2003 IFA Congress: Montreal, Canada

Kinematic Changes Following Static Perturbation in People Who Stutter

Aravind Kumar Namasivayam and Pascal H.H.M. Van Lieshout

Oral Dynamics Laboratory, Graduate Department of Speech-Language Pathology University of Toronto, 500 University Ave, Toronto, ON M5G 1 V7, Canada.


The purpose of this study was to determine whether or not persons who stutter (PWS) could compensate to the presence of a static (bite-block) perturbation. We hypothesized that, if PWS have a limited ability in utilizing sensory information then they would be unable to adequately control their lip movements to deal with the fixed jaw position. Results indicated that PWS like their matched controls were able to compensate for the presence of a static perturbation. However, there were differences between the two groups in terms of their use of motor control strategies during compensation.

  1. Introduction
In recent years, stuttering has been studied from a sensory-motor perspective. This perspective can be broadly divided into two opposing views. The first view states that PWS have an extra reliance on sensory feedback. Neilson and Neilson (1991) and Pindzola (1987) hypothesized that PWS may have difficulties in executing fast speech movements and hence use a more feedback driven motor control strategy (in comparison to normal speakers) wherein one movement is completed before the next one begins in such a way that the sensory feedback from the first movement is used to adjust forthcoming muscle commands for the next movement (see also Van Lieshout,'1995; Van Lieshout, et al., 1993). This type of strategy would typically lead to a stronger than normal temporal segregation (asynchrony) of individual movements that need to be coordinated for a given speech gesture. Support for the notion that PWS may use a different (more feedback driven) control strategy was provided by Van Lieshout et al. (1994), who reported consistent asynchronies for PWS in the timing of peak velocity of lips and jaw and in other studies by Van Lieshout (Van Lieshout et al., 1993; Van Lieshout, et al., 1996), where PWS were found to show a consistent delay in the increase of muscle activity.

This hypothesis of extra reliance on feedback is somewhat in contrast to the second view which states that PWS have limited abilities to process sensory information (Archibald & De Nil, 1999). This view is supported by some recent studies which have shown that PWS differ from people who do not stutter (PNS) in the acquisition and processing or use of sensory information (De Nil & Abbs, 1991; Archibald & De Nil, 1999). Such a problem may coexist with a potential difficulty in speech motor coordination (De Nil, 1995; Ward 1997). Evidence for a relationship between sensory feedback and movement coordination comes from limb control studies showing that sensory information is important for maintaining control and coordination of multi-joint or multi-limb movement (Ghez et al., 1990). Thus, it seems possible that a “faultyâ__ sensory system could underlie a difficulty in movement control and discoordination in PWS (Bauer et a1., 1995).

One way to study the underlying sensory-motor mechanisms in speech motor control is through the use of oral articulatory perturbation paradigms (Folkins & Canty, 1986; McFarland & Baum, 1995). For example, previous studies carried out on fluent speakers by Smith and Mclean-Muse (1987), Folkins and Canty (1986) and others (Folkins & Linville, 1983) have demonstrated a displacement-velocity scaling in the presence of bite-blocks where an increase in amplitude was correlated with an increasein peak velocity for labial closing movements such that the duration of movements remained constant. In this regard, if PWS have a limited sensory-motor system as some researchers have suggested (Archibald & De Nil, 1999) or use a more feedback driven motor control strategy (Neilson & Neilson, 1991; Pindzola, 1987; Van Lieshout et al., 1996) then we could expect their ability to compensate for perturbations to be compromised or at least to react to perturbations differently than fluent speakers who have intact sensory-motor systems.

The experiment reported here utilizes a static (bite-block) perturbation paradigm with two rate conditions (fast and normal) to test the claim that PWS unlike control speakers have limitations in using sensory information and/or use alternative strategies in movement control. If they have limitations in sensory processing, we expect them not to be able to compensate adequately to the presence of a bite-block regardless of rate. This would be evident in group differences in movement parameters (amplitude, peak velocity, and duration) and variability (higher). On the other hand, we hypothesize that if PWS have a greater dependency on afferent feedback for motor control, then they have two options in the combined presence of perturbation and a fast rate of speech. Either, they refuse to go faster because they want to stick to their feedback driven control strategy, which would make them slower in both normal and fast rate conditions. Or, they do speed up, but at that point it is difficult (if not impossible) to use the feedback control strategy and they will experience problems in controlling speech movements as indicated by an increase in movement variability.

  1. Methods


Five adults who stutter (PWS; mean age 26.12 years; standard deviation 9.14) and five adults who do not (PNS; mean age 25.2 years; standard deviation 4.27) formed the subject pool. PWS were rated for severity using the stuttering severity index (SSI: Riley 1972). All subjects were matched for age, sex and educational level and reported no problems in vision, hearing acuity, language, or voice quality other than stuttering prior to participating in this study.


Four pronounceable nonsense words ([bapi], [bipa], [bapiter] and [bipapter]) formed the test stimuli (adapted from the EMMA speech motor assessment protocol-Van Lieshout & Moussa, 2000). Each stimulus was repeated for 12 seconds at 2 self-paced speech rates (normal and fast, but intelligible).


The bite-blocks were custom made for each subject from a vinyl polysiloxane impression putty material (3M) and for all speakers an inter-incisor distance of about 12m was maintained.


The experiment was run on two separate days. On the first day two jaw-free baseline (JFl and JF2) recordings were carried out (not reported here). On the second day the subjects spoke in a jaw-free condition (J F3), immediately followed by the insertion of the bite-block between the 1” premolars (referred to as the bite-block [BB] condition). The JF3 and BB condition constitute the data for the experiment presented here. The bite-block condition was then followed by a 10-minute practice period (reading aloud with the bite-block in place) subsequent to which a post-practice (PBB) recording was carried out. The changes following practice with the bite-block in place will not be reported in this paper.

Data recording

Kinematic and acoustic signals were recorded simultaneously using an Electromagnetic Midsaggital Articulograph (EMMA) or AG100 (Carstens Medizinelektronik, GmbH). For this study, the receivers were placed on the upper lip, lower lip, jaw, tongue blade, tongue body and  tongue dorsum. This instrument tracks position of these receivers using an alternating magnetic field (for more details see Van Lieshout et al., 1994).


In the present study, for a given trial each stimulus was repeated for 12 seconds, which typically even at slow rates provides at least 8 repetitions for each target word. Kinematic data were derived for each movement cycle (= successive opening and closing movement) using custom made algorithms and an arithmetic mean and standard deviation (SD) were calculated for the repeated cycles of a target word (for details see Van Lieshout & Moussa, 2000). For this study, we calculated five dependent variables from the closing movements of upper lip (UL) and lower lip (LL): movement amplitude, peak velocity, duration, the percent of total movement time taken to reach peak velocity or TTP%, and a measure of cyclic spatio-temporal variability called the cST I (see Perkell et al., 2002, Van Lieshout et al., 2002; for more details on these measures). TTP% reflects speed-dependent asymmetries that may arise from changes in motor control strategy (Adams, et al., 1993; Bullock & Grossberg, 1988). The cSTI measure reflects variability in successive cyclic movement patterns, similar to the more general STI measure developed by Smith (Smith et al., 1995). The means and SD’s for all the words were tabulated for each subject and a mixed analysis of variance was carried out for the above-mentioned dependent variables with GROUP as between-subject factor and with repeated measures on the within-subject factors CONDITION (IF and BB) and RATE (normal and fast). Separate ANOVAS were carried out for the five dependent variables and their within-trial SD. In addition, we performed a Spearman correlation analysis on the dependent variables within each subject to study their relationships.

  1. Results and Discussion
We found main GROUP effects for the mean (F (1,8) = 12.22, p: 0.008) and standard deviation (F(1,8) = 5.50, p = 0.047) of UL amplitude, UL mean peak velocity (F (1,8) = 9.63, p = 0.014), LL CSTI (F (1,8) = 18.24, p: 0.002) and the standard deviation of LL TTP% (F (1,8)=11.83, p: 0.008). In general, these results indicate that PWS were producing larger and faster upper lip movements and more variable lower lip movements compared to normal speakers. We did not find a significant GROUP X CONDITION interaction effect for either lip, but there was a trend for higher UL peak velocity values in PWS in the bite-block condition compared to normal speakers (F (1,8) = 3.59, p=0.094). GROUP X RATE interactions were found for the mean (F ( 1,8) = 9.59, p=0.014) and standard deviation of UL duration (F(1,8) = 6.12, p=0.038), the standard deviation of UL TTP% (F (1,8) = 9.49, p=0.015), and the standard deviation of LL duration (F (1,8) = 9.00, p=0.017). In general, with respect to movement duration these interactions indicated that PWS were slower and more variable in their normal rate condition compared to control speakers, but they showed a clear decrease in duration and at the same time became less variable when speaking faster. PN S subjects did not alter duration (or its variability) as a function of rate to the same extent as PWS (Figure 1.a & b). These results imply that speaking rate affects articulatory kinematics and its variability differently in PWS and PN S. The lack of a significant GROUP X CONDITION interaction for any of the variables indicates that both PWS and PNS subjects showed similar compensation patterns in the presence of a bite-block perturbation. However, there was a trend for (subtle) group differences as PWS tended to use slightly faster upper lip movements during the bite-block condition compared to PNS.


Figure 1. (a & b): Depicts significant GROUP x RATE interaction for mean duration (in Msec) of upper lip movements (Figure 1.a) in people who stutter (PWS) and fluent speakers (PNS). A similar (but non-significant) trend was found for the lower lip (Figure 1.b) movements.

Several other interesting kinematic compensatory changes were observed for speakers in both groups in the presence of a bite-block. Our results differed from findings reported in the literature on kinematics of speaking in the presence of a bite-block. For example, in the study by Folkins and Canty (1986) a displacement-velocity scaling was found, where an increase in amplitude was correlated with a proportional increase in peak velocity for labial closing movements. This resulted in an invariant duration of lip closing gestures across jaw-free and bite-block conditions. In the present study, we also found an significant increase for both lips in amplitude (LL: F(l,8) = 10.18, p: 0.012; UL: F(l,8) = 22.06, p: 0.001) and peak velocity (LL: F(1 ,8) = 16.19, p= 0.003; UL: F(l,8) =55.69, p< 0.001) in the bite-block condition compared to the jaw-free situation. The correlation between these two measures in general was found to be significant (8/ 10 and 7/10 subjects showed a strong (r > .70) correlation for UL and LL respectively). However, movement duration actually decreased for both lips (LL: F (1,8) = 15.96, p=0.003; UL: F (1,8) = 23.22, p=0.001).

Furthermore, the presence of a bite-block induced a significant increase in movement cycle symmetry (i.e., near equal acceleration and deceleration times: TTP% = 52.23%) for both PWS and PNS in the lower lip data (F(l,8) = 10.81; p = 0.011). These kinematic changes correlate well with the findings reported in the literature on the relationship between movement rate and velocity profile function curves (Munhall, Ostry, & Parush, 1985; Adams, et al., 1993). According to Adams et al. (1993), one possible explanation for the change in velocity profile symmetry across normal and fast rates of speech is that the motor control system might have shifted from a strategy that is predominantly closed-loop at normal movement speeds to a more open-loop at faster movement speeds (Bullock & Grossberg, 1988). In this line of interpretation, the insertion of the bite-block could have created a similar shift towards open-loop control, at least in terms of the larger and faster movements that were found for this condition.

In terms of movement variability, it is important to note that the presence of a bite-block increased the variability of movement amplitude (trend in LL: F(l,8) = 4.81, p = .059; significant for UL: F(1,8)= 11.99, p: 0.008) and peak velocity (LL: F(l,8) = 8.81, p = 0.017; UL: F(l,8) = 27.77, p<0.00l). However, at the same time a significant decrease in variability was found for both lips in movement duration (LL: F(l,8) = 18.75, p = 0.002; UL: F(l,8) = 12.12,p = 0.008) and TTP% (LL: Section 6. Neurological and Speech Motor Basis of Stuttering 335

F(1,8) = 6.58, p = 0033; UL: F(l,8) = 6.87, p = 0.030). This decrease in variability for duration and TTP% was paralleled by a significant decrease in the CSTI measure for UL ‘F(l,8) = 15.21, p = 0.004), but not lower lip. In other words, the bite-block forced people to make larger and faster movements, which for UL became more consistent in their spatio-temporal pattern (see Figure 2. (a & b). Figure 2.b also shows how the decrease in cSTI in LL is less clear for PWS compared to PN S, which may explain the lack of an overall LL CONDITION effect for CSTI.


Figure 2. (a & b): Depicts cyclic spatio-temporal index (CSTI) of upper (Figure 2.51) and lower lip (Figure 2.b) movements in people who stutter (PWS) artdfluertt speakers ( PNS ) for jaw free and bite-block conditions.

The main effects described above were further qualified by RATE X CONDITION interactions for movement variability. The decrease in UL cSTI in the presence of a bite-block reported above was less clear for the fast rate than the normal rate (F (1,8) = 7.51, p = 0.025). A similar effect was seen for the variability of UL amplitude (F(1,8) = 5.98, p = 0.040). Variability of UL duration in the bite-block condition was not affected by rate, but in the jaw-free condition it was greater at normal rate compared to fast rate. This interaction was also significant (F(1,8) = 12.62, p = 0.007). Overall, the fast speaking rate was characterized by significantly shorter movement duration and less variable articulatory movements in comparison to normal speaking rate in both groups.

Regarding the time scale of compensation to the bite-block, it was found that for all speakers changes in movement amplitude were observed right from the first repetition of the first (bite-block) trial. Similar findings have been reported for vowel formant frequencies under bite-block conditions by Kelso and Tuller (1983). To illustrate this further, a plot of the change in movement amplitude over time for the first trial subsec uent to the insertion of the bite-block is shown in Figure 3. It can be clearly seen that the compensatory changes occur right fromthe start of the first (bite-block) trial.

  1. Conclusions
Our findings indicate that PWS were producing larger and faster upper lip movements and more variable lower lip movements compared to normal speakers. In terms of bite-block compensation, individuals of both groups showed similar kinematic patterns regardless of rate, contrary to any of the hypotheses we formulated at the end of the introduction to this paper. However, there were subtle differences in the manner of compensation as PWS overall showed (a trend) for faster (and also larger) upper lip movements in the bite-block condition. PWS showed slower and more variable movements only for their preferred (normal) speech rate. In the fast rate condition, movement characteristics of both groups became more similar. Main effects for the bite-block indicated larger movement amplitudes and peak velocities, as well as shorter durations across speakers.


Figure 3. Depicts amplitude of upper lip movements (in mm ) in a PWS for repeated utterances (within a single trial) of /bapi/ under jawfree (JF) and bite-block (BB) conditions. Compensatory changes in the upper lip in terms of increased amplitude are evident even in the first repetition of the first trial with the bite-block in place.

It has been suggested that peak movement velocity is an estimate of articulatory effort based on a cost optimization analysis of a dynamical model of single-axis frictionless movements (Perkell et al., 2002). Based on these results, it can be argued then that the presence of a bite-block in general induced more effortful movements in the upper lip. (especially for PWS) and to a lesser extent in the lower lip. Interestingly, upper lip movements in the bite-block condition were also found to be more stable in terms of cyclic movement patterns. Rate influenced mainly movement durations, as expected. However, it was interesting to notice that PWS indeed showed longer curations than control speakers during the normal rate conditions, but were able to shorten duration for fast rates. This would fit the “compensatory feedback control strategyâ__ claim mentioned in the introduction, except for the fact that their movements at fast rates become more instead of less stable. Further analysis showed that compensatory changes in movement amplitude occurred rapidly (at the start of the first perturbed trial) in both groups.

The observed increase in stability could be a direct consequence of either the increase in movement amplitude or the decrease in duration, at least from a coordination dynamics perspective (Peper & Beck, 1998; Van Lieshout, Hulstijn, & Peters, in press). Given the observation that in the fast rate condition the stability gain for the bite-block was reduced, implies that the overall increase in stability for the bite-block is (non-linearly) coupled to movement amplitude (e.g., Van Lieshout et al., 2002). The finding that the lower lip showed less of an effect on movement amplitude in the bite-block condition and a smaller increase in movement stability (for CSTI) seems to corroborate this assumption. It is to be noted that the lower cSTI values in the upper lip unlike the lower lip are less dependent on biomechanical constraints imposed by the bite-block and show a more direct consequence of compensatory changes in the system.

To summarize, these results suggest that PWS can adequately compensate for the presence of bite-block perturbations in terms of appropriate kinematic changes and that there may not be any gross sensory-motor limitations in their speech production system. However, the presence of more effortful articulation in PWS along with other subtle differences in the presence of perturbation suggests a less efficient underlying control system, in comparison to fluent control speakers. The lack of a significant group difference in bite block compensation could also be taken as support for claims that sensory information plays no or only a very limited role in this type of studies (e.g., Saltzman & Kelso, 1987). For example, according to the virtual target theory (Lofqvist & Gracco, 1997) compensation can be achieved without the need for any afferent sensory information if the typical virtual target of lower lip movements would fall within the extended movement range induced by the bite block. In the most basic interpretation of this theory, lower lip movements would just continue a bit longer before they reach the upper lip, provided peak velocity remains the same (using the same motor program). However, that is not what is seen in our data, where movement duration actually decreased and peak velocity values increased for both lips in the presence of a bite block. The Task Dynamics model with its built-in compensatory mechanisms as part of a gestural configuration would be able to explain such changes, even in the absence of feedback (Saltzman & Kelso, 1987). However, how both theories would explain the differential response across lips for the two groups (PWS having a stronger UL response to perturbation) remains unclear.

Adams, S. G., Weismer, G., & Kent, R. D. (1993). Speaking rate and speech movement velocity profiles. Journal of Speech and Hearing Research, 36, 41-54.

Archibald, L. & De Nil, L. F. (1999). The relationship between stuttering severity and kinesthetic acuity for jaw movements in adults who stutter. Journal of Fluency Disorders, 24, 25-42.

Bauer, A., Jancke, L., & Kalveram, K. (1995). Mechanical perturbation of the jaw during stutterers’ and nonstutterers’ fluent speech. In H. F. M . Peters, C. W. Starkweather, & Bosshardt (Eds), Proceedings of the First World Congress on Fluency Disorders (p. 31-34).

Bullock, D., & Grossberg, S. (1988). Neural dynamics of planned arm movements: Emergent invariants and speed-accuracy properties during trajectory formation. Psychological Review, 95, 49-90.

De Nil, L. F. (1995). The role of oral sensory feedback in the coordination of articulatory movements in adults who stutter: A hypothesis. In H. F. M. Peters, C. W. Starkweather, &

De Nil, L. F., & Abbs, J. H. (1991). Kinesthetic acuity of stutterers and non-stutterers for oral and non-oral movements. Brain, 114, 2145-2158.

Folkins, J. W., & Canty, J. L. (1986). Movements of upper and lower lips during speech: Interactions between lips with the jaw fixed at different positions. Journal of Speech and Hearing Research, 29, 348-356.

Folkins, J. W., & Linville, R. N. (1983). The effects of varying lower-lip displacement on upper-lip movements: Implications for the coordination of speech movements. Journal of Speech and Hearing Research, 26, 209-217

Ghez, C., Gordon, J ., Ghilardi, M. F., Christakos, C. N., & Cooper, S. E. (1990). Roles of proprioceptive input in the programming of arm trajectories. Cold Spring Harbor Symposia on Quantitative Biology, 55, 837-847. I 338 Theory, research and therapy in fluency disorders

Kelso, J. A. S., & Tuller, B.(l983). “Compensatory articulation under conditions of reduced afferent information: A dynamic formulation. Journal of Speech and Hearing Research, 26, 217-224.

Lofqvist & Gracco, V. L. (1997). Lip and jaw kinematics in bilabial stop consonant production. Journal of Speech, Language, and Hearing Research, 40, 877-893.

McFarland, D.H., and Baum, S. R. (1995). Incomplete compensation to articulatory perturbation.

Journal of the Acoustical Society of America, 97 (3), 1865-1873.

Munhall, K. G., Ostry, D. J., & Parush, A. (1985). Characteristics of velocity profiles of speech movements. Journal of Experimental Psychology: Human Perception and Performance, 11, 457-474.

Neilson, M. D., & Neilson, P. D. (1991). Adaptive theory of speech motor control and stuttering. In H. F. M. Peters, W. Hulstijn, & Starkweather (Eds.), Speech motor control and stuttering (pp.l49-157). Amsterdam: Elsevier Science Publishers.

Peper, C. E., & Beek, P. J . (1998). Are frequency-induced transitions in rhythmic coordination mediated by a drop in amplitude? Biological Cybernetics, 79, 291-300.

Perkell, J . S., Zandipour, M., Matthies, M., & Lane, H. (2002). Economy of effort in different speaking conditions. I. A preliminary study of intersubject differences and modeling issues. Journal of the Acoustical Society of America, 112, 1627- 1641.

Pindzola, R. H. (1987). Durational characteristics of the fluent speech of stutterers and nonstutterers. Folia Phoniatrica, 39, 90-97.

Riley, G. D. (1972). A stuttering severity instrument for children and adults. Journal of Speech and Hearing Disorders, 37, 314-322.

Schmidt, R. A. (1988). Motor control and learning. Chapaign, Illnois, Human kinetics publishers, Inc.

Saltzman, E. & Kelso, J. A. (1987). Skilled actions: a task-dynamic approach. Psychological Review, 94, 84-106.

Smith, A., Goffman, L., Zelaznik, H.N., Ying, G.S., & McGillem, C. (1995). Spatiotemporal stability and patterning of speech movement sequences. Experimental Brain Research, 104, 493-501.

Smith, B. L., & McLean-Muse, A. (1987). Effects of rate and bite block manipulations on kinematic characteristics of children’s speech. Journal of Acoustical Society ofAmerica, 81, 747-754.

Van Lieshout, P. H. H. M. (1995). Motor planning and articulation in fluent speech ofstutterers and nonstutterers. Unpublished doctoral dissertation, University of Nijmegen.

Van Lieshout, P. H. H. M., & Moussa, W. (2000). The assessment of speech motor behavior using electromagnetic articulography. The Phonetician, 8] , 9-22.

Van Lieshout, P. H. H. M., Hulstijn, W., & Peters H. F. M (1996). 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 Lieshout, P. H. H. M., Hulstijn, W., & Peters H. F. M (in press). Searching for the weak link in the speech production chain of people who stutter: A motor skill approach. In B. Maassen, R. Kent, H. F. M. Peters, P. H. H. M. van Lieshout & W. Hulstijn (eds.), Speech motor control in normal and disordered speech. Oxford, UK: Oxford University Press.

Van Lieshout, P.H.H.M., Rutjens, C.A.W., & Spauwen, P.H.M. (2002). The dynamics of interlip coupling in speakers with a repaired unilateral cleft-lip history. Journal of Speech, Language, and Hearing Research, 45, 5-19.

Van Lieshout, P. H. H. M., Alfonso, P. J., Hulstijn, W., & Peters, H. F. M. (1994). Electromagnetic Midsaggital Articulography (EMMA). In F. J . Maarse, A. E. Akkerman, A. N. Brand, L. J . M. Muddlor & van der Stelt (Eds), Computers in psychology: applications, methods and instrumentation (pp.62-76). Lisse: Swets and Zeitlinger.

Van Lieshout, P. H. H. M., Peters, H. F. M., Starkweather, C. W., & Hulstijn, W. (1993). Physiological differences between stutterers and nonstutterers in perceptually fluent speech: EMG amplitude and duration. Journal of Speech and Hearing Research, 36, 55-63.

Ward, D. (1997). Intrinsic and extrinsic timing in stutterer’s speech: Data and implications. Language and Speech, 40, 289-310.


In preparation for the 2018 World Congress the IFA is implementing Japanese translations of some pages on the site. Choosing Japanese below to see these translations.

Not all pages are translated, but you can use Google translate to see a machine translation using the switch below

Google Translate

Follow the Joint World Congress