A Neurological Basis for Stuttering and the Symptoms of Stuttering



Gregory J Snyder & Monica Strauss Hough


East Carolina University

Greenville, North Carolina


Inadequacies of the Behavioral Perspective regarding Evoked Fluency via Speech Feedback


Past stuttering research perspectives have attempted to account for the evoked fluency phenomenon via speech feedback as behavioral disruptions in the speech output of the person who stutters (Black, 1951; Lee, 1951; Brubaker, 1952; Fairbanks & Guttman, 1958; Coblenz & Agnello, 1965; Ham & Steer, 1967; Wingate, 1970).  Specific behavioral accounts for the fluency enhancing effects of speech feedback include distraction (Sheehan, 1970), slowed speech rate (Starkweather, 1978), external source of rhythm via slowed speech rate (Starkweather, 1983), prolonged vowel or sonorant production (Wingate, 1976); and by increasing the role of proprioception in speech production (Van Riper, 1982). 


However, most explanations of the speech feedback phenomenon via the behavioral paradigm are potentially confounded.  Explanations employing the distraction hypothesis offer no predictive value and subsequently hold little scientific explanatory significance (Stuart, 1999).  Any explanation touting exogenous rhythmic cues must account for the effective yet barely perceptual delay of 50 milliseconds (Lotzman, 1961; Kalinowski et al., 1995), or the zero-delay characteristics of frequency altered feedback (Howell et al., 1987).  Perspectives stressing the Lombard effect or slowed articulatory rate appear to be invalidated by data by later research finding that speech feedback remains effective regardless of speech or articulatory rate (Stuart et al., 1996).


The inclusion of the visual modality associated with visual choral speech (Kalinowski et al., 2000) negates most, if not all, behavioral explanations of the speech feedback phenomenon.  If the effortless production of immediate and stable natural sounding fluency is the long standing holy grail of stuttering treatment, then our theoretical understanding of the speech feedback phenomenon is necessary for the advancement of stuttering research and management.


Neurological Explanations of Developmental Stuttering


When stuttering is perceived as a behavioral disorder, accounting for the speech feedback phenomenon is difficult.  However, researchers have often indicated that the stuttering disorder may go beyond the behavioral aspects of stuttered speech (Conture, 1991; Jancke et al., 1997; Van Lieshout 1996a, 1996b; Kalinowski et al., 2000).  Such a notion appears to be validated with research suggesting that developmental stuttering disorder has a neurological etiology (Braun et al. 1995; Fox et al., 1996, 2000;  Wu et al., 1995, 1997; Salmelin et al., 2000) rather than behavioral.  From this perspective, developmental stuttering becomes a problem of deviant neural functioning, with its primary symptoms being behavioral in nature.  The clarification of neurological etiology creating behavioral symptomotology allows for entirely new explanations of the evoked fluency via speech feedback phenomenon.


Stuttering and Motoric Initiation


Developmental stuttering, a disorder long-since associated with verbal apraxia (Kent, 2000), has been consistently correlated with deviant pre-motor and supplementary motor area (SMA) functioning (Braun et al.., 1995; Fox et al., 1996, 2000; Wu et al., 1995, 1997).  The SMA is frequently associated with speech planning and cognitive initiation—including the initiation of speech gestures (Ikedeta et al., 1995; Mushiake et al., 1990; Roland, 1993; Tanji et al., 1994).  This proposed faulty SMA functioning and possible speech initiation dysfunction may be paralleled in recent magnetoencephalographical data, which revealed deviant neural activation patterns in those who stutter (Salmelin et al., 2000).  While normal speakers activated left prefrontal areas associated with speech planning prior to the activation of motor and dorsal premotor areas associated with speech-motor execution, stuttering cohorts consistently activated the motor cortices prior to the speech planning cortices (Salmelin et al., 2000).  One interpretation of this finding implies that observable stuttering behaviors represent the execution of incomplete or malformed speech plans.


Stuttering and Linguistic Formulation


While stuttering is not found during the onset of speech production (Karniol 1992, 1995; Yairi, 1983), the disorder appears to co-develop with the onset of complex integration of linguistic components into speech production (Brown, 1973; Karniol, 1992, 1995; Slobin, 1970).  From this perspective, it becomes difficult to solely identify stuttering as a behavioral or motoric disorder.  While the symptoms associated with developmental stuttering appear to be motoric in nature (Bloodstein, 1995; Starkweather, 1987), they also appear to be influenced by sentence formulation (Karniol, 1995), word position (Bloodstein, 1995), syntactic complexity (Karniol, 1995), and are documented to co-occur with general language deficiencies (Karniol, 1995).  Karniol (1995) provides a thorough review of the linguistic aspects of developmental stuttering and suggests that the linguistic elements associated with stuttering may indicate faulty processing or integration of suprasegmental sentence planning in speech production. 


While stuttering has been identified as a left hemispheric disorder (Andrews et al., 1972; Luessenhop et al., 1973), there is also evidence of left hemisphere non-affective or grammatical prosodic processing (Sproat, 1995).  The formulation and integration of such suprasegmental devices with the motor aspects of speech programming appears to be necessary for the production of fluent speech (Musalem, 1990).  The deficient integration of non-affective suprasegmental information within the speech motor program may serve as a tenable hypothesis for the deviant left prefrontal activations associated with stuttering (citation).  This account suggests that the malformation or faulty integration of non-affective suprasegmental information inhibits the motor or gestural programming and production of fluent speech.

Fluent and Stuttered Speech Production


Evidence suggests that choral speech, perhaps the most effective form of speech feedback (Bloodstein, 1995; Starkweather, 1987), alters neural functioning (Fox et al., 1996, 2000; Wu et al., 1995, 1997) specifically that of the supplementary motor area (Fox et al., 1996, 2000; Wu et al., 1995, 1997), auditory cortices (Fox et al., 1996, 2000; Salmelin et al., 1997, 2000), and the cerebellum (De Nil et al., 2001).  While never formally tested, variants of speech feedback (delayed auditory feedback, frequency altered feedback, visual choral speech, delayed visual feedback) all appear to approximate choral speech via differing modalities and temporal environments.  This discussion makes the fundamental assumption that the implementation of speech feedback variants (auditory feedback, visual feedback) is similar to that of choral speech, and therefore evokes similar neural alterations to the stuttered neural state. 


Theories of the Production of Speech and Stuttering


Liberman et al., (1967, 1985) suggest the basic unit of speech production is the speech gesture, representing a uniquely arranged laryngeal and vocal tract configuration perceived by listeners via the speaker’s speech signal.  However, it is unclear whether the Motor Theory of Speech Perception and Production (Liberman et al., 1967, 1985) included various linguistic or suprasegmental elements into its largely motoric theoretical perspective.  Another perspective of speech production that may account for the phylogenic development of speech production, thereby inferring the inclusion of linguistic and suprasegmental devices, is the Frame/Content Theory of Speech production (MacNeilage, 1998).  While potentially oversimplifying the Frame/Content Theory, the speech frame could be associated not only with the behavioral production of speech frame cycles, but also with the inclusion of linguistic content as an integral part of speech production.


With the inclusion of the current theories of speech production into the stuttering paradigm, a theory describing stuttering as a failure to fluently initiate speech becomes inadequate.  Instead, stuttering should be described as a failure to initiate the most fundamental unit of speech production, such as the speech gesture (Liberman et al., 1967, 1985), or the speech frame (MacNeilage, 1998).  Therefore, the failure on behalf of the premotor system to fluently initiate the speech frame or gesture is hypothesized to be correlated with faulty processing of left hemispheric suprasegmental integration with the speech premotor system. 


Common Elements in Speech Feedback


The observable symptoms of developmental stuttering are behavioral, and appear to be initiatory in nature (Webster, 1998).  This perspective may be supported with the nature of stuttering behaviors themselves, as they appear more frequently in the initial syllable or gesture of each word or phrase, as well as the nature of speech production itself (Starkweather, 1987).  Furthermore, the nature of speech feedback is such that the symptom altering effects appear only after the introduction of the second speech signal to the speaker.  Thus, with endogenous sources of speech feedback, the primary speech signal must be initially produced in order to be fed back to the speaker, thereby altering the subsequent gestural productions.


The inclusion of the visual modality within the speech feedback phenomenon necessitates changes in the previously mentioned explanations of evoked fluency via speech feedback.  The amalgamation of the neural stuttering data suggesting that developmental stuttering’s primary symptoms may include faulty pre-motor programming of speech resulting in a failure to fluently initiate speech suggests that speech feedback may somehow alter or avoid the hypothesized behavioral consequences.  The perception of exogenously initiated speech gestures, regardless of their sensory modality, may serve as a form of gestural priming.  If the disorder of stuttering is initiatory in nature, then the perception of an exogenously initiated speech gesture or opening speech frame may prime fluent gestural production in those who stutter.  This gestural priming may activate endogenous motor initiation of a speech gesture or frame via an alternate motor neural network (Goldberg 1985, 1992; Goldberg & Bloom, 1990).


Gestural Priming via Delayed Visual Feedback


Eight adults who stutter (4 males, 4 females, mean age = 30.14, SD = 10.21) participated in this research.  Participants reported either normal or corrected vision, and had no other diagnosed speech, language, hearing, or attentional disorders.  While all participants had a history of speech therapy, only one was currently enrolled.  Participants exhibiting at least a three percent dysfluency rate in the nonvisual feedback (NVF) control speaking condition were included in this study.  This inclusion criterion has been routinely used in other stuttering research [8].  Only the perspective participants meeting this criterion were included in this study; one adult failed to meet this standard and was subsequently excluded from the data set. 


For each speaking condition, participants were asked to read passages from junior high school science textbooks, all of which have been used in previous research [24].  Each passage, consisting of approximately 300 syllables, was divided into phrases consisting of 10 to 15 words, and printed on large double-sided cue cards.  Participants sat at a table (approximately 75 cm in height), and were asked to silently read and memorize a phrase of comfortable length (ranging from 3 to 7 words).  Participants would then look up from the cue card and speak the phrase they had just silently read and memorized.  Practice trials were allowed in each speaking condition until participants reported feeling comfortable with each experimental task.  Participants were asked to speak at a normal rate, and not to use any speaking techniques that could alter, control, or reduce stuttering.  Both the speaking conditions and passages were balanced using a Latin Square. 


As mentioned, the NVF speaking condition served as the control condition for this study.  A vertically positioned mirror (24 cm high by 33 cm wide, placed no more than 46 cm from the face of participant) was used to create synchronous self-generated visual feedback (SVF) for the second speaking condition.  During this speaking condition, participants were instructed to "look at your reflection and focus on the movement from your lips, mouth, tongue, and jaw to initiate speech" prior to speaking the memorized phrase.  A third speaking condition tested fluency enhancement via asynchronous self-generated visual feedback (AVF).  A 3Com HomeConnect Universal Serial Bus netcam (model # 0776) was connected to a Winbook XL2 laptop computer, running Microsoft's Windows 98 (version 4.10.1998) with 128 megabytes of random access memory, and a 300-megahertz Pentium II processor.  The laptop was equipped with a 14-inch active matrix liquid crystal display set at 800 by 600 pixels resolution.  The laptop was placed on the table no more than 46 cm from the participant, with the netcam placed approximately 61 cm from the participant, where it was mounted slightly above eye-level.  A perceived asynchronous visual delay was achieved primarily by setting the (3Com) netcam to the highest picture quality factory default setting.  Although the (visual) time delay may vary due to variables such as the intensity of ambient light and superfluous movement of each participant, it was estimated to be approximately 0.5 seconds.  In order to quantify the nature of this time delay more precisely, a digital video camera (Sony #DSR-PD100) recorded a strobe flash (RS #42-3048) which was captured by the 3Com netcam and then displayed on the Winbook XL2 computer.  This video was then digitized using Broadway Pro (version 5.10.9) into an .avi file (at 75 megabytes-per-minute), and then analyzed with Ulead’s Video Editor (version 6.0).  This technique allowed identification of the video frame of the original strobe flash, as well as the image of the strobe flash displayed on the laptop computer monitor; the sampling rate was 30 frames per second.  Five strobe flash tokens were averaged to generate an average visual time delay of 0.36 seconds (SD = .054 seconds).  Participants in the AVF speaking condition followed a nearly identical protocol for the other experimental conditions, as they were instructed to look at their image on the laptop screen, pause for the asynchronous visual feedback to catch up with their body movements so each participant had direct visual access to their mouth on the laptop screen, and finally to "look into your image on the screen and focus on the movement of your lips, mouth, tongue, and jaw to initiate speech" prior to speaking the memorized phrase.  During the experimental speaking conditions, participants perceived visual speech feedback; no exogenous auditory feedback was introduced during any speaking condition.  All participants were video recorded using a Hi-8mm video camera (Sony #CCD-TRV75), and a lapel microphone (RS #33-3003) that was attached no more than 15 cm from their mouth with an approximate orientation of 00 azimuth and –1800 altitude.  Results can be found in Figure 1.




click here to see figure one



Stuttered syllables were counted from the first 300 syllables of each speaking condition.  Stuttering was operationally defined as whole- and part-word repetitions, prolongations, or inaudible postural fixations (Kalinowski et al., 1996; Stuart et al., 1996).  The mean values of stuttering frequency as a function of visual feedback condition are presented in Figure 1.  Specifically, the mean stuttering frequency was 37.13 stuttered syllables (SE = 10.55) for the NVF speaking condition, 14.88 stuttered syllables (SE = 6.41) for the SVF speaking condition, and 7.38 stuttered syllables (SE = 3.33) for the AVF speaking condition. 


As evident in Figure 1, approximately 60% and 80% reductions of stuttered syllables occurred in the SVF and AVF speaking conditions respectively.  A one factor repeated measure analysis of variance (ANOVA) was conducted to investigate mean differences in stuttered syllables as a function of visual feedback.  The results of this analysis revealed a significant main effect of visual speech feedback [F(2,14) = 11.972, Greenhouse-Geisser p = .005, h 2 = .631].  Bonferroni post hoc pairwise comparisons conducted on the significant main effect revealed that the NVF condition was significantly different from both the SVF and AVF speaking conditions (p = .025); the frequency of stuttered syllables between the SVF and AVF speaking conditions were not significantly different from each other (p = .300).


Gestural Priming via Manual Gestural Productions


Eight people who stutter (6 males, 2 females, mean age = 30.75 years, SD = 9.3) participated in this research.  Participants reported either normal or corrected vision, right-hand dominance, and no other diagnosed speech, language, hearing, or attentional disorders.  Study participants were asked to use their right hands to control hand gesturing in all experimental speaking conditions.  An inclusion criterion of minimally three percent dysfluency rate in the control speaking condition was utilized in this study.  Two perspective participants failed to meet this criterion, and consequently were not included in the dataset.  The experimental materials and protocol of this study are nearly identical to the previously mentioned study.


During certain experimental speaking conditions of this study, each participant was given a hand puppet and asked to produce hand gestures preceding, and thereby initiating, speech production.  Participants produced gestural priming (GP) by manually mimicking speech gestures with the hand puppet immediately before speech initiation.  Specifically, participants were instructed to initiate manual hand puppet gestures (performed with the participants’ right hand) such that the puppet was gesturing each syllable immediately before speech production.  Manual hand puppet gesturing continued concurrently with the participants’ speech until the phrase was completed.  Special instructions were given to initiate manual gestures slightly before vocal initiation of the memorized text.  A no GP (NGP) speaking condition served as the control for this study.  The second speaking condition consisted of participants manually producing hand-puppet gestures outside of their own visual field (beneath the large stimulus cue cards), creating the self-generated gestural priming outside their visual field (SGPO), and subsequently providing no visual priming or feedback.  During this speaking condition, participants also were instructed to "manually initiate puppet gestures" prior to speaking the memorized phrase.  A third speaking condition tested fluency enhancement of a manually self-generated gestural primes inside the participants’ visual field (SGPI), subsequently providing visual priming and feedback.  For this speaking condition, study participants were instructed to “manually initiate puppet gestures first” and to “visually focus on the movement of the puppet’s lips, mouth, and jaw” prior to speaking the memorized phrase.  For the fourth speaking condition, the experimenter provided manual hand puppet gestures that served as visual priming or feedback for the study participants (EGPI).  In this speaking condition, study participants were instructed to wait for the experimenter to begin manually producing puppet gestures, and then to “visually focus on the movement of the puppet’s lips, mouth, and jaw” prior to speaking the memorized phrase.



click here to see figure two



The mean values of stuttering frequency were 47.88 stuttered syllables (median = 45, SE = 13.30) for the NGP speaking condition, 23.75 stuttered syllables (median = 8, SE = 10.64) for the SGPO speaking condition, 12 stuttered syllables (median = 4, SE = 7.75) for the SGPI speaking condition, and 19.88 stuttered syllables (median = 4.5, SE = 9.22) for the EGPI speaking condition. 


As evident in Figure 2, there was an approximate 50% reduction of stuttered syllables with SGPO, a 75% reduction of stuttered syllables with SGPI, and a 58% reduction of stuttered syllables with EGPI.  A one factor repeated measures analysis of variance (ANOVA) was conducted to investigate the mean differences in stuttered syllables as a function of gestural priming.  The result of this analysis revealed a significant main effect of gestural priming prior to speech production [F(3,21) = 12.929, Greenhouse-Geisser P = .001, ?2 = .649].  Fisher's least significant difference post hoc comparisons conducted on the significant main effect revealed that the NGP condition was significantly different from the SGPO, SGPI, and EGPI speaking conditions (P = .007, P = .004, and P = .005 respectively); the frequency of stuttered syllables between the SGPO and SGPI were also found to be significantly different from each other (P = .036).

Theories of Speech Production and Stuttering


Gestural Priming Induces Fluency


The research conducted suggests that the visual perception of speech or manual gestures enhances fluent speech in those who stutter.  Furthermore, while the perception or production of manual gestures was found to enhance fluency, the simultaneous perception of self-produced manual gestures appears to enhance fluency more effectively than either perception or production alone.  The nature of the subsequent enhanced fluency is similar to other findings within the auditory speech feedback phenomenon (Kalinowski et al., 1993, 1995), as the perception of an exogenously initiated speech gesture, regardless of sensory modality, appears to induce fluency by the hypothesized neural alterations associated with the stuttering data.


While not definitive, it appears as if the perception of exogenously initiated speech gestures serves to prime the preparation, planning, or programming of the speech gesture, subsequently resulting in enhanced fluent speech in those who stutter.  The perception of these exogenously initiated gestures appears to remain effective, regardless of its perceptual sensory modality.


Gestural Priming Appears to Work on a Gesture-by-Gesture Basis


The immediate and transient nature of the speech feedback phenomenon appears to remain constant regardless of the modality of speech feedback.  As noted in Snyder and Hough (in press), gestural priming appears to be effective only when it is perceived or produced immediately before gestural production.  This finding, consistent in auditory feedback, visual feedback and hand gesturing, suggests that fluency enhancement via gestural priming appears to provide efficacy on a gesture by gesture basis.  This finding appears congruent with the dramatically immediate yet transient nature of the speech feedback phenomenon. 


Gestural Priming found in Speech Feedback may alter Neural Functioning


The previously described initiatory gestural nature of the speech feedback phenomenon suggests that gestural priming may induce fluency primarily by altering neural functioning associated with speech motor production.  It should be noted that this is not a new perspective, as the concepts of gestural priming and initiation parallels a suggested relationship between the supplemental motor area (SMA) and stuttering.  Research frequently includes the SMA as a neural component of stuttering [Braun et al., 1995; Fox et al., 1996, 2000; Kroll & De Nil, 1998; Wu et al., 1995, 1997]; the SMA frequently is associated with speech planning and cognitive initiation, including the initiation of speech gestures [Ikeda et al., 1995; Mushiake et al., 1990; Roland, 1993; Tanji et al., 1994].  It has been suggested that the SMA plays an integral role in the planning of self-initiated movements, but not externally initiated or guided movements [Halsband et al., 1994; Passingham et al., 1989].  When electrically stimulated, the SMA has been found to impede the initiation of speech [Ojemann, 1994].  If a dysfunctional SMA is a neurological component of stuttered speech, then inefficient, deficient, or inversed sequential processing from the speech planning neural circuit may contribute to why people who stutter often fail to properly initiate their own speech.  This notion also may account for why an exogenous speech signal, containing gestural priming, is a necessary component for the initiation and maintenance of natural sounding fluent speech from people who stutter.  Thus, one may hypothesize that gestural priming, which inherently contains cues for speech coding, induces fluency by circumventing self-initiated articulatory planning, thereby stimulating activity within the left inferior frontal cortex prior to execution of the motor plan.


As neuroimaging data shows alterations and normalization of previous stuttered neural functioning when a person who stutters is presented with speech feedback (Fox et al., 1996, 2000; Wu et al., 1995, 1997), it seems reasonable to suggest that gestural priming, regardless of its modality, would yield similar results.  One specific alteration may evoke the previously described abnormal sequential activation between the left inferior frontal area (associated with articulatory coding) and the dorsal and premotor cortices (associated with speech execution) (Salmelin et al., 2000).  It is hypothesized that altered neural functioning associated with the SMA may enhance fluency by subsequently impacting the neural activation patterns between the left frontal area and the motor cortices, thereby potentially restoring the proper sequential activation pattern.


The Dual Premotor Hypothesis, Gestural Priming, and the Evoked Fluency via Speech Feedback in those who Stutter


If the speech feedback phenomenon effectively initiates individual speech gestures via exogenous gestural priming, a possible explanation for this phenomenon may be found in Goldberg and Bloom’s Dual Premotor System Hypothesis (1990), which delineates between a medial and lateral premotor system.  A theoretical explanation of the neurological etiology of stuttering lies in a dysfunction of the medial premotor system, which is believed to be responsible for initiating self-generated (speech) motor plans (Goldberg & Bloom, 1990; MacNeilage, 1999).  It is proposed that gestural primes evoke fluency by bypassing the medial premotor system dysfunction via a lateral premotor system.  Externally perceived gestural primes enter into the speech planning circuit via the lateral premotor system (Goldberg, 1985, 1992; Goldberg & Bloom, 1990), which is rich in afferent multisensory connections (Pandya, 1987).  Dysfunction within the functioning of the medial premotor system may yield speech plans that are either incomplete or non-existent (Salmelin et al., 2000).  The behavioral manifestations of stuttering (repetitions, prolongations, postural fixations) are produced as the body's natural compensatory correction response (Kalinowski et al., 2000) to the execution of an incomplete or errant speech motor plan.  However, an intact lateral premotor system may be capable of initiating a speech gesture with exogenous gestural priming, thus providing transient gestural fluency enhancement when the medially initiated speech plan cannot. 


While gestural priming may elicit fluent gestural production in those who stutter, the underlying non-affective suprasegmental dysfunction is hypothesized to remain.  Just as the behavioral aspects of stuttered speech are considered as symptomatic or compensatory to stuttering on a neural level (Kalinowski et al., 2000), so is the deviant speech motor processing associated with the production of stuttered speech gestures (Snyder & Hough, in press).  It is hypothesized that faulty (linguistic) supra-segmental formulation negatively affects the speech plan prior to any motor involvement, resulting in the deviant systemic motor functioning and activation patterns which are associated with the symptoms of the stuttering disorder (i.e., stuttering behaviors).



Andrews, G., Quinn, P.T., & Sorby, W.A.  (1972).  Stuttering: An investigation into cerebral dominance for speech.  Journal of Neurology, Neurosurgery & Psychiatry, 35, 414-418.


Black, J.W., The effect of delayed sidetone upon vocal rate and intensity, Journal of Speech and Hearing disorders, 16 (1951), 56-60.


Bloodstein, O.  (1995). A handbook on stuttering (5th ed.).  Chicago: The National Easter Seal Society.


Braun, A.R., Varga, M., Stager, S., Schulz, G., Selbie, S., Maisog, J.M., Carson, R.E., Ludlow, C.L.  (1997).  Altered patterns of cerebral activity during speech and language production in developmental stuttering: An H215O positron emission tomography study.  Brain, 120, 761-784.


Brown, R.  (1973).  A first language: The early stages.  Cambridge, MA: Harvard University Press.


Brubaker, R., An experimental investigation of speech disturbance as a function of the intensity of delayed auditory feedback.  Ph.D. dissertation, University of Illinois, Urbana, 1952.


Coblenz, H., & Agnello, J.G., The effects of delayed auditory feedback on the manner of consonant production.  Convention address, American Speech and Hearing Association, 1965.


Conture, E.G.  (1991).  Young stutterers’ speech production: A critical review.  In H.F.M. Peters, W. Hulstijn, & C.W. Starkweathers (Eds.), Speech motor control and stuttering.  (pp. 365-384).  New York: Elsevier.


De Nil, L.F., Kroll, R.M., & Houle, S.  (2001).  Functional neuroimaging of cerebellar activation during single word reading and verb generation in stuttering and nonstuttering adults.  Neuroscience Letters,  302, 77-80.


Fairbanks, B., & Guttman, N., Effects of delayed auditory feedback upon articulation, Journal of Speech and Hearing Research, 1 (1958), 12-22.


Fox, P.T., Ingham, R.J., Ingham, J.C., Hirsch, T.B., Downs, J.H., Martin, C., Jerabek, P., Glass, T.G., & Lancaster, J.L.  (1996).  A PET study of the neural systems of stuttering.  Nature, 382, 158-162.


Fox, P.T., Ingham, R.J., Ingham, J.C., Zamarripa, F., Xiong, J.H., Lancaster, J.L.  (2000).  Brain correlates of stuttering and syllable production.  Brain, 123, 1985-2004.


Goldberg, G.  (1985).  Supplementary motor area structure and function: Review and hypothesis.  Behavioral and Brain Sciences, 8, 567-616.


Goldberg, G.  (1992).  Premotor systems, attention to action and behavioral choice.  In: Neurobiology of motor program selection, ed. J. Lein, C.R. McCrohan, & W. Winlow.  Pergamon Press.


Goldberg, G., Bloom, K.  (1990).  The Alien Hand Sign: Localization, Lateralization, and Recovery.  American Journal of Physical Medicine & Rehabilitation, 69, 228-238.


Halsband U, Matsuzaka Y, and Tanji J.  Neuronal activity in the primate supplementary, presupplementary and premotor cortex during externally and internally instructed sequential movements. Neurosci Res, 1994;20: 149-55.


Ham, R., & Steer, M.D., Certain effects of alterations in auditory feedback, Folia Phoniatricia, 19 (1967), 53-62.


Howell, P., El-Yaniv, N., & Powell, D.J.  (1987). Factors affecting fluency in stutterers.  In H.F.M. Peters & W. Hulstijin (Eds.), Speech motor dynamics in stuttering (pp. 361-369).  New York: Springer-Verlag.


Ikeda A. Luders HO. Shibasaki H. Collura TF. Burgess RC. Morris HH 3rd, and Hamano T. Movement-related potentials associated with bilateral simultaneous and unilateral movements recorded from human supplementary motor area.  Electroencephalogra Clin Neurophysiol, 1995;95: 323-34.


Jancke, L., Bauer, A., Kaiser, P., & Kalveram, K.T.  (1997).  Timing and stiffness in speech motor control of stuttering and nonstuttering adults.  Journal of Fluency Disorders, 22, 309-321.


Kalinowski, J., Armson, J., Roland-Mieszkowski, M., & Stuart, A.  (1993).  Effects of alterations in auditory feedback and speech rate on stuttering frequency.  Language and Speech, 36, 1-16.


Kalinowski, J., Armson, J., & Stuart, A.  (1995). Effect of normal and fast articulatory rates on stuttering frequency.  Journal of Fluency Disorders, 20, 293-302.


Kalinowski J, Dayalu V, Stuart A, Rastatter M.  Stutter-free and stutter-filled speech signals and their role in stuttering amelioration.  Neurosci Let, 2000;293: 115-118.


Karniol, R.  (1992).  Stuttering out of bilingualism.  First Language, 12, 255-283.


Karniol, R.  (1995).  Stuttering, Language, and Cognition: A review and a model of stuttering as suprasegmental sentence plan alignment (SPA).  Psychological Bulletin, 117, 104-124.


Kent, R.D.  (2000).  Research on speech motor control and its disorders: a review and prospective.  Journal of Communication Disorders, 33, 391-428.


Kroll RM, and De Nil LF. Positron Emission Tomography Studies of Stuttering: Their relationship to our theoretical understandings and clinical understandings of the disorder. J Speech Lang Path Aud, 1998;22: 261-70.


Lee, B.S., Artificial Stuttering, Journal of speech and Hearing Disorders, 16 (1951), 53-55.


Liberman, A.M., Cooper, F.S., Shankweiler, D.P., Studdert-Kennedy, M.  (1967).  Perception of the Speech Code.  Psychological Review, 74, 431-461.


Liberman, A.M., & Mattingly, I.G.  (1985).  The motor theory of speech perception revised.  Cognition, 21, 1-36.


Luessehop, A.J., Boggs, J.S., Lorowit, L.J., & Walle, E.L.  (1973).  Cerebral dominance in stutterers determined by WADA testing.  Neurology, 23, 1190-1192.


Lotzman, V., Zur anwendung variierter verzogerungszeiten bei balbuties, Folia Phoniatrica, 13 (1961), 276-312.


MacNeilage, P.  (1998).  The frame/content theory of evolution of speech production.  Behavioral and Brain Sciences, 21, 499-546.


Mesulam, M.M.  (1990).  Large-scale neurocognitive networks and distributed processing for attention, language, and memory.  Annuls of Neurology, 28, 597-613.


Mushiake H, Inase M, and Tanji J.  Selective coding of motor sequence in the supplementary motor area of the monkey cerebral cortex.  Exp Brain Res, 1990;82: 208-10.


Ojemann GA. Cortical stimulation and recording in language.  In: Kertesz A, editor.  Localization and Neuroimaging in neuropsychology. New York: Academic Press, 1994, 33-55.


Pandya, D.  (1987).  Association Cortex.  In: The Encyclopedia of Neuroscience, ed. G. Adelman, Birkhauser.


Passingham RE, Chen YC, and Thaler D. Supplementary motor cortex and self-initiated movement.  In: Ito M, editor.  Neural Programming (Taniguchi Symposia on Brain Sciences ), 12th ed.  S. Karger: Basal, 1989, 13-24.


Roland PE. Brain Activation.  New York: Wiley-Liss, 1993.


Salmelin, R., Schnitzler, A., Schmitz, F., Freund, H.J.  (2000).  Single word reading in developmental stutterers and fluent speakers.  Brain, 123, 1184-1202.


Salmelin, R., Schnitzler, A., Schmitz, F., Jancke, L., Witte, O.W., & Freund, H.J.  (1998).  Functional organization of the auditory cortex is different in stutterers and fluent speakers.  Neuroreport, 9, 2225-2229.


Sheehan, J., Stuttering: Research and Therapy.  New York: Harper & Row, 1970.


Slobin, D.I.  (1970).  Universals in grammatical development in children.  In G.B. Glores d’Arcais & W.J.M. Levelt (ed.), Advances in psycholinguistics (pp. 174-186).  Amsterdam: North Holland.


Sproat R.  (1995).  Computational interpretations of neurolinguistic observations.  In Gazzaniga MS, (Ed.).  The Cognitive Neurosciences (pp. 931-942). Cambridge MA: MIT Press.


Starkweather, C.W., Recent research on stuttering and vocalization: clinical and theoretical implications.  Short course, American Speech and Hearing Association, 1978.


Starkweather, C.W., Tracking a changing tempo.  Convention Address, American Speech and Hearing Association, 1983.

Starkweather, C.W.  (1987).  Fluency and Stuttering.  New Jersey: Prentice-Hall, Inc.


Stuart, A. (1999).  The distraction hypothesis and the practice of pseudoscience: A reply to Bloodstein (1998).  Journal of Speech, Language, and Hearing Research, 42, 913-914.


Stuart, A., & Kalinowski, J.  (1996).  Fluent speech, fast articulatory rate, and delayed auditory feedback: Creating a crisis for a scientific revolution?  Perceptual and Motor Skills, 82, 211-218.


Tanji J, and Shima K, Role for supplementary motor area cells in planning several movements ahead. Nature, 1994;371: 413-16.


Van Lieshout,  P.H.H.M., Hulstijn, W., & Peters, H.F.M.  (1996a).  Speech production in people who stutter—testing the motor plan assembly hypothesis.  Journal of Speech and Hearing Research, 39, 76-92.


Van Lieshout,  P.H.H.M., Hulstijn, W., & Peters, H.F.M.  (1996b).  From planning to articulation in speech production—what differentiates a person who stutters from a person who does not stutter.  Journal of Speech and Hearing Research, 39, 546-564.               


Van Riper, C., The Nature of Stuttering, 2nd ed. Englewood cliffs, N.J.: Prentice-Hall, 1982.


Webster, W.G.  (1998).  Brain Models and the Clinical Management of Stuttering.  Journal of Speech-Language Pathology and Audiology, 22, 220-230.


Wu, J.C., Maguire, G., Riley, G., Fallon, J., LaCasse, L., Chin, S., Klein, E., Tang, C., Cadwell, S., Lottenberg, S.  (1995).  A positron emission tomography [18F]deoxyglucuse study of developmental stuttering.  Neuroreport, 6, 501-505.


Wu, J.C., Maguire, G., Riley, G., Lee, A., Keator, D., Tang, C., Fallon, J., Najafi, A.  (1997).  Increased dopamine activity associated with stuttering.  Neuroreport, 8, 767-772.


Wingate, M.E., Effect on stuttering of changes in audition, Journal of Speech and Hearing Research, 13 (1970), 861-873.


Wingate, M.E., Stuttering: Theory and Treatment.  New York: Irvington, 1976.


Yairi, E.  (1983).  The onset of stuttering in two- and three-year old children: A preliminary report.  Journal of Speech and Hearing Disorders, 48, 171-177.