impair participants’ ability to perform auditory speech discrimination tasks (Boatman, 2004, Boatman and Miglioretti, 2005) but phonological judgments that likely involve verbal working memory and articulatory rehearsal (Boatman, 2004, Romero et al., 2006). On the other hand, a recent rTMS study showed that stimulating the superior portion of the left ventral premotor cortex (svPMC), a region found to be activated during both syllable perception and production in recent fMRI studies (Wilson et al., 2004, Pulvermuller et al., 2006), impaired auditory syllable identification (Meister, Wilson, Deblieck, Wu, & Iacoboni, 2007). The results were interpreted to suggest that the premotor cortex is an essential component in speech perception and that it may contribute to it through a top–down modulation of temporal cortex. However, it should be noted that the auditory syllable identification task was performed in the presence of masking noise which reduced performance overall and therefore impacts on the interpretation of the results. Because Meister et al. (2007) did not examine the effects of premotor stimulation on speech processing in the absence of masking noise, it is still unclear whether the motor system is functionally activated under normal speech processing conditions and, if not, whether motor system involvement is only functional in the presence of sensory challenge or is activated more generally when task demands (beyond increasing signal to noise) are increased.
To extend and refine the observations from the Meister et al. (2007) study, we examined the influence of left svPMC in speech perception under normal listening conditions and a number of auditory speech tasks. We used 1 Hz low-frequency rTMS and frameless stereotaxy to temporarily disrupt the activity of svPMC and examined participants’ performance on a phoneme identification task, a syllable discrimination task and a phoneme discrimination task. All these tasks involved the same set of nonsense syllables, presented without masking noise, but differed in the use of phonemic segmentation and verbal working memory demands. We hypothesized that the phoneme identification task, in which participants were asked to judge whether a syllable began with /p/ or /b/, could be performed solely based on an acoustic-auditory analysis of the stop consonant voicing at the onset of the syllable, without the need to recruit the motor system (Boatman, 2004, Burton et al., 2000). In the syllable identification task, because the presented syllable pairs differed (or not) only in their first initial phonemes (e.g., /put/–/but/), a similar acoustic–phonetic analysis is also likely required together with minimal verbal storage to discriminate phonetic contrasts and compare the two syllables. Finally, in the phoneme discrimination task, participants were asked to determine whether the initial phonemes of syllable pairs were the same or different. Because half of the syllable pairs differed in more complex ways from just their initial segment (e.g., /put/–/bon/), it is likely that the listener had to segment the initial phonemes from the remainder of the syllables and then compare them in order to make a same/different judgment (Burton and Small, 2006, Burton et al., 2000). We hypothesized that specific effects of rTMS over the left svPMC on accuracy and reaction times should be stronger in the phoneme discrimination task, the motor system being likely recruited to assist in phonological segmentation and working memory processes (Romero et al., 2006). According to the above-mentioned theories of speech perception, this would argue against the view that speech perception relies exclusively on the auditory system and the acoustic properties of speech but also would refine a possible mediating role of the premotor cortex in speech perception under normal listening conditions in segmenting the speech stream into constituent phonemes.
Section snippets
Participants
Ten healthy adults (seven females; mean age ± SD, 27 ± 5 years) participated in the study. All but one were native speakers of Canadian French. The other one was a native speaker of French. All participants were right-handed (Oldfield, 1971), had normal or corrected-to-normal vision and reported no history of speaking or hearing disorders. Participants were screened for neurological, psychiatric, and other medical conditions, and contraindications to TMS (Wassermann, 1998). Written informed consent
Perceptual scores
The main effect of task was significant (F(3, 27) = 4.34, p = .05 – see Fig. 2), with a lower percentage of correct responses in the phoneme discrimination task than in the other tasks (on average, 98% (±1) in the control task, 98% (±2) in the phoneme identification task, 98% (±2) in the syllable discrimination task and 89% (±4) in the phoneme discrimination task; all p’s < .03). Neither the stimulation mode (F(1, 9) = 0.56) nor the interaction between the two variables (F(3, 27) = 0.20) were significant.
Reaction times
Discussion
Compared to sham stimulation, low-frequency rTMS applied over the svPMC had an effect on the response latencies in the phoneme discrimination task, which resulted in slower RTs without affecting the accuracy of the response. No specific effect caused by identical stimulation over the same cortical region was observed in the phoneme identification and the syllable discrimination tasks, nor in the visual matching task. Before we discuss these results, it is important to consider some inherent
Acknowledgments
We thank Jean-Luc Schwartz and Jérome Aubin for their help. This study was supported by research grants from NSERC and CIHR (Canada) to V.L.G. and CNRS (France) to M.S., and by a Richard H. Tomlinson Research Fellowship and a travel grant from the ‘PPF: Interactions multimodales’ to M.S.
References (56)
- et al.
The evolutionary origin of the language areas in the human brain. A neuroanatomical perspective
Brain Research Reviews
(1997) - et al.
Cortical memory mechanisms and language origins
Brain and Language
(2006) Cortical bases of speech perception: Evidence from functional lesion studies
Cognition
(2004)- et al.
An fMRI investigation of syllable sequence production
Neuroimage
(2006) - et al.
Functional neuroanatomy of segmenting speech and nonspeech
Cortex
(2006) - et al.
Phonetic perceptual identification by native- and second-language speakers differentially activates brain regions involved with acoustic phonetic processing and those involved with articulatory–auditory/orosensory internal models
Neuroimage
(2004) - et al.
A comprehensive review of the effects of rTMS on motor cortical excitability and inhibition
Clinical Neurophysiology
(2006) - et al.
Imaging speech production using fMRI
Neuroimage
(2005) - et al.
Towards a functional neuroanatomy of speech perception
Trends in Cognitive Science
(2000) - et al.
Dorsal and ventral streams: A framework for understanding aspects of the functional anatomy of language
Cognition
(2004)
The motor theory of speech perception revised
Cognition
The essential role of premotor cortex in speech perception
Current Biology
Processing of audiovisual speech in Broca’s area
Neuroimage
Transcranial magnetic stimulation in cognitive neuroscience: Virtual lesion, chronometry, and functional connectivity
Current Opinion in Neurobiology
Perception of matching and conflicting audiovisual speech in dyslexic and fluent readers: An fMRI study at 3T
Neuroimage
Functional specialization for semantic and phonological processing in the left inferior prefrontal cortex
Neuroimage
Language within our grasp
Trends in Neurosciences
Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: Basic principles and procedures for routine clinical application. Report of an IFCN committee
Electroencephalography and Clinical Neurophysiology
Phonological and lexical motor facilitation during speech listening: A transcranial magnetic stimulation study
Journal of Physiology–Paris
Gestural drift in a bilingual speaker of Brazilian Portuguese and English
Journal of Phonetics
Listening to talking faces: Motor cortical activation during speech perception
Neuroimage
Clustered functional MRI of overt speech production
Neuroimage
Meta-analyzing left hemisphere language areas: Phonology, semantics, and sentence processing
Neuroimage
Risk and safety of repetitive transcranial magnetic stimulation
Report and suggested guidelines from the international workshop on the safety of repetitive transcranial magnetic stimulation, June 5–7, 1996
Electroencephalography and Clinical Neurophysiology
Seeing and hearing speech excites the motor system involved in speech production
Neuropsychologia
Neural responses to non-native phonemes varying in producibility: Evidence for the sensorimotor nature of speech perception
Neuroimage
Top–down and bottom–up processes in speech comprehension
Neuroimage
From monkey-like action recognition to human language: An evolutionary framework for neurolinguistics
Behavioral and Brain Sciences
Cited by (128)
Disruption of speech motor adaptation with repetitive transcranial magnetic stimulation of the articulatory representation in primary motor cortex
2021, CortexCitation Excerpt :This train of subthreshold pulses successfully reduces cortical excitability for a further 10–15 min (Möttönen & Watkins, 2009). Previous work has shown that the temporary interference induced by this form of stimulation over speech motor cortex can lead to increased response latencies and decreased accuracy during task performance, and suppressed electroencephalogram (EEG) responses to unattended stimuli (Meister et al., 2007; Möttönen et al., 2013; Sato et al., 2009; Tang et al., 2021). The hand area in left M1 was selected as a control site because: 1) it is close to the tongue area and, therefore, stimulation produces similar levels of noise and scalp sensations to those heard and felt when TMS pulses are applied over the tongue area; 2) it has the same cytoarchitecture (Brodmann area 4; Eichert et al., 2020).
Sensorimotor contributions to working memory differ between the discrimination of Same and Different syllable pairs
2021, NeuropsychologiaCitation Excerpt :Thus, the total length of stimuli was 2 seconds. While segmentation is not necessary for successful syllable discrimination, it should be noted that segmentation is known to modulate activity in anterior aspects of the dorsal stream (Burton et al., 2000; Locasto et al., 2004; Sato et al., 2009; Thornton et al., 2017). To avoid any potential confounds of segmentation effects, syllable pairs within an individual trial were allowed to differ only by the initial consonant.