Friday, June 03, 2005

It's (not) the basal ganglia, stupid!

Brown, Ingham, Ingham, Laird and Fox have commented on Alm's work: "[He] proposed a model in which the core dysfunction of stuttering was suggested to be an “impaired ability of the basal ganglia to produce timing cues for the initiation” of speech motor activity (p. 325). Unfortunately, this proposal provided no predictions about whether particular nuclei/circuits of the basal ganglia would be over- or underactivated during stuttering. [Our] meta-analysis data did not provide strong indications either favoring or opposing this model." They then go over various activitation signals in various experiments, and discuss its possible relevance to a dysfunction in the basal ganglia. But conclude that "Although the basal ganglia may certainly be playing a contributing role in stuttering, this role is in need of elucidation in future studies."

Doesn't look promising for PDS as they don't agree?? ON THE CONTRARY, I think this is GREAT progress in PDS research. At least to me! :-) Their (constructive) discussion shows that PDS research has moved closer to the core of PDS and research issues can FINALLY be decided by inconsistency or experimental findings, and not on authority or intuition. They are able to ask concrete and by-experiment answerable questions. They are able to do this, because brain imaging is delivering them DIFFERENCES between PDS and control subjects. So you can actually construct a theory and decide whether it is correct or incorrect. I find this very exciting.

So what's my view on the basal ganglia theory? Is it right or not? I have to admit, that I haven't YET read word-by-word Alm's theory and their article. So I might say something factually wrong. But, what I find very striking is that both works by Alm and Brown & InghamS et al. seems complementary. Brown & InghamS might have a theory (using partially efference copies) to understand brain activation patterns, but they remain silent of why and what it is that causes "failure to properly initiate the motor plan". And that's exactly the end product of Alm's work: "Stuttering is suggested to be caused by a disturbance of the medial system, in most cases in the basal ganglia. The core dysfunction is proposed to be impaired "go-signals" from the medial system, supposed to trigger the next motor segment in speech". BUT I am not sure he can explain why there should be brain anomalies as found by Sommer et al. , Foundas et al. , and Jancke et al. . And experimentally it seems to be on weak grounds, as Brown & InghamS say, "[Our] meta-analysis data did not provide strong indications either favoring or opposing this model." So it seems strong conceptually but no clear experimental signal, which should have been found by now??

So my pinch-of-salt is: The basal ganglia is involved but there is no dysfunction. To understand this, we need to talk about postmen! :-) Imagine two postmen are responsible for delivering your mail. But you only see one of them delivering mail to you. So you blame the other one for not delivering mail. But he might be a hardworking postman, but unfortunately he is the postman that your dog doesn't like! So both postmen are "functional", but it's your dog who makes the selection. The same might happen for the basal ganglia. It might work perfectly well and projects time signals into speech areas perfectly well, but that the area is not integrating them well. And this might be the anomaly area found. Paradoxically, most components of the basal ganglia theory would survive neatly like explaining fluency inducing effects i.e. the different system projects to a properly working area that integrates the signals well. Interestingly, factors that impact the functioning of the basal ganglia would still play a role, because a strong performance by the basal ganglia could well give PDS people enough timing signal to speak fluently, but a bad day or stress or emotions makes it worse. This would actually explain why the basal ganglia has not been singled out by experiments.

I am just making guesses, so please comment if you disagree. But I like this way of explaining things, and my brain has not YET found anything that rules it out.


1 comment:

Per Alm said...

Hi Tom,

Your blog is an excellent way of spreading information — I was not aware of the recent article by Brown, Ingham, Ingham, Laird and Fox. Thanks! I also appreciate your reflections very much.

I agree with your comment that it is very possible that the basal ganglia are involved in stuttering even if, in many cases, there is no real dysfunction of the basal ganglia per se. The basal ganglia are a link in a chain of structures that often is summarized as "the basal ganglia-thalamocortical circuits". This circuit starts and ends in the cerebral cortex. The main parts of the basal ganglia motor circuit are: the sensorimotor cortex —> the basal ganglia nuclei —> the VL nucleus of the thalamus —> the supplementary motor area (SMA). My review-paper in Journal of Communication Disorders has the main title "Stuttering and the basal ganglia circuits", where the word "circuits" was meant to widen the scope to include the complete circuit (but I admit it was a bit vague...). In the more recent discussion, in my thesis, I put forward the concept "the medial premotor system", which anatomically is equivalent to the motor part of the basal ganglia-thalamocortical circuits.

From my personal point of view, I think the well-known fluency enhancing conditions provide to strongest argument for involvement of the medial premotor system in stuttering. For example, the effects of metronome pacing and chorus speech on stuttering are in line with what is known about basal ganglia motor disorders. Delayed and frequency-altered auditory feedback has also been found to improve speech for persons with Parkinson's disease.

Tom writes: " BUT I am not sure he [Alm] can explain why there should be brain anomalies as found by Sommer et al. , Foundas et al. , and Jancke et al.".

I think these reported brain anomalies are very interesting and important. There are still many questions that need to be answered in this path of research, and a need for replications, but taken together these 3 studies points to a valid finding of structural anomalies, including sensorimotor regions. I think that at the present stage of knowledge any interpretation of the role of these anomalies in stuttering has to be speculative. With this reservation in mind, a possible integration of these findings and the basal ganglia hypothesis was suggested in section 10.3 of my basal ganglia-paper. I will here present a more developed version of this suggestion. The idea is that the basal ganglia generates go-signals based on the input from widespread cortical regions, but the input from the sensorimotor cortex can be assumed to be especially important. If the signals from the sensorimotor cortex to the basal ganglia are disrupted, distorted, or weak, the basal ganglia will not be able to do their part of the job. The result will be distorted output from the basal ganglia to the SMA, which will make it difficult for SMA to generate the final go-signals for the speech segments.

To build on your illustrative postmen metaphor: you do not get your mail, or you get damaged mail, because the postmen that should bring the mail from the sender (the sensorimotor cortex) to the post office (the basal ganglia) are not doing their job. This might be a common reason for the mail problems, but dysfunctions in any part of the chain may give a similar result: the mails are lost or damaged. I think this is the case also in stuttering. For example, stuttering with adult onset has been reported after lesions in various parts of this circuit, like the basal ganglia, the thalamus, and the SMA. However, the characteristics of the disorder can be expected to be somewhat different depending on location and nature of the dysfunction.

Tom writes: " So it [the basal ganglia hypothesis] seems strong conceptually but no clear experimental signal, which should have been found by now??"

In fact, the meta-analysis by Brown, Ingham et al. did find differences in basal ganglia activation between stuttering persons and controls. The controls showed a weak left globus pallidus activation, which was not shown by the stuttering persons. The interpretation of this is not clear, but a difference has been shown.

Actually, I think it is not surprising that there still is a lack of clear experimental evidence for the basal ganglia hypothesis. Firstly, most published studies have not been designed with the purpose to investigate the role of the basal ganglia. Further, there are several potential problems when studying stuttering by means of functional brain imaging:

1) Heterogeneity. The exact nature and location of the dysfunction can be expected to vary between different cases. Therefore, the resulting patterns of activation in the basal ganglia circuits can be expected to vary.

2) Difficulties to interpret the pattern of activation: basic dysfunction vs. secondary effects? For example, if stuttering is viewed as a problem of speech automaticity, it can be expected that the stuttering person will be using compensatory strategies when trying to speak, involving increased conscious control of speech. I think that large parts of the typical pattern of increased cerebral activation in persons who stutter is a reflection of this: compensatory strategies involving parts of the right hemisphere and the cerebellum.

3) The complexity of the basal ganglia circuits. Interpreting functional brain imaging of the basal ganglia is a very complex task. One complicating factor is that the neural activation seen in brain imaging seems to reflect the activity of incoming synapses, not the firing rate of the neurons within the structure. This has important implications for structures which receives strong inhibitory projections, like the external part of the globus pallidus (GPe). This means that if brain imaging shows activation of GPe, it may reflect increased activity of the inhibitory input to the GPe. The activity of the GPe-neurons is actually reduced when the brain imaging indicates activation. It gets even more complex when there is a combination of inhibitory and excitatory projections. In summary, it is challenging task...

4) Limited spatial and temporal resolution. Brain imaging only gives a rough picture of what is happening in the brain. The fine details can not be seen.

OK, back to the globus pallidus issue: Can I make any interpretation of the brain imaging activation of globus pallidus (GP) in controls, and lack of activation in stuttering persons? Let's start with the controls. If the sensorimotor cortex sends input to the striatum, the striatal neurons may respond with increased firing in both the indirect pathway, to the GPe, and the direct pathway, to the GPi. This would be seen as increased brain imaging activation of the total GP. However, the firing rate of the GPe would be reduced because of the inhibiting input, causing reduced inhibition of the subthalamic nucleus (STN), which, in turn, would send increased diffuse activating input to the GPi. This projection would further increase the activation of the GP seen in brain imaging. In summary, according to this model would brain imaging activation of the GP in controls reflect general activation of the medial premotor system. The lack of brain imaging activation of the GP in stuttering persons could therefore be interpreted as impaired activation of the medial premotor system.

Furthermore, an important function of the GP is to provide a diffuse general inhibition of the cortical target (SMA), with focal activation of a specific response. The pattern shown by the controls is consistent with an increased general inhibition of the SMA, in combination with a focal activation of a specific response. The pattern shown by the stuttering persons is consistent with the combination of impaired general inhibition of the SMA, and impaired focal activation of the intended response. (This state can also be described as reduced signal-to-noise ratio.) Reduced general inhibition of the SMA would be expected to be seen as increased brain imaging activation of the SMA. Interestingly, Brown, Ingham et al. write in their recent paper: "Perhaps the most basal ganglia-specific effect seen in the meta-analysis was the overactivation of the SMA (as well as cingulate motor area) in stutterers." (p. 114) So, in conclusion, the lack of GP activation in stuttering persons seems to fit very well with the observed SMA overactivation! (This model is also in line with the reduced striatal metabolism reported by Wu et al., 1995, based on a PET-study of four stuttering persons.)

In summary, the reported structural anomalies and the reported brain imaging patterns could be integrated with the basal ganglia hypotheses, into the following model: Structural anomalies related to the (left) sensorimotor cortex for the speech organs results in impaired input to the basal ganglia, causing reduced basal ganglia activation, and, in turn, impaired inhibition and overactivation of the SMA.

Hmmm, quite interesting. Discussing and comparing complementary viewpoints can be very fruitful.

I would like to add that it is likely that also factors within the basal ganglia will increase or decrease the risk for stuttering in the proposed model. For example, in my basal ganglia paper, section 9.5, I suggest that a high number of dopamine receptors type D2 in the striatum will increase the risk for stuttering, especially in combination with a low D1/D2 ratio. It is also suggested that a peak in the number of D2 receptors around 3 years of age is a factor behind many cases of transient childhood stuttering. I high number of D2 receptors in the striatum will tend to reduce the diffuse background inhibition of the SMA. A low number of D1 receptors will tend to weaken the focal activation of the intended response. In summary, a low D1/D2 ratio would contribute to a low signal-to-noise ratio in the output from the basal ganglia to the SMA, with a tendency towards overactivation of the SMA.

So my guess is that we will find different types of stuttering. In some cases stuttering might be mainly related to structural anomalies of the sensorimotor cortex, in other cases the basal ganglia dopamine system will play a major role. The response to pharmacological treatment will be different depending on the nature of the basic dysfunction.

Objections or comments?

Thanks Tom, for providing this forum.

Per Alm, PhD
Lund University