What is Stuttering?

A Close Look at the Actual Speech Disruptions

By Per A. Alm, Ph.D.
Uppsala University, Sweden
per.alm@uu.se

When following the current debates on stuttering, it seems clear that there is no general agreement about what stuttering actually ‘is’—no agreement at all. Is stuttering, at its core, an active inhibition of attempts to talk, based on cognitive anticipation and negative emotions? Is stuttering the result of unstable steering of the speech movements, similar to an unstable driver of a car, finally ending up in the ditch? Or, should it not be considered as a disorder, but instead as a natural variation of human communication?

Stuttering is more than speech—but also speech

“Stuttering is more than speech—it’s an experience shaped by emotions, identity, and social interactions.” Yes, absolutely! For many persons who stutter, these additional aspects may have grown to become the largest part, below the surface, as depicted in the traditional iceberg model of stuttering. However, when children begin to stutter, they are typically no more shy or anxious than other children (based on strong research data). Fundamentally, stuttering is diagnosed based on speech disruptions with certain characteristics. Stuttering is not inherently an iceberg; it rather starts off as a beach ball. Still, the central role of speech in our lives implies that such involuntary speech disruptions easily affect our social interaction, our emotions, and our self-esteem. In summary, while the problems of stuttering certainly are not limited to the actual speech motor disruptions, it is important not to forget that the speech motor disruptions are also a part of this condition—and there are reasons to argue that they are the most fundamental aspect, from which the other difficulties arise. It is essential to highlight the situation for people with hidden/covert stuttering as well as the situation for those who have severe problems of communication due to very frequent overt stuttering—that cannot be hidden. “One size does not fit all” is a motto to always keep in mind.

Understanding the disruptions in detail

Despite the apparent centrality of the speech motor disruptions for the stuttering condition, relatively little research has focused on this aspect since the 1990s. Considering that the speech motor disruptions constitute the basis for a diagnosis, it is relevant to compare stuttering with established disorders of movement, whether involving speech or not. In neurology, the precise characterization of anomalous movements—known as motor phenomenology—is the foundation for understanding movement conditions. This is critical for accurate diagnosis, subtyping, and for generating hypotheses about underlying mechanisms and causes. In speech disorders such as stuttering, the motor characterization of the speech disruptions is particularly challenging for three reasons: (1) the relevant movements are largely hidden from view; (2) speech production involves a complex system of moving parts at multiple levels, with continuous interactions; and (3) the movements occur as a very rapid sequence.

Research project in Uppsala, Sweden

Stuttering events are typically categorized as repetitions, prolongations and blocks, but this is a labelling based on the sound, providing little information about what actually happens in the speech apparatus. For example, repetitions may occur from a multitude of causes. At Uppsala University, Sweden, we have been working with these issues since 2014, with development of methods and pilot studies for synchronized recording and analysis of stuttering events. This includes surface electromyography (EMG) of muscles related to the lips and on the tongue, endoscopic video of the vocal folds, electroglottography of phonation, formant analysis indicating tongue movements, and video of the face. Our qualitative and exploratory analysis has focused on in-depth understanding of specific instances of stuttering: What are the first indications that something goes wrong? What may be signs of reactions? In repetitions, the final completed attempt can be used as a template to compare with the disrupted attempts—what are the differences that might have resulted in the disruption and the restart? Overall, the analysis is an attempt to reconstruct the events in as much detail as possible.

Figure: Example showing a part-word repetition of the Swedish word ‘skuret.' The blue marker (solid line) in the lower right shows the normal activity of a muscle opening of the lips after the /ū/ sound.The dashed red marker indicates the point of missing DLI activation in the disrupted attempt.

Results and proposals from Uppsala

A summary and overview of these initial studies in Uppsala was recently published (Alm et al., 2025, freely available). While these studies should be considered as preliminary, with a need for consolidation by further studies, they do provide information that may elucidate the nature of stuttering as a phenomenon. A large part of the article consisted of a comparison of the characteristics of stuttering (the phenomenology) with established movement disorders with characteristics that overlap with stuttering. This comparison also included the subjective experiences. For this comparison, we focused on two disorders: dystonia and motor block disorders. Dystonia is a condition characterized by dysregulated muscle tone, causing tense involuntary contractions. Motor block disorders typically affect automatized sequential movements: The movements are suddenly ‘stuck’, with a temporary inability to move forward to the next movement in the sequence. The most established motor block disorder is freezing of gait in Parkinson’s disease. In freezing of gait, the experience is typically being described as “the feet being glued to the floor.” Of particular interest is that both dystonia and motor block disorders often show variability and task specificity—the risk for symptoms depends on the type of movement, the context, the level of stress and time pressure, etc. Furthermore, both disorders are related to the basal ganglia, a structure central for automatized movements and being implicated also in stuttering.

In summary, the analysis of our data and the comparison with known movement disorders resulted in the following proposals:

1. Stuttering may be viewed as a “motor block disorder.” The subjective and objective characteristics of stuttering largely align with the characteristics of motor block disorders. This suggests that the core aspect of stuttering may be a transient inability to activate the next speech movement—getting stuck in the present position, or stopping and restarting.

2. Reactions, tension, tremor, and physical concomitants. Fixed postures (motor blocks) tend to result in volitional attempts to break the posture and move forward. Such attempts (reactions) often result in elevated static muscular tension, that may elicit oscillatory (pulsating) muscular activity, which also can be described as tremor. The typical tremor frequency is about 8 Hz, but also multiples may occur, such as 16 Hz. Tremor can often be observed as silent ‘vibrations’ during fixed positions, but can also result in a rapid opening and closing of the airway, which is manifested as a fast repetition of sounds (about 7 to 9 Hz). We suggest that the tremor results from a normally functioning cerebellum trying to compensate for the difference between the motor commands from the cortex and the actual position of the articulators—but overcompensates in both directions. Tremor may, in some instances, break a fixed position and allow continued speech. The studies suggest that widespread tremor constitutes a large part of the physical concomitants in stuttering.

3. Levels of muscular tension. The level of muscular tension in articulatory muscles is predominantly within the normal range for speech. Very high levels of tension tend to be associated with tremor.

4. Immediate mechanisms of stuttered disruptions. An immediate mechanism is here defined as the last step in the chain of events resulting in a stuttered speech disruption. These immediate mechanisms occur in the mechanical speech apparatus. A range of different immediate mechanisms for the disruptions have been observed, for example the following proposals:

a) Repetitions for the repair of missing muscular activation. A frequent pattern in part-word repetitions is that speech is disrupted because one or several required speech muscles are not activated when needed, or activated too weakly. The word is restarted until all required muscles are activated.

b) Fast sound repetitions. Fast sound repetitions, at about 7 to 9 Hz (or an interval of about 0.125 seconds) are likely to be the audible manifestation of a tremor opening and closing the airway.

c) Silent pauses from laryngeal blocks. Blocks of the airflow at the laryngeal level typically result from complete constriction of the airway at the level of the vestibular folds, just above the vocal folds.

d) Silent pauses from articulatory blockage of airflow. The airflow can also be blocked by sustained closure of the lips or sustained pressing of the tongue against the roof of the mouth.

e) Silent pauses from laryngeal openings. Silent ‘blocks’ frequently result from involuntary opening (abduction) of the vocal folds, precluding phonation. These “laryngeal openings” are characterized by the loss of air, but typically without physical concomitants—presumably because the person cannot react by building up air pressure.

f) Voiceless prolongations. Laryngeal openings can also result in prolongation of voiceless consonants before a vowel, e.g./ssssssssssolemn/. In this case the voiceless consonant is prolonged until the vocal folds are in the position for phonation.

g) Prolongations with articulatory muscles stuck in a fixed position. However, most prolongations (and probably all prolongations of voiced sounds) are the result of articulatory muscles being stuck in a fixed position. Typical examples are sustained contraction of the orbicularis oris lip muscle, closing the lips and resulting in a prolongation of /mmmm/, sometimes even when lip closure is not intended. Another example is prolongation of /L/ as a result of sustained contraction of tongue muscles.

h) Slow sound repetitions. Brief bursts of sounds, voiced or voiceless, can occur as a result of sustained contraction of articulatory muscles blocking the airflow, resulting in silent pauses. The bursts of sounds occur when the contraction is temporarily reduced, so that the air pressure from the lungs makes air briefly pass the obstruction.

i) Repetitions as a result of ‘blocks.’ Repetitions of longer sequences tend to occur as a response to some type of interruption, at some level, hindering the continuation.

About the current terminology

It is clear from the list above that the traditional categorization of stuttering events as repetitions, prolongations, and blocks provides limited information about the actual nature of the speech disruption. In particular, the word ‘blocks’ is ambiguous. In the current literature it appears that the term block is typically used to signify a blockage of the airflow, even though it also seems to be used as an umbrella term for silent pauses or for stuttered disruptions in general. However, in the textbook from Charles Van Riper in 1939, he described a block as a temporary inability to move the speech musculature: “the stutterer finds himself unable to move a certain speech structure when it is necessary for him to do so” (p. 121). Interestingly, this is an exact description of the symptoms of a motor block disorder. To reduce this ambiguity, we suggest using the term ‘block’ with a specification. For example, a “laryngeal block” would signify a hard closure of airway at the laryngeal level, while a “motor block” would refer to the temporary inability to move forward in the motor sequence.

Implications for brain theories of stuttering

Active inhibition theories. When considering stuttering in relation to various neuroscience theories, it is important to consider to what extent the observed motor symptoms correspond to what can be expected from the theories. One influential set of theories postulates that active inhibition of speech is a central aspect of stuttering—possibly as a result of negative expectations or anxiety. This idea has been linked to a general model by the psychologist Adam Aron, in which the command center for the inhibition of action (the “brake pedal”) is located within the right hemisphere frontal cortex, sending stop-commands to the basal ganglia. It is quite possible that this or similar mechanisms contribute to the situational variability of stuttering, but as a main explanation there are two issues to consider. First, the neural mechanism outlined by Aron is assumed to inhibit actions as a whole, not specific muscles. In contrast, it was frequently observed that specific muscles did not get activated while phonation and other muscles were unaffected until the speech attempt was interrupted and restarted a little later. Second, it is difficult to see how a cortical right hemisphere mechanism of active inhibition would result in the experience of prolonged loss of motor control, for several seconds. This phenomenon rather points towards the basal ganglia as the center for the disruption.

Motor control theories

Another set of influential theories are based on the assumption that people who stutter have an imprecise system for speech motor control and/or an inefficient feedback control system. This is assumed to somehow result in errors of articulation and feedback corrections that lead to speech disruptions. There are many different formulations on this theme. However, to the best of our understanding, the observations from the studies seem to differ from what would be expected from motor control theories of stuttering. First, if the control system is imprecise, we would expect to see articulatory errors in the fluent speech before instances of stuttering, which we typically do not. Second, we often observed sustained static contractions of articulatory muscles during stuttering. This is contrary to the theory, which predicts instability. Third, the observed motor errors in stuttering are more likely to be ”all-or-nothing” than gradual, which rather points towards some gating mechanism than a correction mechanism. Fourth, motor control models have difficulties explaining blocks at the beginning of utterances, when there is no feedback or no errors to correct.

Proposal of brain mechanism

As discussed above, the subjective and objective phenomenology of stuttering align well with the phenomenology of motor block disorders such as freezing of gait. It is therefore reasonable to consider the proposed neurological mechanisms of freezing of gait also in relation to stuttering. One proposal that stands out for having strong explanatory power and being based on empirical evidence has been presented by Shine et al., (2013). In short, the core of this proposal is that the motor block occurs when there is a transient functional decoupling between the cortical networks and the basal ganglia. Normally there is continuous and reciprocal exchange of signals between these two levels. The communication is based on a common rhythm of oscillations within the cortex and the basal ganglia. One possible source of functional decoupling might be that the cortical level and the basal ganglia level somehow get out of sync, resulting in a temporary loss of communication and stalling of the motor system.

Conclusions

Based on these initial studies it is our view that in order to understand stuttering we need to consider the actual nature of the speech disruptions in more detail. The main proposition is that stuttering, fundamentally, can be considered to be a motor block disorder, with the activation of the next motor program being the core problem. Reactions tend to increase the tension and may result in tremor, which is proposed to constitute a large part of the physical concomitants. A multitude of different immediate mechanisms disrupt speech in stuttering, at all possible levels of the speech apparatus.

References

Alm PA, Brösemyr T, Grinde S, Johansson S, Karlsson J, Nordgren G, Olsson R, Rocksten F, Sandsten M, Sör I and White D (2025) Stuttering: the nature of the speech disruptions—a multimodal study of articulation and phonation. Frontiers in Human Neuroscience 19:1623308. doi: 10.3389/fnhum.2025.1623308

Shine, J. M. et al. (2013). Freezing of gait in Parkinson’s disease is associated with functional decoupling between the cognitive control network and the basal ganglia. Brain 136, 3671–3681.

From the Fall 2025 Magazine