Investigations into the cause of stuttering have provided a wide variety of interesting etiological theories over the last century. They have included diverse models such as stuttering as a result of neurological differences in cerebral organization, stuttering as a form of neurotic behavior, stuttering as an operantly-conditioned learned response, stuttering as an improperly activated laryngeal reflex, and stuttering resulting from multiple factors in a combinatorial model where causal factors may vary among individuals. Recent research by Wu, Riley, Maguire, Najafi, & Tang (1997) utilizing Positron Emission Tomography (PET scans) has shown greater than normal levels of blood flow in the corpus striatum region of the basal ganglia in the brain during speech production in people who stutter (PWS) indicating abnormal levels of activity in that region. The authors hypothesize that this finding implicates the possible involvement of the corpus striatum in stuttering. Their hypothesis is further buttressed by the reduction in stuttering noted by individuals taking medication such as haloperidol, risperidone and olanzapine, neuropharmaceutical agents that affect uptake of the neurotransmitter chemical dopamine (Maguire, Riley, Wu, Franklin, & Potkin, 1997). Dopamine is one of several neural activity modulating chemicals utilized by the nerve cells in that region, as well as in others. Several neurological disorders characterized by abnormal movement patterns (movement disorders) are associated with dopamine transport or uptake problems in the basal ganglia.

The corpus striatum is part of an extensive neurologic control system known as the basal ganglia (BG). The basal ganglia consists of multiple gray matter structures (clusters of hundreds or thousands of neurons/nerve cells known as nuclei, with the neurons in each nucleus involved in the same specific activities) and their interconnections. The nuclei of the BG in turn interconnect with multiple cortical and subcortical sites, including areas involved with spontaneous movement, with learned patterns of movement, with emotions, and with memory. The structures comprising the basal ganglia are located deep in the brain beneath the cerebral cortex, surrounding the thalamus and hypothalamus. The basal ganglia are generally believed to be responsible for modulating and facilitating various motor and cognitive programs, including complex movements such as those involved in speech production (Marsden, 1982; Mink, 1996). The basal ganglia structures thought to be important in controlling movement include the caudate nucleus (which along with the lentiform nucleus and the interconnecting neurons forms the corpus striatum), the putamen, the globus pallidus (both internal and external segments), the subthalamic nucleus, and the substantia nigra (pars compacta and pars reticulata).

Disorders or dysfunction within the basal ganglia are commonly associated with impairment of movement or the presence of unwanted movement. It is important to remember, however, that via the extensive interconnections of its component structures with other cortical, subcortical, and thalamic structures, the BG is also involved in multiple cognitive and emotive functions, any or all of which may modulate movement (Hoover & Strick, 1993). Of particular interest are the BG interconnections with the limbic system, thus allowing autonomic functions and certain aspects of emotion to play a modulating role in movement, as well as its reciprocal function, allowing the striatum to affect the mechanisms of the prefrontal cortex, generating cognitive, rather than musculoskeletal influences (Nauta, 1987). The nucleus accumbens of the striatum along with other structures of the ventromedial portion of the BG receive input from limbic structures including the anterior cingulate cortex, the orbitofrontal cortex, the hippocampus, the lateral habenular nucleus, the hypothalamus, and the amygdala.

Many authors have speculated that some type of basal ganglia dysfunction may play a role in stuttering. The earliest mention appears to be that of German researcher, Seeman, in 1934, who felt the striatopallidal pathways (corpus striatum to globus pallidus) left the speech production motor activity subject to interruption via the effects of emotional factors on the autonomic nervous system. Gracco (1991) hypothesized that two major brain systems, the basal ganglia (BG) - supplementary motor area (SMA) system and the cerebellar - premotor system play primary roles in implementing the neuromotor sequencing and modulation necessary for speech production. He cited speech movement disorders characteristic of Parkinson’s Disease as support for the BG-SMA system’s role in scaling the movements of all the articulatory structures to the specific speech movements. Caruso (1991) reviewed the research examining positioning and temporal timing of not only speech production structures but also non-speech structures (eye and finger movements) in both stuttering and nonstuttering individuals. He concluded that for at least a certain subsection of people who stutter, stuttering may "involve a generalized neuromotor deficit that involves temporal control of both speech (e. g. orofacial) and nonspeech (e. g. eye and finger) movements". Caruso hypothesized that the BG-SMA system would be a possible source for the temporal impairments he observed. Abwender, Trinidad, Jones, Como, Hymes, & Kurlan, (1998) compared stuttering and Tourette syndrome (caused by BG disorders) and found enough similarities to suggest that the two might be pathogenetically related. Similarly, Kiziltan & Akalin, (1996) speculated that stuttering may be a form of action dystonia (another movement disorder associated with basal ganglia involvement).

Of particular interest when considering stuttering as resulting from basal ganglia dysfunction is the multitude of interconnections of the BG with other cortical and subcortical systems, thus affecting and allowing cognitive, autonomic, and emotional factors to play a role in modulating motor function, and vice versa. One of the stumbling blocks to a neurologic theory of stuttering is the variability of the disfluency. The majority of PWS can speak with no disfluency when speaking in situations with little or no communicative pressure, such as talking to a child, a pet, or oneself, but will experience noticeable difficulty talking in other situations. If it is a neuromotor disorder, one might expect the disfluency to be present no matter what the situation or setting. The high degree of interconnectivity of the BG to other structures, however, would let emotions, memories, anxiety, and fear have an effect on motor function, as well as allowing improper motor learning and patterning to take place.

Another interesting aspect of movement disorders that appears somewhat homologous to stuttering is the variability in the cause of the basal ganglia dysfunction that is at the root of the movement disorder. Rather than a single unitary cause, the impairment or disease involving the basal ganglia can come from a variety of sources. As in stuttering there are genetic links for some individuals based on a pattern of familial occurrence. Yet a large percentage show no clear genetic links. Diseases, metabolic disorders, lipid disorders, injury or trauma, exposure to toxins or drugs, and vitamin deficiency have all been linked to the etiology of dystonia (Jankovich & Fahn, 1998), and similar patterns are seen for other forms of movement disorders. This is similar to stuttering, in that a clear linkage to any unitary etiology has yet to be shown (and is what has lead, in part, to the multifactor etiological models currently in vogue).

Many types of movement disorders are associated with disease, damage, or malfunction within the basal ganglia. Each is at least partially identified either by the nature of the abnormal movements that are observed, or by a specific impairment in movement. In other words, to at least some extent, each is differentiated from the others by having movement difficulties or abnormalities unique to itself. For example, patients with dsytonia have co-contraction of antagonistic muscles, yielding involuntary, sustained, and often repetitive muscle contractions of limbs, torso, neck, larynx, tongue, eyes, or head. Tourette syndrome (TS) is classified as a tic disorder, with rapid, brief, jerk-like movements or sounds. Parkinson’s disease (PD) is recognized by a triad of symptoms, including rigidity of muscles, tremor, and hypokinesia (reduced range of movement). While each disorder can be differentiated on the basis of symptoms, they often also share symptoms with each other. Some form of tremor accompanies many of the movement disorders or syndromes. Tremor is a rhythmical alternation of contractions in antagonistic muscles, resulting in a rhythmic shaking or stroking movement of the fingers, hand, limbs, larynx, head, etc, generally at a specific rate (for example, in PD at a rate of 3.5 — 7 cycles per second, and at a rate of 6-10 per second in some forms of dystonia). Some of the disorders, such as PD are characterized by tremor at rest (when muscles are not being moved purposefully); others, such as dystonia are characterized by intentional/volitional tremor (observed only when engaged in a task, such as when reaching or walking). There are, therefore, both significant differences and significant similarities in symptoms between the various movement disorders. What is important, however, for the purposes of this paper, is not the difference between the disorders, but the similarities — are there characteristics shared by many basal ganglia based movement disorders that are also observed in the stuttering syndrome that would suggest a similar underlying etiological mechanism?

Tourette syndrome might be a good place to start our examination of movement disorders, for it is one that has a clear association with stuttering. Comings and Comings (1987) reported a high co-occurrence of stuttering in children with TS, ranging up to 31.3%, although other studies have placed the figure at half of that (15.3%; Pauls, Lechman, & Cohen, 1993). The Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (American Psychiatric Association, 1994) includes the following diagnostic criteria for TS: both multiple motor and one or more vocal tics at some time during the illness; tics occurring many times a day, usually in bouts and nearly every day or intermittently throughout a period of more than a year; marked distress or significant impairment in social, occupational, or other important areas of functioning; onset before 18 years of age; and the disturbance is not due to the direct physiologic effects of a substance or general medical condition. The Tourette Syndrome Classification Study Group (1993) has elaborated on the diagnostic criteria, removing the requirements that the symptoms must cause "marked distress" and they called TS a "heterogeneous disorder", indicating the diversity of its symptomatology between individuals. The study group went on to include a list of associated behavioral characteristics seen in individuals with TS: obsessive-compulsive behavior (OCB), attention deficit hyperactivity disorder (ADD), impulsivity, aggressiveness, anxiety, phobias, depression, self-injury, low frustration tolerance, poor socialization, and low self-esteem.

Abwender et al. (1998) have written about similarities between TS and stuttering: these include a tendency for the disorders to run in families; more common in males by a ratio of 3 or 4:1; childhood onset, symptoms may lessen in adulthood; follows a characteristic waxing and waning course; symptoms can be at least partially suppressed voluntarily; and both are influenced by emotional factors with a clear pattern of exacerbation under emotional stress. Abwender and his associates also conducted clinical interviews and a limited neuropsychological assessment with 10 stuttering children and 12 adult PWS. Only one of the individuals, a child, carried a diagnosis of comorbid TS. They reported observing motor tics in 4 of the 10 children (40%) and 7 of the 12 adults (58%). In clinical interviews, 3 of the children and 3 adults reported symptoms consistent with ADD, 5 children and 9 adults reported obsessive-compulsive behaviors in the interview, and 2 children and 9 adults obtained clinically significant scores on an OCB inventory (OCB Inventory; Frankel, Cummings, Robertson, Trimble, Hill, & Benson, 1986). According to the researchers, only one of the 22 subjects presented no evidence of TS-like neurobehavioral features.

A final comment as we leave Tourette syndrome. TS is one of many movement disorders classified as tic disorders. Tics are defined as involuntary sudden repetitive movements or vocalizations. They tend to be paroxysmal, abrupt, and only rarely continual or prolonged. They are extremely variable and differ greatly in location, severity, and/or frequency between patients. Tics typically wax and wane, and are often increased by anxiety, anger, excitement, and fatigue (Singer, 1998). While one of the differential diagnostic features for TS is its chronicity, many other tic disorders, especially those seen in children, are notable for their transitory nature. The Tourette Syndrome Classification Group classifies tics as transient if they are present for less than 12 months. "Transient tic disorder" is the mildest and most common form of tic disorder, with estimates of occurrence ranging from 5-24% of school-aged children. Its very name indicates that it is spontaneously ameliorating, as is stuttering, in a large number of young children. Similarly, while "chronic motor or phonic tic disorder" may consist of exclusively motor or exclusively vocal tics that have been maintained for more than a year, there is still a high level of spontaneous amelioration, of almost 50%. Neurologically-based disorders are not necessarily permanent. Even TS itself shows a high level of remission or at least diminishment. Erenberg, Cruse, & Rothner (1987), in a long term follow up study of 58 individuals with TS, reported that tics virtually disappeared in 26%, diminished considerably in 46%, remained stable in 14%, and increased in 14%.

Dystonia is another form of movement disorder that shares many similarities with stuttering. It is characterized by involuntary, sustained, and often repetitive muscle contractions of antagonistic muscles, causing twisting movements and/or abnormal postures. While classically these have been characterized as slow movements, several forms of dystonia with rapid movements (resembling myoclonus/myoclonic contractions) and more similar to movements seen in stuttering, also exist. The movements may be focal, present only in single structures [e. g. blepherospasms (eyelid movements) or laryngeal or palatal dystonias], or segmental, multifocal, or generalized, (involving whole segments of muscles or muscle systems, such as the torso, head and neck, and/or limbs). Dystonic movements can occur at rest, but generally are initiated or exacerbated by voluntary motor activity (known as action dystonia), and may be exclusive to a particular task or action. Writer’s cramp (graphospasm) would be an example of this type (Marsden & Sheehy, 1990) and similar task-specific dystonias may effect musical instrument-playing, typing, chewing, and speaking. Scolding, Smith, & Sturding (1995) reported a vocal dystonia associated with the rapid speaking style used in auctioneering.

Another form of dystonia that shares similarities with stuttering is a fluctuating form known as paroxysmal dystonia, characterized by a sudden onset of dystonic movements that may be triggered by kinesigesic (movement) or nonkinesigenic events (which may occur unpredictably without any particular precipitant). The contractions seen in paroxysmal dystonia may last only for a few seconds, and may be exacerbated by stress. As is seen with stuttering, the timing and intensity of the movements can be influenced by various factors, including emotion, stress, fatigue, relaxation, and other motor activity (Jankovich & Fahn, 1998). Dystonic movements can sometimes be relieved by rest, self-hypnosis, and various sensory tricks or counterpressure on muscles.

One other form of dystonia that is of particular interest when comparisons are made to stuttering is a form of focal dystonia known as laryngeal dystonia. Laryngeal dystonia is more commonly referred to as spasmodic dysphonia (SD). Originally thought to be psychogenic in origin, the adductor form of spasmodic dysphonia (in which the thyroarytenoid/vocalis muscle undergoes sudden hypercontraction, yielding an effortful and strained speaking pattern interrupted by frequent breaks in phonation) is now viewed by many as a task-specific focal (laryngeal) dystonia. As in stuttering, a gradual progression of symptoms is often seen, In SD, it often begins by affecting only the speaking voice or only the singing voice and may be somewhat situationally linked, but symptoms usually progress to involve both types of vocalization and a pervasive occurrence pattern .

Many patients with SD also have voice tremor or essential tremor in other muscles (another basal ganglia disorder symptom, or at least basal ganglion—cerebellar loop based) which may have preceded the onset of SD by several years. Rosenfield, Donovan and Sulek (1990) reported that essential tremor accompanied the dystonia in 71 out of 100 patients with SD and in 25 of 100 patients with cranial dystonia. Van Riper (1971) reported a high incidence of tremor during speech production in people who stutter, primarily in structures such as the lips, jaws, tongue, and extrinsic laryngeal muscles. While relatively subtle in amplitude, frequency patterns appear to center in the 7-9 cycles per second range, similar to that observed in other movement disorders. Patients with spasmodic dysphonia may also complain of difficulties breathing, and respiratory muscles may be affected even without laryngeal involvement (Lagueny, Burband, & Le Masson, 1995) and result in deep inspiratory gasps, loud breathing, respiratory arrests, and respiratory deregulation, all of which are symptoms described in some individuals who stutter.

Similar to the Abwender et al. study, Kiziltan and Akalin (1996) compared similarities between focal action (laryngeal) dystonia and stuttering. They reported observing abnormal involuntary movements, involving principally the facial and neck muscles, in 23 stuttering patients. They also reported a statistically significant occurrence of stuttering in the family history of patients with idiopathic torsion dystonia. While there doesn’t seem to a great deal of current empirical support for their suggestion that stuttering is a form of laryngeal dystonia, there remains the possibility that the two may reflect different manifestations of a similar underlying etiology. At the Auburn University Speech and Hearing Clinic we have recently encountered an interesting duo of a father with adductor spasmodic dysphonia, and his son who exhibits stuttering. In addition, there is an extensive history of stuttering in the family.

Symptomatology of several forms of movement disorders that have been linked to basal ganglia dysfunction were reviewed. Across the movement disorders examined in the paper, multiple similarities exist, including similarities with behaviors and patterns observed in stuttering:

• The disorders may result in abnormal vocal productions or at least affect structures and systems utilized in speech production, including respiration, laryngeal function, and/or articulatory production.
• There tends to be a higher prevalence of the disorders in males than in females.
• There is a familial history of the specific disorder for a significant number of individuals, indicating a possible genetic transmission pattern.
• There does not appear to be a unitary causal factor for each disorder; the underlying basal ganglia dysfunction may be the result of genetic factors, diseases, metabolic disorders, lipid disorders, injury or trauma, exposure to toxins or drugs, or vitamin deficiency.
• The majority either demonstrate a childhood onset or occur frequently in children.
• For several of the disorders there is a relatively high rate of spontaneous remission.
• Symptoms may lessen in adulthood.
• Many demonstrate a waxing and waning pattern over days, weeks, or months.
• The degree or occurrence of motor impairment is increased under conditions of emotional states or stress, excitement, and complexity of demands.
• Symptoms may be lessened by heightened efforts to voluntarily suppress them, as well as under conditions enhancing relaxation or utilizing novel treatments such as auto-hypnosis or the use of muscular pressure points.
• Several are accompanied by associated symptoms that might include attention deficit disorder, obsessive-compulsive behaviors, impulsivity, anxiety, depression, phobias, low frustration tolerance, poor socialization, and low self-esteem (some or all of which may share a similar neurochemical dysfunction with the movement disorder).

The high degree of interconnectivity of the basal ganglia offers a possible explanation for the variability in disfluency observed for most PWS, as well as the apparent role that prior negative experience and anxiety play in generating disfluency. Interconnections of the BG with limbic system structures, prefrontal cortex, sensorimotor cortex, and cortical motor areas, would let emotions, memories, anxiety, and fear have an effect on motor function, as well as allowing improper motor learning and patterning to take place. The secondary or associated behaviors that accompany stuttering, such as facial, head, trunk, or facial movements and vocal interjections, have generally been considered to be compensatory behaviors learned in efforts to avoid, postpone, or escape from a moment of stuttering. While this is a plausible explanation, an alternate hypothesis would be that some of those behaviors might reflect overflow or abnormal movements or vocalizations generated by an underlying basal ganglia abnormality. There seems to be enough empirical support to continue to keep basal gangliar dysfunction as a leading contender as an etiological factor in stuttering.