fMRI-Studie des neuroradiologischen Instituts der Universität Frankfurt mit Probanden der Kasseler Stottertherapie

Beitrag auf der ASHA Convention 2002 in Atlanta: "Dysfunction of the Right Frontal Operculum in Stuttering"

ASHA The American Speech-Language Hearing Association
Meeting: 2002 American Speech-Language-Hearing Association Annual Convention

Author:

Katrin J. Neumann (Non-Member)
University of Frankfurt/M
Klinik fŸr Phoniatrie und Paedaudiologie
Johann Wolfgang Goethe-UniversitŠt Theodor - Stern
Frankfurt/M, HesseGermanyD 60590
Entered By, Primary Author, Presenting Author


Christine Preibisch (Non-Member)
Department of Neuroradiology, University of Frankfurt/M
Schleusenweg 2-16, 60528 Frankfurt, Germany
Frankfurt/M, HesseGermanyD 60528
Co-Author


Harald A. Euler (Non-Member)
Department of Psychology, University of Kassel
Hollaend. Str. 36-38, 34127 Kassel, Germany
Kassel, HesseGermanyD 34127
Co-Author


Alexander Wolff v. Gudenberg (Non-Member)
Institut der Kasseler Stottertherapie
Am Goldacker 4, 34225 Baunatal, Germany
Baunatal, HesseGermanyD 34225
Co-Author


Anne-Lise Giraud (Non-Member)
Department of Neurology and Neurophysiology, University of Frankfurt
Schleusenweg 2-16, 60528 Frankfurt
Frankfurt/M, HesseGermanyD 60528
Co-Author

Title: Dysfunction of the Right Frontal Operculum in Stuttering

Text:

Abstract of proposal

This study demonstrated with fMRI at 16 subjects who stutter when reading aloud that the right frontal operculum was recruited by every subject, and by none of 16 nonstuttering controls. With a passive visual semantic task, these findings were confirmed also in the absence of disfluency and overt motor output.

Narrative Summary

Contemporary theories of stuttering propose a central nervous dysfunction including dysfunction of motor control, altered cerebral dominance, deficiency of the language production system, sensory impairments, in particular auditory, or a complex combination of motor and linguistic deficits. Neuroimaging studies revealed that disfluent speech is generally associated with deactivations in left hemispheric language and auditory areas, and over-activations in right cortical and left cerebellar motor regions (1;2). To further investigate the cerebral dysfunction underlying stuttering, we performed a fMRI experiment with 16 fluent male speakers, and 16 male stuttering subjects. In a first task, subjects read aloud a text within the 3 sec of presentation or viewed passively letter-like meaningless signs. In a semantic decision task subjects performed a silent judgement of synonyms that did not require explicit motor output. Using an event related design allowed suppression of speech production motion artifacts. Data from all subjects were included in a random-effects analysis to obtain statistical inferences relative to differences between populations. Reading aloud was contrasted with passive viewing, and group comparisons of this difference were performed. Stuttering subjects activated more than controls in the right superior and middle frontal regions including the precentral gyrus, in the right superior temporal region, in both intraparietal sulci, and in the RFO (right frontal operculum). Apart from the parietal region, these results confirm earlier findings. Because subtle differences in the individual subjectsÕ performance might affect our results, we performed individual analyses as a fixed-effects group analysis where we retained only those regions that reached a significant activation threshold in each of the 16 stuttering subjects. This analysis ruled out activations that were not reproducible across subjects. Now we no longer found intraparietal activations. Superior/middle frontal activations were limited to the right precentral gyrus, but temporal activations were more extensive and bilateral. Recruitment of the RFO was confirmed at the exact location found in the random-effects analysis. This approach revealed cortical regions that, compared to controls, were significantly and systematically overactive when each of the stutterers read aloud even without disfluency. To assess the specificity of a regional activation in stuttering, we constrained the fixed-effects analysis to those brain regions that were not activated by any of the fluent speakers and found that the only region that showed this degree of specificity for stuttering was the RFO. Lesions to the RFO result in dysarthria, whereas its stimulation during open brain surgery induces blurred speech or speech arrest. The function of the right inferior frontal region, however, extends beyond motor aspects of speech production to various cognitive tasks. It is implicated in reading in dyslexic patients and is suspected to compensate for a dysfunction of the left frontal cortex. The RFO participates in working memory processes and in inhibition of response, when two interfering simultaneous tasks require a response. The most specific relationship to stuttering probably lies in the involvement of the RFO in a 'repair mode' for anomalies in speech and language, that is, when subjects have to notice and repair grammatical errors in auditorily presented sentences (3). The recruitment of this region fits well with the Ôcovert repair hypothesisÕ of stuttering, which assumes that stuttering people have difficulties selecting the correct phonemes, implying delays in the internal monitoring and covert repair. This repair attempt results in repetition of wrongly selected phonemes while adjusting the choice of the correct phonemes. We therefore propose the RFO as a likely candidate for mediating either the primary speech production inhibition in stuttering, or for compensating dysfunctions of distinct brain regions by inhibiting the manifestation of stuttering, which would be consistent with the fact that disfluent subjects dedicate a considerable amount of their cognitive resources to the prevention of stuttering. A possible compensation suggests that other brain structures should perform less well in stuttering subjects than in fluent speakers. Checking our data for such effects we could not identify a brain region activated consistently less in stuttering than in fluent speakers and providing evidence for a possible dysfunction the stuttering brain would have to compensate for. To test whether overactivity in the RFO in stutterers during reading was related specifically to overt language production we constrained the fixed-effects analysis to those regions that were also significantly more active in stutterers than in unaffected subjects in the synonym decision task. The RFO was the only region that was overactivated in stuttering subjects across tasks, suggesting that the dysfunction in stuttering occurs in processing steps upstream from speech output production. If RFO activation reflects a compensation mechanism, this mechanism may even operate in the absence of speech production demands, e. g. initiation or suppression of articulatory routines takes place even when there is no need for language output and thus possibly prior to critical early steps of speech production, e.g., activation of phonological output codes. Overactivation in the RFO in stutterers is compatible with an abnormal repair process that would interfere with normal speech production. Inner speech possibly occurred during the passive visual semantic task and during reading as a dual process. Yet, during the semantic task, we did not observe the general neural hallmarks of speech-production related brain activity that we would expect if RFO activation were to indicate a compensation for speech-production failure. Moreover, the RFO does not usually participate in inner speech. Our findings hence point to a type of neural activity in the RFO that does not occur during inner or overt speech in fluent speakers, and a dysfunction that is specific to the brain of people who stutter instead of being merely a phenomenon associated with the manifestation of stuttering.
(1) Fox, P.T., et al., A PET study of the neural systems of stuttering. Nature 382, 158-162 (1996).
(2) Braun, A.R., et al., Altered patterns of cerebral activity during speech and language production in developmental stuttering: An H215O positron emission tomography study. Brain 120,761-784 (1997).
(3) Meyer, M., Friederici, A.D., von-Cramon, D.Y., Neurocognition of auditory sentence comprehension: event related fMRI reveals sensitivity to syntactic violations and task demands. Brain Res. Cogn. Brain. Res. 9, 19-33 (2000).