Oromotor Nonverbal Performance and Speech Motor Control

Version	Summary	Created by	Modification	Content Size	Created at	Operation
1		Gary Weismer	--	6295	2023-05-10 00:37:52	\|
2	update references and layout	Rita Xu	-17 word(s)	6278	2023-05-10 04:01:09	\|

This entry is adapted from the peer-reviewed paper 10.3390/brainsci13050768

This position research offers a perspective on the long-standing debate concerning the role of oromotor, nonverbal gestures in understanding typical and disordered speech motor control secondary to neurological disease. Oromotor nonverbal tasks are employed routinely in clinical and research settings, but a coherent rationale for their use is needed. The use of oromotor nonverbal performance to diagnose disease or dysarthria type, versus specific aspects of speech production deficits that contribute to loss of speech intelligibility, is argued to be an important part of the debate. Framing these issues are two models of speech motor control, the Integrative Model (IM) and Task-Dependent Model (TDM), which yield contrasting predictions of the relationship between oromotor nonverbal performance and speech motor control.

speech motor control oromotor nonverbal motor control dysarthria

1. Introduction

Historical Perspective

The significance of nonspeech, oromotor control for theories of speech motor control, and for the clinical diagnosis and management of speech motor control deficits have long been written about in the literature. Relevant publications extend back in time in the literature at least to the 1940, when Emil Froeschels (1884–1972), one of the original academic speech pathologists, advocated chewing as a therapeutic technique for stuttering and other speech disorders ^[1]^[2]. (I use Froeschels as a blurry reference point in time for the beginning of “academic” publications on the topic of oromotor nonspeech postures and gestures and their relationship with speech (and voice) production. An earlier example was published in 1893 by Oskar Guttman in a book titled, Gymnastics of the Voice, a method for song and speech, also a method for the cure of stuttering and stammering ^[3]. This second edition in English, translated from a version (one of at least four), was originally written in German. Guttman, writing on exercises for the tongue, (which he claimed most people regarded as no more than a “helpless lump of meat” (p. 83), suggested back and forth tongue tip movements between the corners of the mouth, slowly at first and gradually ramping up to maximum speed. Examples of other tongue exercises (among 16) included protrusion–retraction cycles and rapid “dotting” of the upper lip with the point of the tongue. Lip exercises included practice for independent movements of the upper and lower lip and a velar exercise of raising and lowering the soft palate with the mouth open wide. Close reading of Guttman’s book suggests it was intended more for singers than speakers, but he clearly meant the oromotor nonverbal exercises to apply equally to speech, as suggested in the subtitle of the book. Other similar manuals and textbooks were written in the late nineteenth and early twentieth centuries. They are written still today). For Froeschels ^[2], the act of silent (mimed) chewing, especially with exaggerated movements of the tongue and lips, harkened back to the biological origins of modern voice and speech production. Preverbal people discovered that these motions could be superimposed on sound made in the throat, resulting in vocal tract signals that changed with the motions. Chewing, in Froeschel’s view, took a patient back to the most basic of oromotor behaviors, a logical starting point for voice and speech therapy. Froeschels (Ref. ^[1], p. 127) offered this anecdote as evidence for this view: “…gum chewing played an interesting part in my studies on the influence of chewing on voice and speech. Two years ago, my wife and I made a trip on a Greyhound bus. The drivers changed regularly, and the new driver always addressed the travelers. I could then predict to my wife, without a single error, whether or not the driver chewed gum, making the differential diagnosis from the impression of a hyperfunctional or a hypofunctional voice, on the one hand, or of a normal voice, on the other. The normal voice indicated to me that the man was used to chewing gum”. The shared anatomical structures for chewing and speech production led Froeschels to claim, “In fact, only one single center (in the anterior central convolution of the brain) dominates the movements of chewing and talking. It stands to reason that two different functions cannot be performed at the same time by one single part of the body. From this, researchers can conclude that, as far as the movements of the mouth are concerned, what has been considered two functions, namely talking and chewing must be one and the same” (Ref. ^[2], p. 9). (The paper appeared in ^[2] and is reprinted in Froeschels’ Selected Papers of Emil Froeschels, 1940–1964 ^[4]). As discussed below, Froeschel’s views are a variation on the common effector arguments for the relevance of oromotor nonverbal control to speech motor control ^[5]^[6]; see the discussion in ^[7]. Obviously, you can chew and speak at the same time, as evidenced by the millions of mothers and fathers who have told their children not to. (A close reading of Froeschels publications on the near identity of motor control for chewing and speech suggests some inconsistencies in his thinking about doing two different things simultaneously with shared effectors. Usually, his view is the one expressed above, but at least one example of a different opinion, that you can chew and speak at the same time, can be found in the chewing papers collected by Froeschels ^[4]. The book also contains papers on deafness, psychoanalysis, stuttering, aphasia, and language characteristics).

2. Definitions

2.1. Operational Definitions

The following definitions are not meant to improve those for the same or similar terms in other publications.

Several definitions are revisions of ones in Weismer ^[6]. In his 2017 publication, Maas claimed that a problem with the oromotor nonspeech/speech production controversy is the absence of precise definitions required to frame the debate ^[8]. Precisely stated (but nevertheless arguable) definitions of the terms, “dysarthria”, “Mayo Clinic approach”, “oromotor nonverbal tasks”, “quasi-speech tasks”, and “explanation” are given in Weismer ^[6]. Herein, I revise and broaden definitions for “dysarthria”, “oromotor non-speech tasks”, and “quasi-speech tasks”. New definitions are provided for “task specificity”, “apraxia of speech”, and “speech motor control”, which surprisingly (to the author as well) were not formally included in Weismer ^[6] but see page 329). A brief note follows some of the definitions.

2.1.1. Speech Motor Control

Speech motor control is defined as the goal-directed, nervous system control of speech production by which movements, and their derivatives, effect shapes and configurations of structures throughout the speech mechanism, including laryngeal and respiratory structures, all planned and executed as a function of time to produce sequences of acoustic phonetic events for the purpose of communication. Acoustic phonetic output of the vocal tract is an integral component of speech motor control processes, not separate or subtractable from them. The speech acoustic signal is intertwined with the movements and their derivatives stated above by calibrating the latter with the former, and the former with the latter. Note: The claim that the speech acoustic output is the goal of speech motor control is based on the work of Guenther ^[9], Perkell ^[10] and Hickok and Poeppel ^[11]. See Bose and van Lieshout ^[12] for a different view.

2.1.2. Motor Speech Disorder

A motor speech disorder is a neurologically based disorder of speech motor control in which movements of the upper airway structures (including the velopharynx) and the constrictions and configurations they produce, as well as movements of the laryngeal and respiratory structures, and/or the planning and programming of those movements, are affected in ways that result in a speech acoustic signal that is phonetically and/or prosodically degraded and therefore cause deficits in speech intelligibility and/or naturalness. Note: “Neurologically based disorders” are those demonstrated on imaging studies (e.g., as in stroke, or multiple sclerosis) or those inferred from signs and symptoms (e.g., as in Parkinson’s disease or childhood apraxia of speech). Examples of signs of a speech motor control disorder include, but are not limited to, disorders of speaking rate, speech rhythm, sound-segment production, speaking fundamental frequency and/or sentence-level pitch modulation, prominence of syllables for lexical stress, and regulation of speech loudness.

2.1.3. Dysarthria

Dysarthria is a type of motor speech disorder, primarily regarded as an execution deficit resulting from muscle weakness, abnormalities of muscle tone, timing and coordination, in which the speech deficit is secondary to documented neurological damage resulting from a number of different diseases. Different types of dysarthria, developed in the perceptual studies of Darley, Aronson, and Brown ^[13] and refined by Duffy ^[14], are thought to be associated with damage to different parts of the nervous system. Note: As documented below, a reliable match between speech symptoms and dysarthria type, or between speech symptoms and disease diagnosis, has not received much support in the literature. This calls into question the suggestion of different speech therapies according to dysarthria type ^[14] as well as the frequent use of dysarthria or disease type as an experimental blocking variable.

2.1.4. Apraxia of Speech

Apraxia of speech (AOS) is a speech planning or programming dysfunction, as in the preparation of a sequence of syllables ahead of their production. AOS is a motor speech disorder in which the speech deficit is not thought to be the result of muscle weakness, tone abnormalities, and so forth. AOS is therefore different from dysarthria, although a simultaneous diagnosis of apraxia of speech and dysarthria can be made in the same patient. AOS is not considered in depth.

2.1.5. Oromotor, Nonverbal Tasks (Nonspeech Oromotor (NSOM) Tasks)

Any performance task, absent phonetic goals, in which structures of the speech mechanism—especially those of the upper airway—are measured for any aspect of force production (maximum force, ‘‘fine’’ (submaximal) force, positional or force accuracy and/or hold duration (endurance), stability, repetition rate, impulse production, tone, range of motion, speed of motion, movement repetition rate and regularity, structural shape/configuration (e.g., lip rounding), and/or tracking accuracy). Note: The term and its acronym “Nonspeech oromotor” (NSOM) is in frequent use in the literature, most often to describe treatments used in the speech clinic, many of which are manualized for clinician use ^[15]. Nonspeech tasks performed by speech structures that are not part of the upper airway, such as laryngeal and respiratory structures, have also been studied, but are not discussed here.

2.1.6. Task Specificity

In motor control, task specificity is accomplished by flexible “tuning” of sensorimotor processes within nervous system networks to achieve a specific behavioral action/goal. Task-specific tuning is implemented by differential activity within the motor control network of fiber tracts and/or cortical and subcortical cells connected by these tracts. The tuning is learned and refined with extensive “practice” to achieve the desired goal; in the case of speech production, “practice” is not used in the sense of mastering a musical instrument or sports skill, rather it is embedded in the production of millions of utterances. The goal of speech motor control is to generate an acoustic phonetic output of the vocal tract that serves the purpose of communication. Note: This definition does not equate “tunings of networks” with specific structures and/or mechanisms.

2.1.7. Quasi-Speech Tasks

Tasks performed, on command, in which structures of the speech mechanism produce phonetic (or phonetic-like sounds, as in sustained vowels or fricatives), single syllables that are not words, diadochokinetic (DDK) tasks using repetition of a single syllable (e.g., /pʌpʌpʌ…/), or syllables with varying phonetic content (e.g., /pʌtʌkʌ…/), or sequences of other novel sounds or phonatory events, which have no lexical or paralinguistic function. Quasi-speech tasks may also include maximum and minimum fundamental frequency (F0), maximum and minimum phonatory intensity as a function of voice F0, F0 glissandos at different rates of change, and a variety of other sounds (e.g., raspberries, hissing sounds, rhythmic sequences of mouth sounds). Note. Quasi-speech tasks are difficult to define as the question of what is lexical or paralinguistic is subject to debate. For example, an argument can be made for hissing as lexical and/or paralinguistic, depending on context. The sound of hissing is not a word but may serve a communication purpose just like the word “hiss”. Ziegler et al. ^[7] refer to these quasi-speech events as non-speech, which in the current opinion is a reasonable way to categorize these productions. Quasi-speech for these events instead of non-speech because it is relevant to the idea of a continuum of tasks ranging from clearly not speech, to increasingly more speech-like events until the endpoint, “speech” is reached (see ^[16] discussion of different DDK tasks as points along such a continuum). The common factor among various quasi-speech tasks is that they require production of an acoustic signal, which may or may not be phonetic. The definition is not meant to apply to prelingual children because they may produce segmental and phonatory behaviors before understanding the directions to perform such tasks. Ziegler et al. ^[7] refer to repetitive syllable tasks as “nonspeech” tasks. I mostly agree with that classification—DDK tasks are not speech in the sense that they are not composed of sounds meant to be comprehended—but choose to use the narrower term to include other tasks as listed above.

2.1.8. Subsystems

Partitions of peripheral speech mechanism structures conceptualized as independent, functional components of speech production. Note: The partitions take various forms in the literature.

2.1.9. Mechanism

Nervous system processes for control of action; in the case of speech motor control, the neural processes organized and tuned to produce an acoustic signal having phonetic value for communication.

3. Two Models of Speech Motor Control

3.1. Integrated Model (IM) versus Task Specific/Task Dependent Model (TSM/TDM)

Over the past six decades, many models or theories of speech motor control have been put forth (in the current paper, the terms “model” and “theory” are used interchangeably with full recognition of the important technical distinction between the two). Two models with implications for the relevance of oromotor nonverbal behavior to articulatory behavior are discussed here. The Integrated Model (IM) and the Task-Specific or Task Dependent Model (TSM, TDM) are at odds in this debate. They are worth discussing because they each have potential value in guiding the development of experimental hypotheses that are relevant to theories of speech motor control and clinical practice. Other, more formal (computational) models of speech motor control are available ^[9]^[17], and in fact, the IM and TDM share some assumptions with these models.

Models that differ in assumptions and content are tested against each other by contrasting predictions. The best-known example of prediction “contests” in speech science research pits the motor theory of speech perception against “auditory” theories of speech perception (for review, see ^[18]^[19]). In brief, the original formulation of the motor theory held that in humans, a species-specific, dedicated brain module served the function of speech perception ^[20]^[21]. The module does not use general auditory mechanisms in the brain to perceive speech. Rather, the module represents the objects-to-be-perceived as articulatory gestures—phonetic identities. These identities are automatic outputs of the module, and do not require transformations of the physical signal entering the central auditory pathways. The module, a product of evolution, is dedicated to the perception of speech and reflects the tight coupling in humans of speech production and speech perception. In contrast, auditory theories of speech perception suggest that general auditory mechanisms are employed to perform the perception of speech in humans, much as they are assumed to be for the perception of any auditory event; a special, species-specific analysis mechanism is not required.

The two theories of speech perception make opposing experimental predictions. The motor theory predicts that animals should not respond to variations in speech stimuli in the same way as humans—animals do not possess the special mechanism that would allow them to do so. In contrast, auditory theories predict the perception of speech signals is likely to be similar in both animals and humans because both use general auditory mechanisms to process the signals. This is a simplified statement of opposing predictions made by the two theories, but in general the predictions are consistent with their contrasting structure. Other theories of speech perception exist as well but are not discussed here; see ^[18]^[19].

This brief presentation of a well-known theoretical debate between special versus non-dedicated mechanisms in speech perception is directly relevant to the structure and predictions derived from the IM and TDM. The IM and TDM make opposing predictions as well, some as clear cut as the one just described for the case of speech perception. I show how the structure of the IM and TDM generate these predictions. To the best of my knowledge, previous publications on the nonspeech-speech issue have not described explicitly such predictions and how they derive from the respective structures of the models. Ironically it is possible the specificity of the opposing predictions presented below must be regarded as tentative because they are derived from two “models” that lack specificity.

The current structure of both the IM and TDM do not qualify as true models. More precisely, both are sets of statements that resemble assumptions, like primary premises in deductive reasoning. Both qualify as coarse-grained theoretical frameworks, with the proviso that a widely accepted definition of “theoretical framework” has been difficult to establish for research on speech motor control (see ^[22]).

When an SLP employs oromotor nonverbal or quasi-speech tasks to diagnose and/or treat a speech production deficit in a person with dysarthria, it is reasonable to ask, what are the reasons for this choice? Stated otherwise, what evidence is available and defensible for the effectiveness of oromotor nonverbal tasks in diagnosing a speech production disorder in dysarthria, or treating it with such tasks. (It is interesting that in the surveys that have been done on the use of oromotor nonverbal tasks in diagnosis and treatment of motor speech disorders, the data have focused primarily on whether or not the tasks are used, rather than why SLPs use them ^[23]^[24]. Models and theories, even if provisional, are important to how this question is answered. I do not discount the value of clinician experience in choosing a diagnostic or treatment approach. An SLP’s experiential database of therapeutic outcomes, and what seems to have caused (or not caused) them is important and should be included as a significant component of clinical decision making ^[25]^[26]^[27].

3.1.1. Structure of IM

The IM assumes, explicitly, that general motor processes are recruited for the production of speech. “Ziegler has interpreted our position as a task-independent model of motor control and in particular of speech motor control…researchers believe their model is better described as an integrative model in which some nonspeech motor tasks share principles with speech and some do not—that is, the speech motor system is integrated into a more general motor system. This leads researchers to postulate overlapping neural and behavioral systems for the control of speech and volitional nonspeech tasks” (Ref. ^[28] p. 38). Linguistic units such as phonetic segments are independent of motor processes; a domain-independent motor system transforms the units to action. By “domain-independent” motor processes, I mean motor control mechanisms that are not specialized for specific actions, such as (for example) speech production, playing an instrument, or finger spelling. Rather, the motor control capabilities of the nervous system are sufficiently powerful and flexible to be recruited and shaped for any and all actions. To be fair, Ballard et al. ^[28] allow that overlap of motor control for volitional, nonspeech and speech behavior is only partial, in that some nonspeech tasks are relevant to speech motor control, others are not. Which nonspeech tasks are not relevant to speech motor control are not suggested. Ballard, Granier, and Robin (Ref. ^[29] p. 983, emphasis added) put this view succinctly: “We rely on such theoretical stances as that presented by Folkins and Bliele (1990) and Saltzman (1986) which claim that the motor system is not necessarily organized around presumed units of language or speech. Rather, it is assumed that the motor system has its own cognitive architecture that is activated and monitored, in part, by the language system”. Speech motor control is therefore thought to be based on principles common to general movement control, regardless of what the movement is in service of. Similar views, though more nuanced for the case of speech motor control are presented in ^[30]^[31].

This nesting assumption, criticized for the case of speech motor control in ^[6], and more recent publications (e.g., Ref. ^[7]), drives the idea that speech motor control can be studied most directly by eliminating the speech acoustic signal that results from movement of the articulators, larynx and respiratory system. The speech acoustic signal is linguistic (e.g., Ref. ^[32]) and should be separated from evaluation of the movement control that generates linguistic events--phonetic behavior (e.g., voice onset time [VOT], vowel, diphthong, nasal, and semivowel formant frequencies), the various acoustic contributors to word and sentence stress, speech timing, and so forth. Stated otherwise, these “linguistic” events are not in the domain of speech motor control. In addition, the assumption eliminates, to a large extent, the relevance of movement type to understanding its control. Oromotor nonverbal tasks, whatever form they take, are stripped of the complexity of speech production movements (such as multi-articulator coordination over time and space) and therefore capable of allowing a basic analysis of effector control. In other words, such tasks are designed to move the analysis up to the top of the hierarchy—the general principles of motor control, free of specific goals.

It follows that IM rejects the idea that processes and perhaps the anatomy of speech motor control are modular (see ^[33]^[34]). It has been argued that “speech is not special”—echoing the theoretical difference between the motor theory of speech perception and auditory theories—and so does not require speech tasks to study and understand speech motor control (Ref. ^[8], p. 347). As argued below, the proper study of speech motor control requires tasks that include an acoustic phonetic, linguistically relevant product of underlying movements and forces, but in no way requires a specialized speech production neural module parallel to the one proposed in the motor theory of speech perception.

The IM also assumes that measures of speech motor control are well-served by measures of the parameters mentioned above (e.g., force, distance, tracking ability), in both oromotor nonspeech and speech tasks. Such measures have been used extensively in studies of finger, hand, arm, leg, and other non-oral motor behaviors, and in oromotor behavior as well.

Another assumption of IM is that a logical study of speech motor control, as well as other motor control tasks, is performed by isolating components of an ensemble of peripheral effectors that are organized to produce speech and other oromotor behaviors. This view is motivated by two explicit, supporting assumptions. One is that different orofacial subsystems—meaning the lips versus jaw, the jaw versus tongue, and lips versus tongue, the velopharyngeal port versus any of these structures “Subsystems” is used here following ^[35]. Rong et al. ^[36] have used the term to include the respiratory, phonatory, resonatory, and articulatory subsystems, which is more consistent with typical textbook descriptions of how the speech mechanism is partitioned ^[37]—can be differentially impaired in dysarthria. For example, speakers with dysarthria may have different degrees of impairment distributed across subsystems, often referred to as “differential impairment of the articulators” ^[34]^[38]. As discussed in Weismer ^[6], for this concept to make sense requires measurement of each isolated subsystem impairment, followed by a quantitative model that combines the results across the designated subsystems to estimate the likely deficit of the integrated, upper airway effector ensemble in the production of speech. The measurement of “isolated” subsystem impairment, seen through the lens of the IM, must be performed with oromotor nonspeech tasks, because in speech production, movements of the lips, tongue, and jaw do not function independently. Weismer (Ref. ^[6], p. 321) cited a passage from ^[39] endorsing this view. Subsystem movements of the lips, jaw, and tongue for speech production are coordinated over time and space, and, in the case of the jaw and tongue, and lower lip and jaw, are coupled anatomically and functionally, albeit in a loose-to-moderate way that depends on the speech sound (or syllable) production goal. It may be argued that tongue and lip movements can be isolated from movements of the jaw during speech, by fixing the position of the latter with a bite block. However, the matter of fixing the jaw in one position and therefore neutralizing it as a contributor to articulatory events is more complicated than it seems at first glance. Folkins and Zimmermann ^[40] showed that electromyographic (EMG) signals from jaw closer muscles were present for phonetic events requiring closure of the vocal tract, even when jaw position was fixed with a bite block. More recently, Dromey, Richins, and Low ^[41] reported the qualitative observation that, in their bite block experiment, some participants occasionally “bit down” on the block when they produced sentences. Although Dromey et al. did not specify when the participants did this, it would be interesting to know if the bitedowns were intermittent, possibly timed to phonetic events that in speech require closure of the vocal tract. Isolating the jaw, tongue, or lips during speech production, with all parts moving, does not seem to be possible, at least not currently. Weismer (Ref. ^[6], pp. 320–322; pp. 332–333) presents other significant problems with subsystems analysis. It should be pointed out that ^[36] performed an analysis of each component of more traditionally defined subsystems (respiratory, phonatory, resonatory, and articulatory) using multiple speech and speech-like measures. Rong et al. tested a tentative hypothesis for their participants with ALS that the measures for each subsystem represented its isolated impairment for speech production.

3.1.2. Predictions from the IM

Predictions concerning speech production deficits follow from the structure and assumptions of the IM.

First, the IM predicts that a componential analysis of at least the three upper airway subsystems (tongue, lips, jaw) will provide useful information regarding a person’s oral communication skills. The prediction should apply to both speech intelligibility scores, as a “general” measure of speech deficit, and its underlying (“specific”, as discussed in Section 3.1) phonetic details, such as perceptually determined segmental errors ^[42] or automated estimates of segment “goodness” derived from the speech acoustic signal ^[43]. A challenge for this prediction is, as discussed above, the absence of a suitable metric of the overall deficit of orofacial muscular control based on componential analysis. This is so because the problem of “summing” the deficits of the three (or more) isolated parts and expressing the sum as an index of overall oromotor deficit has not been attempted and may not be conceptually feasible. (I am not arguing that the problem has zero potential to be resolved; “not attempted” does not strictly mean it has not been tried, but the absence of any such publications in the literature suggests either the absence of an attempt or, if attempted, a lack of success. However, a long-term research program can be envisioned to establish the proper way to express deviation from “normal” for each subsystem, perhaps using different tasks and establishing the best ones from the large number of potential oromotor, nonspeech gestures (see ^[6]^[44]). The research program could pursue formulae that combine subsystem component deficits that predict an overall, quantitative deficit of upper articulatory function for speech production, and even one that might account for specific speech sound deficits. How subsystem impairments that are quantitatively similar (e.g., 20% deviation from normal function for each subsystem) are combined for an estimate of overall oromotor deficit is a major issue that would need to be resolved. There is no a priori reason to expect 20% deviation from normal in a jaw oromotor nonverbal task to mean the same thing to speech production as a 20% deviation in tongue performance. A long-term goal would be to show that the quantitative nonspeech deficit maps onto speech intelligibility measures. I am not favorably inclined to this sort of work, because I think the effort would ultimately be futile. However, in the end, a carefully developed proposal for this kind of work requires fair consideration; science is so much more valuable than opinion. Instrumental measures for the nonspeech behaviors would be far preferable for the work, but my use of “…perceptual and/or instrumental methods” is meant to acknowledge the often-used requests to patients to, “wag your tongue back and forth as fast as possible”, “open and close your mouth as fast as possible”, “press your tongue into your cheek as hard as you can, resist the pressure I apply”, and so forth. In Ziegler et al. ^[7], multi-task performance, including speech and non-speech tasks (by their definition DDK is considered a non-speech task), addresses a similar question by performing a statistical analysis on a large group of speakers with dysarthria and showing that speech and non-speech performance form mutually exclusive groupings).

Second, based on the continuum notion of NSOM tasks, ranging from very different from speech movements (e.g., steady force efforts exerted by each of the upper airway articulators, including the measurement of maximum forces or pressures, and/or endurance of maintaining maximum or submaximum forces), to movements that are similar to speech production gestures (e.g., soundless gestures like those observed for phonetic segments), performance on more speech-like NSOM tasks should predict speech intelligibility scores and phonetic “goodness” more accurately when compared with less speech-like tasks. A problem with this experiment is identifying how components (subsystems) of oromotor control can be partitioned for time-varying tasks, such as a soundless lingua-alveolar gesture.

The IM makes predictions for the treatment of neurogenic speech production disorders, which follow from the predictions outlined above for their diagnosis. For example, the componential assumption predicts that training of oromotor skills, in the absence of phonetic output of the vocal tract, will result in improved speech production skills either globally (e.g., improved speech intelligibility) and/or locally (e.g., at the segmental level of analysis). Implications of the IM for componential, oromotor training at the segmental level, such as training focused on the lips for improvement of labial consonant production or configuration for vowels, are (to the best of my knowledge) not considered explicitly in publications advocating the IM, but data have been published that are not consistent with the hypothesis ^[7]^[45]^[46].

Finally, a strong form of the IM predicts that therapy with a goal of improving articulatory skills in persons with dysarthria will be equally effective by training NSOM tasks versus NSOM tasks plus articulation, or NSOM tasks plus auditory training. The “NSOM task alone” condition would have to be paired with another form of non-oromotor training performance (e.g., a rhythm or reaction time task) to balance the conditions comparisons.

3.1.3. Structure of the TDM

Ziegler ^[47] presents a “task-dependent model” (TDM) of speech motor control. The TDM contrasts with what he calls a “task-independent model”, which is similar if not identical to the IM. Stated otherwise, the “task-independent” model is consistent with the recruitment of general motor processes, as defined above, to implement the movements, forces, and sensorimotor mechanisms observable in speech production. In contrast, Ziegler’s task-dependent view is that speech motor control is implemented by motor processes unique to speech production. Ziegler argues that speech motor control is unlike motor control for non-verbal, oromotor actions such as those listed in a compendium published in ^[44].

The structure of the TDM, as currently developed, is coarse-grained in the same way as the IM. Like the IM, the model is best described by a series of statements, rather than by a more formal structure.

First, in the TDM, sensorimotor control processes for speech production are qualitatively different from those for oromotor control of non-speech actions. Motor control processes for the actions produced by structures of the speech mechanism are specific to speech production. In this view, the idea of a continuum of oromotor movements and configurations that vary in similarity to those required to produce speech does not make sense. Importantly, the sensorimotor processes in the TDM—that is, in speech motor control—include the speech acoustic signal and its integration with respiratory, laryngeal, velopharyngeal and articulatory movements, applied forces, and configurations. They also include sensorimotor processes for generating and regulating airflows and pressures and integrating them with movements and configurations. Sensorimotor processes in the TDM, including feedback from the speech acoustic signal and tactile and proprioceptive sensors (among other sensory information from structures of the upper airway, larynx, and respiratory system) are not separable from the gestures observed in speech production. They are an integral component of speech motor control ^[9]^[10].

Second, the TDM as currently conceptualized does not specify in precise neural terms how task dependency fits in with speech motor control processes. One possibility, discussed above, is that a dedicated nervous system network of nuclei and fiber tracts serves as the mechanism of speech motor control (i.e., it is a module). An alternative is that speech motor control may be vested in a learned and overpracticed “tuning” of interconnected nuclei and fiber tracts, developed from early childhood on, repeatedly refined by the millions of calibration samples and refinements that come with talking.

Third, the listener’s requirements for an intelligible speech signal enter the domain of, but not necessarily mechanisms for, speech motor control. These requirements are part of the goals of speech motor control, in the sense that they can dictate details of the task. Speaking metaphorically, the TDM allows, by alternate tunings of the neural substrate for oral communication, the speaker to know and adapt to what the listener needs ^[48].

Finally, the TDM is not consistent with the componential approach to the study of speech motor control. In the TDM, the goals of speech production—the production of acoustic signals useful for communication—are not accomplished by independent mappings of subsystem oromotor nonverbal performance to specific phonetic segments.

3.1.4. Predictions of TDM

Several predictions emerge from the TDM that are opposite to those from the IM. First, the TDM predicts that measures of oromotor control for isolated structures such as the lips, tongue, and jaw, derived from non-speech actions,: will not predict an overall or detailed analysis of a speech production deficit, either summed across the structures or within single structures. A measure of verbal deficit in speech motor control is speech intelligibility, or other coarse-grained perceptual measures including but not limited to scaled estimates of severity. A “detailed analysis” might include perceptual evaluation of speech sound accuracy and prosodic characteristics, acoustic measures of different classes of speech sounds, measures positions, displacements, and velocities of movable upper airway structures or analysis of imaged tongue positions or configurations. Stated otherwise, the TDM predicts that componential analysis will not sum to yield a variable that will predict speech intelligibility, or the accuracy of speech sounds, even when the evaluation of a specific, isolated component is matched to a speech sound group. For example, in the TDM perspective isolated evaluation of the tongue by means of oromotor, nonspeech gestures will not predict the correctness of tongue gestures for lingual consonants.

The TDM suggests several clinical predictions. For example, in a properly controlled experiment, therapeutic gains in speech production skills of persons with dysarthria should be greater among a group of individuals receiving speech production training, compared with persons trained to improve oromotor, nonverbal skills. An alternative version of this experiment is to compare therapeutic gains across two groups, one receiving oromotor, nonverbal training plus speech production training, the other speech production training plus training of a motor skill that does not involve oromotor structures, such as control of a measure of handgrip (e.g., target accuracy, or stability). The design of experiments such as these present challenges, many of which concern experimental control variables across groups such as comparability of severity, clinician competence, and stability of the underlying disease responsible for dysarthria. Assuming the experiment can be controlled, the concept of the latter experiment is probably closer to clinical reality than the former, which compares only oromotor, nonverbal training to speech production training.

Another prediction consistent with the TDM reverses the IM prediction that measures of oromotor, nonverbal control can be summed across subsystems to estimate the magnitude of a deficit in speech motor control. More precisely, estimates of overall speech motor control deficits, by speech intelligibility and/or severity scores, will not predict performance of oromotor, nonverbal control, summed across subsystems. A more fine-grained version of this prediction is that obstruent errors that differ in magnitude across place of articulation, will not predict the presence of differential weakness, displacement, speed, and so forth, in NSOM tasks that parallel the place-specific phonetic impairment.

3.2. Structure and Predictions from the IM and TDM: Summary

The IM and TDM have been discussed here because of their explicit application to theoretical and clinical understanding of motor speech disorders. Neither the IM nor TDM are true models; rather, both are a series of statements and assumptions that offer coarse-grained accounts of typical and impaired speech motor control in dysarthria (as well as in other speech deficits not covered here).

Selected experimental predictions derived by the current author from the structure and assumptions of the IM and TDM have been presented. Comparison of the predictions from the IM and TDM shows them to be in conflict. This is good for scientific and clinical communities, because the experimental evaluation of well-defined, opposing predictions should contribute to a qualitative, and eventually quantitative enhancement and refinement of either (or both of) the IM or TDM.

The IM and TDM have been described here as if separated by an impermeable conceptual wall. This is overly simplistic, but a starting point from which to determine experimentally if and how the two views can be melded (e.g., see ^[31]). Some formal models of speech motor control may represent execution goals in articulatory terms, such as vocal tract constriction location and degree, but the output of the model is an acoustic signal (Ref. ^[17], their Figure 1). Moreover, the speech acoustic signal is used to refine the shape, location, and degree of constriction—much as in Guenther’s ^[9] formal model. Nevertheless, a model with two (or more) simultaneous goals, even if not independent, may point the way to a hybrid model of speech motor disorders with IM and TDM principles.

References

Froeschels, E. Hygiene of the Voice. Arch. Otolaryngol. Head Neck Surg. 1943, 38, 122–130.
Froeschels, E. The Chewing Approach in Speech and Voice Therapy; Weiss, D.A., Beebe, H.H., Eds.; S. Karger: Basel, Switzerland, 1951.
Guttman, O. Gymnastics of the Voice for Song and Speech: Also a Method for the Cure of Stuttering and Stammering; Werner, E.S., Ed.; The Voice Press: Albany, NY, USA, 1893.
Froeschels, E. Selected Papers of Emil Froeschels; North-Holland Publishing Company: Amsterdam, The Netherlands, 1964.
Weismer, G. Assessment of Oromotor, Nonspeech Gestures in Speech-Language Pathology: A Critical Review. In Telerounds #35 (Videotape); University of Arizona: Tucson, AZ, USA, 1997.
Weismer, G. Philosophy of Research in Motor Speech Disorders. Clin. Linguist. Phon. 2006, 20, 315–349.
Ziegler, W.; Schölderle, T.; Brendel, B.; Risch, V.; Felber, S.; Ott, K.; Goldenberg, G.; Vogel, M.; Bötzel, K.; Zettl, L.; et al. Speech and Nonspeech Parameters in the Clinical Assessment of Dysarthria: A Dimensional Analysis. Brain Sci. 2023, 13, 113.
Maas, E. Speech and Nonspeech: What Are We Talking About? Int. J. Speech Lang. Pathol. 2017, 19, 345–359.
Guenther, F.H. Neural Control of Speech; The MIT Press: Cambridge, MA, USA, 2016; ISBN 9780262336987.
Perkell, J.S. Five Decades of Research in Speech Motor Control: What Have We Learned, and Where Should We Go from Here? J. Speech Lang. Hear. Res. 2013, 56, 1857–1874.
Hickok, G.; Poeppel, D. Dorsal and Ventral Streams: A Framework for Understanding Aspects of the Functional Anatomy of Language. Cognition 2004, 92, 67–99.
Bose, A.; Van Lieshout, P. Speech-like and Non-Speech Lip Kinematics and Coordination in Aphasia. Int. J. Lang. Commun. Disord. 2012, 47, 654–672.
Darley, F.; Aronson, A.; Brown, J. Motor Speech Disorders; W.B. Saunders: Philadelphia, PA, USA, 1975.
Duffy, J.R. Substrates, Differential Diagnosis, and Management. In Motor Speech Disorders, 4th ed.; Elsevier: St. Louis, MI, USA, 2020.
Staiger, A.; Schölderle, T.; Brendel, B.; Ziegler, W. Dissociating Oral Motor Capabilities: Evidence from Patients with Movement Disorders. Neuropsychologia 2017, 95, 40–53.
Lancheros, M.; Pernon, M.; Laganaro, M. Is There a Continuum between Speech and Other Oromotor Tasks? Evidence from Motor Speech Disorders. Aphasiology 2023, 37, 715–734.
Parrell, B.; Ramanarayanan, V.; Nagarajan, S.; Houde, J. The FACTS Model of Speech Motor Control: Fusing State Estimation and Task-Based Control. PLoS Comput. Biol. 2019, 15, e1007321.
Fitch, W.T. The Biology and Evolution of Speech: A Comparative Analysis. Annu. Rev. Linguist. 2018, 4, 255–279.
Samuel, A.G. Speech Perception. Annu. Rev. Psychol 2011, 62, 49–72.
Liberman, A.M.; Cooper, F.S.; Shankweiler, D.P.; Studdert-Kennedy, M. Perception of the Speech Code. Psychol. Rev. 1967, 74, 431–461.
Liberman, A.M.; Mattingly, I.G. The Motor Theory of Speech Perception Revised. Cognition 1985, 21, 1–36.
Gafos, A.; van Lieshout, P. Editorial: Models and Theories of Speech Production. Front. Psychol. 2020, 11, 1238.
MacKenzie, C.; Muir, M.; Allen, C. Non-Speech Oro-Motor Exercise Use in Acquired Dysarthria Management: Regimes and Rationales. Int. J. Lang. Commun. Disord. 2010, 45, 617–629.
Miller, N.; Bloch, S. A Survey of Speech–Language Therapy Provision for People with Post-Stroke Dysarthria in the UK. Int. J. Lang. Commun. Disord. 2017, 52, 800–815.
Dollaghan, C.A. The Handbook for Evidence-Based Practice in Communication Disorders; Paul H. Brookes Publishing Company: Baltimore, MD, USA, 2007.
Trinder, L.; Reynolds, S. Evidence Based Practice: A Critical Appraisal; Blackwell Publishing: Oxford, UK, 2000.
Thoma, A.; Eaves, F.F. A Brief History of Evidence-Based Medicine (EBM) and the Contributions of Dr David Sackett. Aesthet. Surg. J. 2015, 35, 261–263.
Ballard, K.J.; Robin, D.A.; Folkins, J.W. An Integrative Model of Speech Motor Control: A Response to Ziegler. Aphasiology 2003, 17, 37–48.
Ballard, K.J.; Granier, J.P.; Robin, D.A. Understanding the Nature of Apraxia of Speech: Theory, Analysis, and Treatment. Aphasiology 2000, 14, 969–995.
Tremblay, P.; Deschamps, I.; Gracco, V.L. Neurobiology of Speech Production: A Motor Control Perspective. In Neurobiology of Language; Hickok, G., Small, S., Eds.; Academic Press: San Diego, CA, USA, 2016; pp. 714–750.
Grimme, B.; Fuchs, S.; Perrier, P.; Schöner, G. Limb versus Speech Motor Control: A Conceptual Review. Mot. Control 2011, 15, 5–33.
Folkins, J.W.; Moon, J.B.; Luschei, E.S.; Robin, D.A.; Tye-Murray, N.; Moll, K.L. What Can Nonspeech Tasks Tell Us about Speech Motor Disabilities? J. Phon. 1995, 23, 139–147.
Gick, B.; Stavness, I. Modularizing Speech. Front. Psychol. 2013, 4, 977.
Barlow, S.M.; Abbs, J.H. Force Transducers for the Evaluation of Labial, Lingual, and Mandibular Motor Impairments. J. Speech Hear. Res. 1983, 26, 616–621.
Barlow, S.M.; Bradford, P.T. Measurement and Implications of Orofacial Muscle Performance in Speech Disorders. J. Hum. Muscle Perform. 1992, 1, 1–31.
Rong, P.; Yunusova, Y.; Wang, J.; Green, J.R. Predicting Early Bulbar Decline in Amyotrophic Lateral Sclerosis: A Speech Subsystem Approach. Behav. Neurol. 2015, 2015, 183027.
Hixon, T.; Weismer, G.; Hoit, J. Preclinical Speech Science: Anatomy, Physiology, Acoustics, and Perception, 3rd ed.; Plural Publishing: San Diego, CA, USA, 2020.
Moore, S.; Rong, P. Articulatory Underpinnings of Reduced Acoustic–Phonetic Contrasts in Individuals with Amyotrophic Lateral Sclerosis. Am. J. Speech Lang. Pathol. 2022, 31, 2022–2044.
Robin, D.; Solomon, N.; Moon, J.; Folkins, J. Nonspeech Assessment of the Speech Production Mechanism. In Clinical Management of Sensorimotor Speech Disorders; McNeil, M.R., Ed.; Thieme: New York, NY, USA, 1997; pp. 49–62.
Folkins, J.W.; Zimmermann, G.N. Jaw-Muscle Activity during Speech with the Mandible Fixed. J. Acoust. Soc. Am. 1981, 69, 1441–1445.
Dromey, C.; Richins, M.; Low, T. Kinematic and Acoustic Changes to Vowels and Diphthongs in Bite Block Speech. J. Speech Lang. Hear. Res. 2021, 64, 1794–1801.
Platt, L.J.; Andrews, G.; Howie, P.M. Dysarthria of Adult Cerebral Palsy. J. Speech Lang. Hear. Res. 1980, 23, 41–55.
Tu, M.; Wisler, A.; Berisha, V.; Liss, J.M. The Relationship between Perceptual Disturbances in Dysarthric Speech and Automatic Speech Recognition Performance. J. Acoust. Soc. Am. 2016, 140, EL416–EL422.
Kent, R.D. Nonspeech Oral Movements and Oral Motor Disorders: A Narrative Review. Am. J. Speech Lang. Pathol. 2015, 24, 763–789.
Schliesser, H.F. Alternate Motion Rates of the Speech Articulators in Adults with Cerebral Palsy. Folia Phoniatr. Logop. 1982, 34, 258–264.
Langmore, S.E.; Lehman, M.E. Physiologic Deficits in the Orofacial System Underlying Dysarthria in Amyotrophic Lateral Sclerosis. J. Speech Hear. Res. 1994, 37, 28–37.
Ziegler, W. To Speak or Not to Speak: Distinctions between Speech and Nonspeech Motor Control. Aphasiology 2003, 17, 99–105.
Lindblom, B. Role of Articulation in Speech Perception: Clues from Production. J. Acoust. Soc. Am. 1996, 99, 1683–1692.

© Text is available under the terms and conditions of the Creative Commons Attribution (CC BY) license; additional terms may apply. By using this site, you agree to the Terms and Conditions and Privacy Policy.

Upload a video for this entry

Information

Subjects: Others

Contributor MDPI registered users' name will be linked to their SciProfiles pages. To register with us, please refer to https://encyclopedia.pub/register :

Gary Weismer

View Times: 264

Update Date: 10 May 2023

Table of Contents

Video Upload Options

Confirm