Stuttering 2024: updates for research and practice

Outline of the morning:

• (1) Current research understanding of stuttering
• (2) Early intervention of stuttering with preschool children who stutter (CWS)
• What we do
• Evidence of what works
• (3) What are the effective components of stuttering treatment in older people who stutter (PWS)

Sources for this presentation

• Much of the material for this presentation derive from the following three publications:
• Bloodstein, O., Bernstein Ratner, N. & Brundage, S.B. (2021). A handbook on stuttering, 7^th ed. San Diego: Plural.
• Chang, S-E., Bernstein Ratner, N. et al. (in review). Stuttering: Our current knowledge, research opportunities, and ways to address critical gaps. Neurobiology of Language. (Proceedings of NIDCD workshop on stuttering, June 2024)
• Godsey, A. & Bernstein Ratner, N. (in press). All in good time: Parent-child turn-taking in early stuttering. American Journal of Speech-Language Pathology.

​

Are we any closer to understanding what causes stuttering?

YES!!! We think that stuttering is…

First, a working definition of stuttering:

• Stuttering is more than observed disfluency: we often describe stuttering as having “ABC features”
• Behaviors (both primary stuttered speech segments and associated physical reactions (sometimes called secondary behaviors)
• Affective (emotional) features (feelings, fears, anxiety) that accompany both speaking and stuttering
• Cognitive (thought) features (beliefs about stuttering, what makes it better/worse, etc.
• We seem to be making more progress on understanding why the behaviors emerge
• We think we know how the A’s and C’s emerge, but not all therapies have been paying attention to them.

​

The ABCs: a diagram

Learned

(almost certainly)

Learned

(almost certainly)

Probably NOT learned

(with exception of secondary/escape behaviors

A genetically-transmitted disorder

• Candidate genes (e.g., Kang, et al. 2010 NEJM; Drayna lab ); in areas of Chromosome 12
• But they are not found in most PWS
• There is likely to be more than one chromosomal locus
• Exactly what is being inherited?
• NOT CLEAR.
• These are not simple sequences that can be turned on or off – they control broad functions (e.g., liposomal storage) – no “knock-outs” or gene therapy is likely in near future.

What are these genes doing?

• Chow et al. (2020) found that expression patterns of GNTPG and NAGPA in the brain correlated with brain regions where group differences (PWS/PWNS/CWS-Persist/CWS-Recover/CWNS) were found in gray matter volume, suggesting a potential mechanism by which the genes in stuttering may lead to its symptoms and recovery profiles.
• Benito-Aragón et al. (2020) found that expression of GNTPG was spatially correlated with cortical networks shown to be involved in stuttering.
• Furthermore, Chow et al. (2021) found that, compared to controls, participants who stutter with rare heterozygous AP4E1 variants showed reduced volume in the corpus callosum, brain areas where gene expression of AP4E1 is relatively high.
• SO, the story is starting to come together

8

A “stuttering” mouse? Yes, Barnes, et al.(2016) https://www.youtube.com/watch?v=05YQXWzVFm8

What has this taught us?

• Recent work using this “designer” mouse suggests that astrocytes, which are brain (glia) cells highly expressed in the cerebellum play a role in stuttering (Adeck et al., 2024; Han et al., 2019; Maguire et al., 2021; Turk et al., 2021).
• Astrocytes in the mutant mouse vocal motor circuits are smaller and have reduced complexity (Adeck et al., 2024).
• Again, we are finding a link from genes to structure to function.

So, why should we care about genetics?

• It seems to still be “news” that stuttering can be inherited rather than learned or acquired through trauma (☹)
• That is mostly what the public media seemed to find interesting
• If we can identify the gene(s), and THEN what they code for (what they do), we MAY be able to develop better drug therapies and understand the disorder better
• NOW, we can counsel parents and PWS that stuttering is nobody’s “fault”: it is genetics, anatomy and physiology.

​

​

So why are some twins discordant for stuttering?

• No genetic condition is 100% concordant even for identical twins.
• This is due to “environment”, but to a geneticist, that is NOT upbringing or experience!
• It is the physical environment of the dividing embryonic cells and gestational environment!
• We really don’t have evidence that experience or parenting either cause or impact the life course of stuttering
• Although they can clearly impact learned reactions, affective/cognitive features of stuttering and Quality of Life (QoL).

Anatomical/physiological differences not just between PWS/PWNS, but CWS-Persistent and CWS-Recovered

• General findings:
• Both white and grey matter differences
• Connectivity differences (networks among brain regions)
• Lateralization of function differences
• Activation and event-related potential differences on speech and language processing tasks

All are now documented in both adults and children who stutter

​

Illustrative differences (Chow, et al., 2023)

It’s not just about areas of the brain, it’s about connections among them… (Chang & Guenther, 2020)

​

• Lower volume in certain areas of the cortex, and reduced connectivity between regions – a faulty switchboard, in some respects – sometimes you get a good connection, sometimes you don’t.
• And unlike a switchboard, people develop strategies for dealing with a “noisy signal”

Stuttering involves more than speech production: disruption of Event Related Potentials (ERPs)

• N400 – responds to semantic (meaning) violations…
• I like to drink tea with sneakers (red line)

• P600 – responds to grammatical/syntax violations…
• The boys is running (red line)

Sample findings of differences among CWNS/CWS-P/CWS-R (Usler & Weber-Fox, 2015)

Expressive language skill predicts recovery from stuttering (Leech, et al., 2017;2019; Singer, et al., 2022)

• Growth over time (see also Hollister, et al., 2017)

​

• Language profiles at stuttering onset (Singer, et al., 2022)

​

Implications for clinical work

• We also know that tested language abilities, language sample analysis measures and tested articulation scores distinguish between CWS-P and CWS-R
• We also know that there is high comorbidity between stuttering and other developmental communication disorders
• All children referred for stuttering should get a complete communication evaluation, not simply a fluency “work-up”

​

The intersection of language processing and motor stability: the spatial-temporal index

Work by Anne Smith, Christine Weber (Fox) [Purdue] and colleagues

Practiced motor movements:

Ex: Handwriting…

And its regularity across exemplars

Subject’s Task: Say,

“Buy Bobby a puppy”

Measure: stability of repetitive movements to obtain a spatial temporal index (STI)

​

Results:

Children are variable over many attempts, adults are not

​

CWS and motor performance

• Motor coordination in children who stutter is particularly destabilized when they are faced with language demand.
• This instability predicts persistence (Usler, Smith & Weber, 2016)

​

• Stuttering is a unique disorder – it appears to involve BOTH SPEECH AND LANGUAGE

​

Motor learning is depressed in CWS-P (Tendera et al. 2020; finger tapping sequence task)

Accuracy is lower (BLUE)

Latency (time) is higher

Implications for some therapy �goals

• How long should it take for someone to “master” the techniques you provide (e.g. both fluency shaping and stuttering modification? How soon should the CWS “get off the caseload”?

​

• How long will it take someone to be an “expert”, especially when people are watching and waiting for you to make a mistake?

HOWEVER… stuttering is more than its obvious symptoms

• Because stuttering starts at such a young age,
• And because learning impacts how the child and adult who stutters interacts with the world,
• It is likely that TREATMENT for stuttering will need to go past CAUSES of stuttering.
• And
• It is likely that CAUSES of stuttering will not directly inform most effective treatment.

Thought break: some possible questions to consider

• What are the advantages/disadvantages of using an animal model to understand the neurobiology of stuttering?

​

• Some clients or families are NOT happy to find that a childhood condition is neurologically-based.
• Why might this be?
• How might you address this?

Research and controversies ABOUT therapies:

What are appropriate therapy outcomes?

Which therapies are evidence-based?

Is therapy for early stuttering effective?

• Many children recover from stuttering rather soon after onset.
• This makes it difficult to know if therapies “work” (we need many more cases than we have)
• Additionally, we don’t have great evidence that the “pieces” of therapy programs are actually “mechanisms of action”

Concern:Preschool therapy outcomes research

.

​

​

Some problems with “mechanisms of action” in early stuttering therapies

• Lidcombe therapy: uses parents to reinforce fluent utterances, urges retrials of stuttered utterances (on contingency schedule), without direct instruction.
• Very strong research support, if you count numbers of articles (although many are from the same originating team)
• Outcome data confounded by length of treatment (see next slide) and drop out rates
• BUT: fidelity of parental contingencies is not associated with outcomes (Swift, et al., 2016; Donaghy, et al., 2020).
• That means that kids get better even if parents don’t stick to the contingencies

How long would you give your client to show improvement?

• E.g., From the RESTART Study (2015) [treatment of early stuttering RCT enrolling ~200 children]: “in both groups, most improvement in %SS occurred in the first 3 months of therapy.”

​

• How long would you give your client to show a “flat” response?

“Indirect” therapies

• Started with the Diagnosogenic Theory (don’t call attention to the stuttering), but morphed into more practical issues – like how to do any direct instruction with very young children (stuttering has an average age of onset between 2-3 years)
• RESTART-DCM, Palin PCT: parents utilize “Demands and Capacities” model to reduce communicative stressors on child. E.g.,
• Slow parental rate
• Increased turn-taking latencies
• Simpler language input

​

Evaluating this advice using the classic Illinois Stuttering Project (at FluencyBank.org)

• 53 CWS (14 CWS-Persistent; 39-Recovered)
• At T1: 19,602 CWS child utterances; > 16K maternal utterances
• 27 CWNS
• After baseline (Time 1), all families given uniform guidance to:
• Slow rate,
• Increase turn-taking latencies

Slowing speech and lengthening turns (Allison Godsey, AJSLP, in press):

• Measures:
• Turn-taking latencies (sampled)
• Parental speech rate
• Computed over time-aligned transcripts

• RESULTS (see next slides)
• Parental speech rate slowest to CWS-Persistent (CWS-P) before advisement
• Not predictive of either within-session fluency or diagnostic outcomes

​

​

Temporal findings (Godsey, in press)

• No differences in turn-taking latencies between CWS-P and CWS-R
• Consistent with Lidcombe team observation that latencies reduced in one trial (Amato Maguire, et al., 2023)
• For CWS, no correlations between latencies (or predictability of latency) and stuttering rate
• Parental rate faster to CWS-Recovered

Godsey

38

​

Questioning behaviors (Garbarino, 2022)

• For utterances produced by CWS, answers to adult questions were significantly less likely to contain stuttering-like disfluencies than other utterance types, and this was still true after controlling for utterance length and grammaticality.
• In contrast, for utterances produced by CWNS, answers to questions were significantly more likely to be disfluent

Parents’ language complexity (Burns, 2022 and in ms prep)

• Computed Mother-child difference scores for 5 LSA measures
• NO significant differences between CWS/CWNS or CWS-P/R
• Trend for CWS-Rec to receive more sophisticated input

What do we do with this research?

• First, reconsider the message to families
• They may NOT have the agency we tell them they have
• Problems with guilt when things don’t change
• Next, we need to reconsider the implicit goal we convey:
• A good outcome is no stuttering/recovery; shifting to other goals later implies a “second best” outcome
• We do not do this with other developmental disorders of childhood
• We work on creating CHILD agency and joint problem solving between children and their parents.
• Modestly, I believe we will have more discussions about this going forward, together with some “de-implementation science”

​

Therapies for adults

• We’ve always had short-term evidence of “effectiveness” of fluency-shaping/speech restructuring therapies
• BUT: relapse is common, and mental effort is significant enough to be seen in brain imaging studies!
• Stuttering modification, which targets ABC’s (see next slide) takes a very long time to master
• And can target fluency as an outcome that outweighs affective and cognitive gains
• The “ableism” movement is emerging in the neurodiversity community and will impact how we reconfigure therapy goals (Gerlach-Houck & Constantino, 2022)
• From early stuttering onward, as all preschool programs define “success” as recovery from stuttering symptoms

​

Direct stuttering�Treatment

Fluency Shaping/Speech Restructuring

Possible components of stuttering modification

Mindfulness and�Acceptance &�Commitment�Therapies

The ABC’s of stuttering: a mandate for research going forward (Chang, et al., in press)

• Reminder: stuttering is an interactive system of:
• Behaviors: stuttered moments, secondary behaviors
• Affect: feelings and emotions regarding stuttering, speaking
• Cognition: beliefs and thoughts about stuttering (when you stutter, why you stutter, what you think you can do about stuttering)

How central is fluency to stuttering outcomes?

• A question for clinicians, PWS and families of young children
• To some extent, more informed therapies will need to answer a yet-unsolved question:
• Is “recovery” from stuttering, or predictable response to treatment available in all cases?
• Few medical or educational (or sports) interventions yield predictable results
• But we often blame either professionals or clients for less than optimal outcomes

Finally, the role of Common Factors in therapeutic interventions

What does all this mean?

• In psychology, there is a robust body of research to suggest that the CLINICIAN MATTERS MORE THAN THE THERAPY.
• A good clinician:
• Has options to offer the client (knows how to provide more than one type of therapy)
• Negotiates mutually agreed-upon goals of treatment
• Monitors to assess outcomes of therapy (even if lots of evidence suggests it should work)
• Changes course as necessary
• Keeps up with the research literature

Summing up what we know about stuttering and its treatment (Chang, et al., in press)

That’s a lot of information:�Questions? Thoughts?

Write me at nratner@umd.edu if you have additional questions.