The Cerebral Cortex and Stuttering, Part 2

Per Alm has recently published a chapter in a book focusing on cluttering. As he points out, since stuttering and cluttering are overlapping but contrasting disorders, understanding cluttering may help in understanding stuttering. A lot of the presentation below, representing a theoretical framework for the production of speech, owes much to this chapter. However, I take full credit for any misrepresentations that I may have made.

In this post, we focus on the medial cortical areas that are involved in the production of speech. Specifically, these are the motor cortical areas, which are typically divided into three regions having different functional roles:

• pre-motor area (PMA)
• supplementary motor area
• primary motor cortex(M1)

The premotor cortex guides movements of the vocal apparatus by integrating sensory information. It helps to regulate motion by dictating an optimal position to the motor cortex for any given movement.

The supplementary motor area lies just in front of the primary motor cortex. What had previously been regarded as the supplementary motor area is actually composed of two anatomically and functionally distinct parts: the supplementary motor area proper (called the SMA) and the pre-SMA (see figure*).

The Pre-SMA and the PMA function in processes that are preparatory to speech. The pre-SMA is involved in acquiring new motor sequences, being more active when the sequence is novel as opposed to one that had already been learned. Higher level planning and sequencing takes place in the anterior pre-SMA, while the posterior pre-SMA is involved in the sequencing of sounds and syllables.

The SMA, along with M1, functions in the initiation and execution of movement. The SMA is responsible for the coordination of sequences of motor movements such as those involved in speech. It is concerned with actions under internal control, for example the retrieval of a motor sequence from memory involving the pronunciation of a word.

SMA neurons are more active when the task requires the arrangement of multiple movements in the correct sequence and correct temporal order. Specifically, the timing of an articulation and the speech rate is controlled by the SMA with support from both the basal ganglia and the cerebellum. And it is this timing mechanism that has been implicated in stuttering.

The purpose of the primary motor cortex (M1) is to connect the brain to the lower motor neurons via the spinal cord.

Aside from the motor cortical areas mentioned above, the anterior cingulate cortex (ACC)** plays the role of a central executive in the production of speech The ACC (as well as the preSMA) are involved in assembling the phrase, from the selection of words and word forms to the sequencing of the words.

In addition, the ACC is involved in the high-level monitoring of speech errors (along with the SMA) through auditory connections. There seems to be a lack of such monitoring among clutterers who seem to be unaware of their disfluency as pointed out in Per Alm’s chapter.

Stutterers, on the other hand, may not have this problem. Instead they seem to be able to monitor their speech errors and are aware of their disfluency to such an extent that they ultimately regress into secondary stutttering patterns such blocking, etc. However for stutterers, the auditory aspects, feedback, and timing associated with this monitoring may be deficient as shown in a number brain imaging studies.

The ACC, the pre-SMA, and the SMA represent an assembly center for spontaneous speech that retrieves the linguistic components from the left lateral cortex regions such as Broaca’s and Wernicke’s areas. And as discussed in a previous post, the selection of a single word from a number of competing alternatives is further facilitated by the basal ganglia through a winner-take-all function.

Brain imaging studies have shown that stutterers exhibit deficiencies in the left cortical areas discussed above, while the right hemisphere tends to be overactive. However, the greater right hemisphere activity may merely reflect a compensatory mechanism due to the left hemisphere deficiencies. Similarly, it is unclear at this point as to whether: (1) the activity level deficits in these left cortical regions are a secondary effect of dysregulation of the basal ganglia circuits, (2) the left cortical activity level deficits cause dysregulation of the basal ganglia, or (3) deficiencies in both these areas of the brain simultaneously contribute to disfluency.

______________________________________
* This figure, taken from Per Alm’s chapter, represents the medial portion of the brain’s left hemisphere, i.e., split in half from front to back; the figure in the Part 1 post represents the exterior of the left hemisphere of the brain.

** More specifically, that part of the ACC identified as the cognitive ACC.

The Cerebral Cortex and Stuttering, Part 1

We have thus far focused heavily on the basal ganglia as a potential root cause of stuttering. In the next two posts, we discuss the possible role of structures in the cerebral cortex involving speech processes that may either be affected by or directly contribute to disfluency.


The functional location of speech is principally in the left hemisphere of the cerebral cortex for the great majority of right-handed people. For left-handed people, the picture is less clear; some show a specialization for speech in the left hemisphere, while others specialize in the right, and for still others, both hemispheres contribute just about equally. In the following we focus on the vast majority of individuals who specialize chiefly in the left hemisphere.

A number of brain imaging studies have shown that stutterers exhibit a deficient involvement of the left cortical hemisphere in speech activities and greater involvement of the right. An excess of testosterone in newborns due to stress at the time of birth might well be one of the most common causes of slower development in the left hemisphere resulting in greater participation by the right.

Located in the left lateral cortex regions, as shown in the figure below, Wernicke’s and Broca’s areas are involved in the initiation of speech.

There are three sub-areas within Wernicke’s area. The first responds to spoken words (including the individual’s own); the second responds only to words spoken by someone else (but is also activated when the individual recalls a list of words); and the third is associated with producing speech. Basically Wernicke’s area relates to the representation of phonetic sequences, regardless of whether the individual hears them, generates them himself, or recalls them from memory.

Broca’s area is another part of the complex network involved in developing an articulation plan. It is concerned with the meanings of words (i.e., semantics), how words are combined to form phrases and sentences (i.e., syntax), and the specific sounds associated with words (i.e., phonology).

 Broca's area is associated with the serialization of coordinated action of the speech organs.

Various injuries or deficiencies in Broca’s and Wernicke’s areas lead to aphasia which is the partial or total loss of the ability to articulate thoughts and ideas. For example, injuries to Broca’s area may result in agrammatism which typically involves a lack of use of syntax in speech production resulting in laboured speech. Individuals with Wernicke's aphasia have difficulty recalling correct words for the context or they may coin meaningless words.

Being articulate (as opposed to fluent)* depends upon the rich complex of information generated in lateral areas of the cortex, namely Wernicke’s and Broca’s areas. This information is then transferred to medial cortical areas adjacent to Broca’s area such as the motor cortex, which governs the mechanics of speaking. Current theories of stuttering do not implicate the lateral areas as root causes of disfluency. However, stuttering may affect being articulate by virtue of the interruptions of one’s train of thought and the fact that excessive energy and effort is devoted to the mechanics of speaking, which takes away from the focus on content (i.e., what you say).

In addition to the left lateral areas of the cortex, the right hemisphere also comes into play for speech. Prosody refers to the intonation and stress with which the basic units of a language (called phonemes) are pronounced. Feeling and attitude are conveyed through prosody as well as the melody, inflection, and intonation of one's voice and by varying the pitch, inflection, timbre, stress contours, and the rate/amplitude of speech. These aspects of speech enable a speaker to convey and a listener to determine intent, attitude, feeling and meaning. These capacities, both for the speaker (governing the style of presentation, i.e., how you say it) and the listener (interpreting the style), are predominantly mediated by the right half of the cerebral cortex.

Once again the effort/energy spent on stuttering, attempting to speak fluently, or using fluency shaping techniques impairs the ability to convey emotion through one’s speech. As a result, stutterers may speak and behave in a way that seems flat and emotionless.

In the next post, various aspects of the medial cortical areas as relates to speech will be discussed.

*************************************************
* While the aphasias discussed above are regarded by the medical establishment as fluency problems, I prefer to distinguish between being fluent and being articulate.

Subgroups and Drug Testing Trials

The existence of subgroups among stutterers may lead to the appearance of reduced efficacies for any therapeutic treatment under test.  For example, some stutterers may respond to dopamine blockers such as atypical antipsychotic drugs while others may respond to dopamine agonists such as amphetamines.  Including both of these subgroups within a drug testing trial will lead to reported results that are less robust.

In other areas of medical research, attempts have been made to identify individuals for inclusion in drug trials based on some measurable characteristic.  For example, researchers have identified certain genetic traits among breast cancer victims on the basis of which these individuals may have a greater chance of success in the trials.

With regard to stuttering, the level of knowledge has not yet reached the stage whereby  subgroups can be identified on the basis of genetic analyses.  However, there may be other means for identifying subgroups.

For example, in selecting individuals for a trial of a dopamine antagonist such as an atypical antipsychotic, a pretrial might be conducted to choose individuals on the basis of their responses to a benzodiazepine (BZ).  BZs effectively block dopamine by indirect means, so that an individual whose fluency improves on a BZ might be in the same subgroup responsive to an atypical antipsychotic.  The effect of the BZ on fluency should be apparent within half an hour after administration.  On the other hand, an individual whose fluency either deteriorates on BZs or remains the same might be in the amphetamine responsive group (or in a potentially third subgroup responsive to neither treatment protocol).

We can additionally refine this selection approach by further screening for individuals who may be particularly responsive to placebos.  During the pretrial, a subject might be given a BZ on one day and a placebo on another.  An individual who responds positively to both the BZ and the placebo would be eliminated as a trial subject.

The approach outlined above is by no means perfect.  BZs affect the limbic system (involving emotions) more strongly than atypical antipsychotics.  Since the limbic system (via the ventral striatum) may affect fluency (see the “An Anxious Mind Affects Stuttering” posts), the drug trial group chosen by the above approach may still contain individuals (i.e., false positives) who are not responsive to dopamine antagonists.

The pretrial/trial approach may give a more accurate measure of the effect of a drug on a particular subgroup.