Recent Models of Stuttering

1
Recent Modelsof Stuttering
J. Scott Yaruss, Ph.D., CCC-SLPUniversity of
Pittsburgh

• Western Illinois UniversityFebruary 7, 1997

2
Purpose

• To discuss two recent theoretical models of
  stuttering
• Covert Repair Hypothesis(Postma Kolk, 1993)
• Neuropsycholinguistic Model(Perkins, Kent,
  Curlee, 1991)

3
Some Factors That May BeInvolved in Stuttering
Genetic
Environmental
Linguistic
Motoric
Psychological
4
Basic Steps inLanguage Formulation

• Figuring out what you want to say
• Basic message
• Pragmatic intent
• Figuring out how you want to say it
• Grammatical structure
• Specific words
• Sounds in the words
• Prosody

5
Levelts (1989) Blueprint for the Speaker
6
Covert Repair Hypothesis (CRH)(Postma Kolk,
1993)

• Designed to explain the production of speech
  disfluencies by adults who do stutter as well as
  those who do not in the context of general
  language models
• Based upon current psycholinguistic models of
  speech production(e.g., Dell, 1988 Levelt, 1989)

7
Theoretical Foundationsof the CRHI

• Levelts blueprint for the speaker
• Describes conceptualization, formulation,
  articulation, audition, and comprehension
• Explains how speakers monitor the accuracy and
  appropriateness of their speechboth before and
  after speech is produced
• Describes processes speakers can utilize to
  repair detected errors in their phonetic plan by
  interrupting ongoing speech (e.g., Levelt, 1983)

8
Levelts (1989) Blueprint for the Speaker
9
Levelts Monitoring Loops

• Internal Loops
• In conceptualizer, before preverbal message is
  generated
• After phonetic plan is generated by formulator
  (covert speech)
• External Loop
• After speech is articulated (overt speech)

10
Theoretical foundationsof the CRHII

• Dells (1988) Spreading-Activation Model of
  Phonological Encoding
• Connectionist model
• Describes how phonological units are selected
  (and mis-selected)
• When a node is activated at one level, activation
  spreads to all connected nodes at other levels
• Node with highest degree of activation is selected

11
Connectionist Models
12
Spreading-Activation and Phonological Errors
13
Basic Assumptionsof the CRH

• All speakers experience occasional errors in
  their phonetic plan due to mis-selection of
  phonological units
• If errors are detected, they can be repaired
  repairs result in disfluencies
• Disfluencies are the by-product of speakers
  attempt to repair phonetic plans errors

14
Basic Assumptionsof the CRH (cont.)

• Individuals who stutter produce more disfluencies
  because they are assumed to
• have a slow phonological encoding mechanism
• attempt to begin speaking too soon or try to
  speak too quickly for their impaired encoding
  mechanism to make accurate selections
• The result is many phonetic plan errors and many
  opportunities for disfluencies

15
Delayed Phonological Encoding(after Yaruss
Conture, 1996)
16
What the CRH does anddoes not attempt to explain

• Does attempt to explain
• The production of speech disfluencies in the
  framework of normal language processes believed
  to be exhibited by all speakers
• Does not attempt to explain
• The developmental course of stuttering (e.g.,
  repetitions prolongations, etc.)
• The development of accessory behaviors (struggle,
  tension, etc.)

17
CRH Summary

• Speakers have the ability to monitor their
  speech both before and after it is produced
• Speakers sometimes make errors in phonological
  encoding due to the spreading-activation
  mechanism
• If speakers detect these errors, they can
  interrupt speech to repair them before they are
  produced
• The by-product of the interruption is a
  disfluency
• People who stutter produce more disfluencies
  because their phonological encoding mechanism is
  assumed to be impaired so there are frequent
  encoding errors

18
Neuropsycholinguistic ModelPerkins, Kent,
Curlee (1991)

• Speech involves many concurrent language and
  motoric processes
• The timing of the interactions between these
  processes is crucial
• This model, like the CRH, emphasizes timing, but
  now were interested in timing between language
  components
• Also called the temporal dyssynchrony model

19
Neuropsycholinguistic ModelPerkins, Kent,
Curlee, 1991
20
Assumptions of the Neuropsycholinguistic Model

• Disfluencies occur due to a disruption in timing
  between various linguistic formulation and
  motoric execution
• Stuttering is defined as a loss of control
  which occurs when the individual is unable to
  continue speaking, but does not know why

21
Loss of Control

• Some aspects of language formulation are
  accessible to conscious awareness
• Others aspects (e.g., phonological encoding,
  integration) are not
• Loss of control occurs when disruption takes
  place in an aspect of the model that is not
  accessible

22
One Unique Aspects of the Model

• Time pressure
• The pressure a speaker feels to continue speaking
  when he is experiencing a disruption in
  linguistic/motoric integration
• Increases the likelihood that the loss of
  control feeling will occur
• Without abnormal time pressure, speech
  disfluencies may occur, but stuttering will not

23
Summary of Neuropsycholinguistic Model

• Speech requires integration of several aspects of
  linguistic formulation with motoric execution
• Processes can become disrupted
• If disruption is accessible to awareness, then
  a disfluency will occur. If not, the speaker will
  experience a loss of control
• If the loss of control is accompanied by abnormal
  time pressure, stuttering will occur

24
What do these models tell us about the diagnosis
and treatment of individual people who stutter?
25
Not Much!(yet)
However, the general language models on which
they are based can provide guidelines about
specific aspects of language that should be
examine in the diagnosis of individuals who
stutter
26
Levelts (1989) Blueprint for the Speaker
27
Conceptualization
CONCEPTUALIZER
monitoring

• Basic Message
• Does client stutter more on complex messages?
• Does client stutter more on utterances with
  greater propositionality or meaning?
• Pragmatic Intent
• How does social interaction affect fluency?
• Does client stutter more when experiencing
  greater demands on pragmatic or social
  interaction skills?

28
FormulationGrammatical Encoding

• Does client stutter more in situations which
  require more complex utterances?
• Does client stutter more on syntactically complex
  utterances?
• Does client demonstrate problems with syntactic
  development which might interfere with fluent
  speech production?

29
FormulationLexical Access

• Does client exhibit word finding problems that
  might interfere with speech production?
• Does client exhibit frequent mislabelings or
  nonsystematic speech errors which might interfere
  with fluent speech production?
• Are some words harder to retrieve than others?

30
FormulationPhonological Encoding

• Does client frequently produce nonsystematic
  speech errors which might lead to production
  self-repairs or speech disfluencies?
• Does client have difficulty retrieving the
  phonological form of utterances (as in the tip
  of the tongue phenomenon)?

31
Formulation Phonological Development

• Does client exhibit phonological delay which
  might indicate a slow-to-develop linguistic
  formulation system?
• Does client exhibit phonological delay which
  makes communication more difficult and increases
  sensitivity about his/her speech?

32
Articulation

• Does client stutter more when using a faster
  speaking rate?
• Does client stutter more when leaving less time
  for linguistic planning and formulation of
  utterances?
• Does client have difficulties with
  diadochokinetic (DDK) abilities that might
  interfere with ability to rapidly and precisely
  produce speech

33
Summary

• Several recent models of stuttering have
  emphasized the role of language formulation
• These models are important for improving our
  understanding of the nature of stuttering, in
  part because of the research they encourage
• An increased understanding of the language models
  on which these theories are based can help us
  improve the diagnosis of stuttering