Elsevier

NeuroImage

Volume 29, Issue 1, 1 January 2006, Pages 46-53
NeuroImage

The cerebral control of speech tempo: Opposite relationship between speaking rate and BOLD signal changes at striatal and cerebellar structures

https://doi.org/10.1016/j.neuroimage.2005.03.046Get rights and content

Abstract

So far, only sparse data on the cerebral organization of speech motor control are available. In order to further delineate the neural basis of articulatory functions, fMRI measurements were performed during self-paced syllable repetitions at six different frequencies (2–6 Hz). Bilateral hemodynamic main effects, calculated across all syllable rates considered, emerged within sensorimotor cortex, putamen, thalamus and cerebellum. At the level of the caudatum and the anterior insula, activation was found restricted to the left side. The computation of rate-to-response functions of the BOLD signal revealed a negative linear relationship between syllable frequency and response magnitude within the striatum whereas cortical areas and cerebellar hemispheres exhibited an opposite activation pattern. Dysarthric patients with basal ganglia disorders show unimpaired or even accelerated speaking rate whereas, in contrast, cerebellar dysfunctions give rise to slowed speech tempo which does not fall below a rate of about 3 Hz. The observed rate-to-response profiles of the BOLD signal thus might help to elucidate the pathophysiological mechanisms of dysarthric deficits in central motor disorders.

Introduction

Human speech production requires the temporal coordination of more than a hundred of muscles and therefore poses considerable demands on the underlying neural control mechanisms (Levelt, 1993). As compared to other motor domains such as locomotion or upper limb movements, only sparse data, predominantly derived from clinical investigations (Kent, 1997) or electrophysiological studies during brain surgery (Ojemann, 1994), on the cerebral correlates of vocal tract functions during verbal communication are available so far. Among others, the absence of a homologous animal model and the biomechanical complexities of orofacial and laryngeal structures account for the difficulties in exploring the neurophysiological bases of speech production (Barlow and Farley, 1989).

Disorders of the central motor system (upper motor neuron, cerebellum, basal ganglia) are characterized by distinct abnormalities in movement execution each, such as muscular paresis, ataxia or akinesia. These dysfunctions also may compromise articulatory and phonatory functions giving rise, e.g., to slurred and slowed speech utterances and altered voice quality (dysarthria/dysarthrophonia). Syllable repetitions as fast as possible (oral diadochokinesis tasks) proved to be one of the most sensitive clinical tests for neurogenic speech motor deficits (Kent et al., 1987). Most noteworthy, dysfunctions of the upper motor neurons, the cerebellum and the basal ganglia have a differential impact upon this task. Reduced maximum syllable repetition rates can be observed both in patients with spastic and ataxic dysarthria (Ackermann and Hertrich, 2000). While bilateral lesions of the corticobulbar tracts ultimately may give rise to anarthria and/or aphonia, performance of oral diadochokinesis does not fall below 3 Hz in ataxic disorders. By contrast, speech tempo is found largely unimpaired in Parkinson's disease (Ackermann and Ziegler, 1991), and subgroups of patients with basal ganglia disorders even may exhibit “speech hastening”, i.e., involuntary acceleration of speaking rate (Logigian et al., 1991, Hefter et al., 1993). Because of the limitations posed on kinematic and electromyographic recordings in dysarthric subjects, any pathophysiological interpretations of these speech motor deficits still faces considerable difficulties.

In addition to dysfunctions of the central motor system, damage to the frontal operculum of the left hemisphere may also give rise to articulatory deficits (Alexander et al., 1989). Since, however, nonspeech functions of laryngeal and orofacial muscles are unimpaired, these abnormalities have been assumed to reflect compromised “programming” or “planning” of speech motor functions (apraxia of speech; Ziegler, 2002). Besides lesions of the language-dominant inferior dorsolateral frontal lobe, more recent data indicate that this deficit also might be bound to disorders of the anterior insula (Dronkers, 1996).

As a rule, damage to the mesiofrontal cortex does not give rise to dysarthric deficits or apraxia of speech. However, electrical stimulation of SMA has been observed to elicit speech arrest or involuntary vocal emissions during surgical interventions in awake patients (Penfield and Roberts, 1959). In line with these electrophysiological data, SMA lesions may result in vocal epileptic seizures (Ackermann et al., 1996a), acquired dysfluencies (Ackermann et al., 1996b) or reduced spontaneous verbal communication (Krainik et al., 2003). Rather than subserving the generation of motor plans or the innervation of vocal tract muscles, SMA seems to represent a “starting mechanism of speech” that mediates motivational or attentional aspects of verbal behavior (Ackermann and Ziegler, 1995).

Imaging techniques such as positron emission tomograhy (PET) or functional magnetic resonance imaging (fMRI) now provide a further and more advanced tool for the investigation of the cerebral correlates of speech motor control. PET studies were able to demonstrate bilateral hemodynamic activation within sensorimotor cortex, cerebellum, and SMA as well as left-sided responses of the anterior insula and the left pallidum during repetition of single nouns (Petersen et al., 1988, Petersen et al., 1989, Bookheimer et al., 1998, Wise et al., 1999). Moreover, a previous fMRI investigation of our group found bilateral hemodynamic responses at the level of mesiofrontal and sensorimotor cortex, striatum, thalamus and cerebellum in association with a syllable repetition task (Riecker et al., 2005). Activation of dorsolateral premotor cortex and anterior insula was restricted to the left hemisphere. Most noteworthy, cortical, striatal and cerebellar areas showed, in line with clinical data (see above), a different relationship each between the BOLD signal and repetition rate. This preceding investigation had asked subjects to synchronize their syllable utterances with click trains applied via earphones. However, human verbal communication is characterized by internally generated rather than externally triggered speaking rates. A variety of clinical and functional imaging data indicate that these two modes of movement sequencing within the domain of upper limb movements are, at least partially, bound to different components of the frontal lobe (rostral part of SMA versus lateral premotor cortex; Halsband et al., 1993, Jahanshahi et al., 1995, Deiber et al., 1999a, Deiber et al., 1999b, Jenkins et al., 2000). In order to further delineate the neural basis of human speech motor control and the pathomechanisms of dysarthric deficits subsequent to cerebral lesions and diseases, brain activation was now measured during self-paced syllable repetitions at different production rates.

Section snippets

Subjects

The present study included eight healthy native German subjects (4 women, age: 21–30 years, mean age = 25.8 years). Informed consent had been obtained in line with the Declaration of Helsinki and the local Institutional Review Board. All participants were right-handed as determined by means of the Edinburgh Handedness Scale. Inclusion criterion was a lateralization index > 70%. None of the subjects had a history of neurologic, psychiatric or medical diseases nor any signs of hearing disorders.

Experimental procedure

Behavioral data

In spite of background scanner noise, the recorded syllable productions were clearly discernible at perceptual evaluation. Acoustic analyses of the obtained utterances revealed only slight variations of sound intensity across syllable trains. In addition, F0 measurements documented rather flat intonation contours with a maximum range of about one semitone. By contrast, pitch fluctuations during conversational speech production extend up to 12 semitones (one octave). The recorded vocalizations

The cerebral network of speech motor control

The cerebral correlates of self-paced syllable repetitions, in terms of the hemodynamic main effects calculated across all six frequencies considered, were found to encompass SMA, sensorimotor cortex, anterior insula, thalamus, caudatum, putamen and cerebellar hemispheres. Striatal (caudatum and putamen) and intrasylvian regions exhibited lateralized BOLD responses towards the left side whereas the other structures showed a rather bilateral activation pattern with a tendency to a left

Acknowledgments

This study was supported by the Deutsche Forschungsgemeinschaft (SFB 550 and B1 and DFG-SPP “Sprachproduktion”).

References (59)

  • D. Poeppel et al.

    Processing of vowels in supratemporal auditory cortex

    Neurosci. Lett.

    (1997)
  • A. Riecker et al.

    Parametric analysis of rate-dependent hemodynamic response functions of cortical and subcortical brain structures during auditorily cued finger tapping: a fMRI study

    Neuroimage

    (2003)
  • J.J. Sidtis et al.

    Predicting performance from functional imaging data: methods matter

    Neuroimage

    (2003)
  • G.W. Thickbroom et al.

    Dual representation of the hand in the cerebellum: activation with voluntary and passive finger movement

    Neuroimage

    (2003)
  • R.A. Weeks et al.

    Comparison of auditory, somatosensory, and visually instructed and internally generated finger movements: a PET study

    Neuroimage

    (2001)
  • R.J. Wise et al.

    Brain regions involved in articulation

    Lancet

    (1999)
  • H. Ackermann et al.

    Articulatory deficits in parkinsonian dysarthria: an acoustic analysis

    J. Neurol. Neurosurg. Psychiatry

    (1991)
  • H. Ackermann et al.

    Akinetic mutism—A review of the literature

    Fortschr. Neurol. Psychiatr.

    (1995)
  • H. Ackermann et al.

    Impaired procedural learning after damage to the left supplementary motor area (SMA)

    J. Neurol., Neurosurg. Psychiatry

    (1996)
  • H. Ackermann et al.

    Acquired dysfluencies following infarction of the left mesiofrontal cortex

    Aphasiology

    (1996)
  • H. Ackermann et al.

    Rate-dependent activation of a prefrontal-insular-cerebellar network during passive listening to trains of click stimuli: an fMRI study

    NeuroReport

    (2001)
  • S.M. Barlow et al.

    Neurophysiology of speech

  • R.M. Birn et al.

    Magnetic field changes in the human brain due to swallowing or speaking

    Magn. Reson. Med.

    (1998)
  • R.M. Birn et al.

    Event-related fMRI of tasks involving brief motion

    Hum. Brain Mapp.

    (1999)
  • S.Y. Bookheimer et al.

    Regional cerebral blood flow during auditory responsive naming: evidence for cross-modality neural activation

    NeuroReport

    (1998)
  • M.P. Deiber et al.

    Mesial motor areas in self-initiated versus externally triggered movements examined with fMRI: effect of movement type and rate

    J. Neurophysiol.

    (1999)
  • M.P. Deiber et al.

    Mesial motor areas in self-initiated versus externally triggered movements examined with fMRI: effect of movement type and rate

    J. Neurophysiol.

    (1999)
  • N.F. Dronkers

    A new brain region for coordinating speech articulation

    Nature

    (1996)
  • R.S. Frackowiak et al.

    Human Brain Function

    (1997)
  • Cited by (0)

    1

    Present address: Department of Neurology, University of Jena, Germany.

    View full text