Lecture Nine Early and Late Childhood Stuttering

1
Lecture Nine - Early and Late Childhood Stuttering

• Orienting Questions How does stuttering change
  over childhood? How does stuttering affect the
  wider spectrum of behaviour?

2
Lecture Outline

• How stuttering helps to understand why fluent
  speech breaks down
• Describe research with children who stutter
• Sensory and motor factors
• Social factors
• How research can inform speech therapy
• Use of external devices to improve fluency
• Using social psychology methodology

3
Prevalence and incidence

• Stuttering will affect approximately 5 of the
  population at some time.
• Disorder affects children disproportionately -
  the age of onset is usually between the ages of 3
  and 7 years.
• Boys more affected than girls. Ratio 31 at onset
• Social factors may be important in this area.

4
Two recent studies that (together) deal with
stuttering from onset to teenage

• The Illinois study (Yairi, Ambrose and
  co-workers) children from around onset of
  stuttering (two years or so of age) to about
  eight years.
• The UCL study from age eight to teenage

5
The Illinois study

• Diagnostic factors Three groups
• Primary
• Secondary
• Other

6
Primary Factors

• Family history
• Persistent stutterers come from families who have
  members who persisted
• Transmission model - single-major locus
  explanation (Ambrose et al, 1993).
• Contrasts with Kidd et al. - sex-modified
  polygenic mode
• Gender
• Boys at higher risk (67 boys)

7
Primary Factors

• Stuttering like disfluencies (SLD) Stutter-like
  disfluencies because typical, but not exclusive
  to people who stutter
• SLD
• part-word repetitions
• single-syllable word repetitions
• disrhythmic phonation
• Other
• interjections
• multiple syllable word and phrase repetitions
• revisions or abandoned utterances
• Dissension about SLD
• Wingate (2001) SLD are not stuttering events
• our own two-class scheme puts some SLD and OD in
  each class
• advocate present separate counts so other people
  can do their way

8
Primary Factors

• Duration of stuttering history
• The longer stuttering goes on, the higher is the
  risk it will persist (esp. girls)
• Disfluency length Extent and duration
• More extensive stuttering events (bu, bu, but)
  are sign of higher risk
• Sound prolongations and blocks
• Apart from early months, prolongation is a factor
  that differentiates children who will eventually
  persist from those who recover.
• Also as shrink, sign recovery in progress.

9
Secondary Factors

• Stuttering severity
• Head and neck movements
• Phonological skills
• Expressive language skills

10
Other Factors

• Concomitant disorders
• e.g. ADHD
• Awareness and affective reactions
• evidence that some young children are aware and
  that this upsets them

11
The UCL study from age eight to teenage

• Illinois work - important information but
  limitations
• Recovery not complete at age 8 (is by teenage)
• No control group, we have limited data on control
  subjects
• Our work shows important changes in disfluency
  and social characteristics happen after age eight
  that relate to diagnosis, assessment and possibly
  treatment.

12
The UCL study from age eight to teenage

• Follow-up study of children who stutter between
  the ages 8 and 18 years.
• First seen when attending clinic for initial
  assessment at around 3 - 8 years
• Cohort constant in terms of treatment received
• Recorded at age 8 or younger so verified that
  stuttering
• Followed up to age at which recovery complete

13
The UCL study from age eight to teenage

• Total number of children who stutter with English
  as a first language involved in the study - 372.
• Not all children have performed the full range of
  tests. Some of reasons
• still participating (not yet reached age
  appropriate for some of the tests).
• Geographical
• Lost contact

14
The UCL study from age eight to teenage Core
Group

• Important subgroup of the children we have seen
  over the years
• Consists of 24 children all of whom have been
  classified as persistent (13) and recovered (11)
  using the following criteria
• Three recordings - 8-10 years, 1012 years and 12
  years
• SSI
• Parental and child rating using adaptation of
  Kully and Boberg questionnaire
• Researchers rating

15
The UCL study from age eight to teenage
Secondary Group

• Consists of 84 children who stutter for whom we
  have a minimum of one audio recording, together
  with history of therapy, initial contact details,
  date of birth, gender and at least one of the
  research projects.
• Sixty-six (79) of secondary group have been
  classified as recovered/persistent
• Children in this group can be transferred to the
  Core Group when at appropriate age / assessment
  completed etc

16
We have the following information/data on the
Secondary Group

• Motor (tapping) 23/84 (27)
• Motor (lip tracking) 14/84 (17)
• Temperament 29/84 (35)
• Handedness 67/84 (80)
• Grammar 24/84 (29)
• Vocabulary 17/84 (19)
• Lexical retrieval 20/84 (24)
• Priming 5/84 (6)
• Intelligence 10/84 12)
• Academic 37/84 (44)
• Five children (6) are no longer participating

• Therapy history 84/84 (100)
• Demographic 40/84 (48)
• Family History 39/84 (46)
• Anxiety 27/84 (32)
• Health 44/84 (52)
• Scans 9/84 (11)
• Age of Onset 55/84 (65)
• Hearing Threshold 40/84 (48)
• Hearing problems 40/84 (48)
• Masking threshold 40/84 (48)

17
Other participants

• There are a further 164 children for whom we have
  a minimum of one audio recording, but incomplete
  data with regard to therapy, contact details etc.
  
• Studies of recovered/persistent use subjects from
  core and secondary groups
• Studies that havent looked at recovered/persisten
  t per se (e.g. some of the hearing tests and
  social studies) used selection of all the
  speakers

18
Sensory and Motor Factors
Do children who persist have a hearing / timing
problem?
Reason that usually motivates investigation of
hearing is because altering hearing improves
speech control (more about this later)
19
Survey data (n 61)

• Parents reported a high incidence of hearing
  problems in the General Health Questionnaire
  conducted as part of demographic assessment.
• Further investigation established a high
  incidence of otitis media with effusion (OME)
• 21.3 overall (13/61)
• with 16.7 persistent (5/30), and 25.8 recovered
  (8/31)
• There was also a high likelihood of OME being
  treated by ventilation tubes
• 38.5 overall (5/13)
• 20 persistent (1/5)
• 50 recovered (4/8).
• Possible that stuttering recovered because
  hearing problem resolved (NB would be peripheral)

20
Data from Hearing Studies

• Used masking technique (probe sound masked by
  white noise)
• results indicate that auditory sensitivity for
  sounds in noise continues to develop through to
  teenage
• a different pattern of auditory development
  exists for the participants who stutter compared
  with participants who do not stutter.
• Backward masking performance at teenage
  distinguishes speakers who persist in their
  stutter from those who recover.

21
Hearing - conclusions
Data that have been collected on a sample (n
61, 30 persistent, 31 recovered) of these
participants show that for recovered speakers at
teenage, there are no hearing problem either at
central or peripheral levels. In speakers who
persist in their stutter, there is a hearing
problem that affects central hearing at teenage
that is specific to speakers who persist in
their stutter, which could be a residual problem
hanging over from OME in early childhood.
22
Data from Timing Studies

• Used various tests from the Dow Moruzzi battery
  designed to assess performance of the cerebellum
• As with hearing, cerebellar performance continued
  to develop through to teenage
• a different pattern of cerebellar development
  exists for the participants who stutter compared
  with participants who do not stutter.
• No studies yet to test fro differences between
  persistent and recovered people who stutter

23
(No Transcript)
24
Conclusions

• Evidence to implicate auditory and speech timing
  factors in the onset and development of the
  disorder.
• Moderating timing mechanisms using altered
  auditory feedback improves fluency.

25
The role of biological, environmental and social
factors in the breakdown of fluent speech
26
Speech is a social phenomenon. It is therefore
important to know what role social factors, and
their changing role over development, play in
stuttering. Stuttering has been shown to be
topic- (Mallard Westbrook, 1986) and
situation-specific (Yaruss, 1997). The disorder
is governed in part by affective factors that
are socially moderated.
27
Pre-school and early years

• Intelligence
• Attitudes toward speaking
• Temperament
• Family Interaction

28
Intelligence - EMPIRICAL

• CWS score significantly lower on intelligence
  tests than do fluent controls. Evident in both
  verbal and non-verbal intelligence tests.
• the intelligence and social class of CWS who are
  receiving treatment is found to be above average

29
Intelligence - PERCEPTION

• Fluent children report CWS as less intelligent
  than controls
• Adults perceive adults who stutter to be have
  average or above average intelligence.

30
Attitude - EMPIRICAL

• Attitudes have been shown (in particular
  circumstances) to be causally-related to
  behaviour
• Stuttering usually starts between three and five
  years.
• Attitudes about many things can change between
  ages 3 and 5

31
Communication Attitudes and Stuttering

• Why do we need to measure the communication
  attitudes of young children who stutter?
• How can the communication attitudes of young
  children who stutter be assessed
• How can information on attitudes be used to
  improve prediction and treatment?

32
Communication Attitudes and Stuttering

• Stuttering usually starts between three and five
  years (Dalton Hardcastle, 1977)
• An instrument that offers an indication of
  attitude in young children would be useful for
  early diagnosis and intervention (Yairi
  Ambrose, 1992 Onslow, 1994).
• Increase of intervention at a young age
  (Lidcombe) etc. Need to be able to assess
  attitude change to measure efficacy of treatment.

33
Communication Attitudes and Stuttering - chicken
or egg?

• Conflicting views as to whether the negative
  communication attitudes of cws reported in the
  literature are cause or effect
• the onset of stuttering is a result of the belief
  that speech is difficult (Bloodstein, 1987
  Brutten Dunham, 1989)
• the negative beliefs that people who stutter have
  about speech is a product of, rather than a cause
  of their dysfluency (Guitar, 1976 Peters
  Guitar, 1991).

34
Computer assessment

• Animation and sound
• Standardised presentation of stimuli
• Response collection and statistical analysis are
  automated

35
Participants

• 143 monolingual English children
• 68 boys 35 girls, mean age 5.2 years (sd .59)
• 22 children with English as an alternative
  language
• 8 boys 14 girls, mean age 5.32 years (sd .58)
• 16 monolingual English children who stutter
• 12 boys 4 girls, mean age 5.83 years (sd .82)

36
A one way ANOVA indicated a significant effect of
language group Post hoc tests indicated that
there was a significant difference between the
mean scores of the monolingual English children
and children who stutter (p.001) and between
EAL children and children who stutter (p.023).
37
Exploratory factor analysis

• Factor 1
• classroom-based situations
• Factor 2
• play and home-based situations
• Factor 3
• talking with known older children and adults
  (relatives and friends)
• Factor 4
• talking with other children and adults

38

• Significant effect of language group across the
  four factors
• Post hoc tests show persistent group to be
  significantly more anxious than
• their monolingual peers on 3 out of the 4
  factors.

39
Conclusions - young children and communication
attitudes

• Identification of stuttering
• Negative attitudes apparent close to onset
• Are those children with more negative attitudes
  toward communication more at risk for persistent
  stuttering?
• COMCAS allows measurement of specific areas of
  communication attitudes.

40
Temperamental characteristics and fluency
breakdown

• Research evidence that temperament contributes to
  cognitive development (Miceli et al, 1998 Singer
  Fagan, 1992)
• Also contributes to language acquisition (Dixon
  Shore, 1997 Dixon Smith, 2000)
• Relationship with speech and language disorders
• delayed language development (Caulfield et al,
  1989 Paul et al, 1990,1997)
• stuttering (Anderson et al, 2003 Embrechts et al
  2000 Lewis Goldberg, 1997

41
Temperament - EMPIRICAL

• Temperament is distinguished from personality
  traits by being much more physiologically based.
• Temperament is a contributing factor to both the
  development and persistence of stuttering.
• CWS exhibit a more vulnerable or sensitive
  temperament, which could possibly be a
  contributing factor in the development,
  maintenance or chances of recovery of stuttering

42
Temperament and Stuttering

• 10 CWS versus 10 controls
• CWS
• 7 boys, 3 girls age range 3 years 7 months - 7
  years 2 months, mean 5years 10 months (sd 1.36
  years)
• Controls
• 7 boys, 3 girls age range 3 years 8 months - 7
  years 2 months, mean 5 years 4 months (sd 1.16
  years)

43
Temperament and Stuttering

• The Behavioral Style Questionnaire (BSQ)
• Parental report
• Measures nine components of temperament
  Activity, Predictability, Approach, Adaptability,
  Intensity, Mood, Persistence, Distractibility,
  Threshold

44
Conclusions

• We found temperament differences others have not
• Anderson found a temperament difference that
  Embrechts did not
• One variable that went in different directions
  between Anderson and Embrechts
• One consistent finding CWS not adaptable

45
Family interaction - does parental speech rate
and/or linguistic style affect fluency?

• Speech Rate
• when parents reduce speech rate their 3 yo
  children also reduce speech rate (Guitar
  Marchinkowski, 2001)
• no difference in the speech rate of parents of
  children stutter compared to controls (Kloth et
  al, 1995, 1998)

46
Family interaction - does parental speech rate
and/or linguistic style affect fluency?

• Linguistic style
• parents of pre-school children at stuttering
  onset use less complex language than controls
  (Kloth et al, 1995)
• children will begin to stutter after
  intervention to enhance expressive language
  (Hall, 1997 Meyers et al, 1990)
• relationship between conversation turn-taking
  style and breakdown of fluency (Winslow Guitar,
  1994)

47
Adolescence

• Bullying
• Self-esteem
• Personality
• Anxiety

48
Bullying - EMPIRICAL

• Two main questions
• 1. Are CWS more likely to be bullied than
  non-stuttering peers?
• 2. Does bullying exacerbate the problem of
  stuttering?
• There only appears to be evidence for the former.

49
Factors that increase the risk of a child being
bullied

• few friends
• low-status friends
• rejection by peers
• these factors often apply to children with
  special educational needs.

50
Why are CWS likely to be bullied?

• often reluctant or unable to participate verbally
  in school activities (or social groups in
  general).
• this may lead to them being seen as shy or
  withdrawn
• would lead to the perception that they have few
  friends

51
Bullying - EMPIRICALRatings by fluent adults of
children who stutter.

• CWS were not readily accepted as members of their
  classroom
• stuttering students who were more able to adjust
  to interpersonal situations responded better to
  therapy

52
Bullying - EMPIRICALRetrospective self-ratings
by adults who stutter.

• 11 of adults who stutter report being bullied at
  school (compared to 30 in general population)
• More recently 74 of 276 adults who stutter
  report being bullied at school
• 6 of these stated that the bullying had a
  long-term negative effect on their stuttering.
• Problems with long-term retrospective ratings

53
Social Exclusion and Bullying

• Single classroom group in each of 16 schools
  across England
• One child who stutters in each class
• Sociometric interview procedure
• Allocation to social groups
• Nomination to behavioural categories

54
Social Exclusion and Bullying

• Social Groups
• CWS twice as likely to be rejected than controls
  (43.518.86)
• CWS four times less likely to be popular than
  controls (6.2525.84)
• Using chi-square analysis both differences are
  significant
• Trend for CWS to adopt the predominant social
  profile of the group

55
Social Exclusion and Bullying

• Behavioural Categories
• Control children twice as likely to be nominated
  as leaders than CWS (12.92 compared to 6.25)
• More CWS nominated as bully victim (37.25) and
  seeks help (25) than controls (10.6 and 13.18)
• Using chi-square analysis significant difference
  in nominations to bully victim category

56
Social Exclusion and Bullying

• Persistent/Recovered
• No significant differences between groups in
  either social or behavioural
• No significant relationship between stuttering
  severity and social preference scores

57
Social Exclusion and Bullying

• Conclusions
• CWS more likely to be bullied and rejected than
  their peers and are less likely to be popular
• CWS tend to adopt the predominate social profile
  of the group.
• No relationship between severity and social status

58
Bullying - EMPIRICAL

• Problems with sociometric interview
• Triangulation of methodologies
• Other methods of assessing social profile
• diaries
• phone logging

59
Self-esteem - BACKGROUND

• Self-perception can be affected by speech
  disorders and can result in low self-esteem
• Many therapeutic techniques often include goals
  to improve an individual's concept of self-worth
• Actual role of self-concept is in stuttering is
  disputed
• once individuals have developed a concept of self
  as a PWS spontaneous recovery becomes less
  likely.
• PWS respond more positively to therapy if their
  speech disorder is integrated as part of their
  self-concept.

60
Self-esteem - EMPIRICALSelf -ratings

• Children
• using a variety of rating instruments, most
  studies report no difference between CWS and
  normative data.
• Adults
• relationship between low self-esteem and high
  rates of stuttering
• a trend toward low self-esteem
• feelings most commonly expressed by PWS are
  anxiety, helplessness, victimization, and low
  self-esteem.

61
Personality - EMPIRICAL

• Investigations using a number of different
  assessments show no differences between
  school-age CWS and controls (CWNS) in personality
  factors related to neuroticism or anxiety.
• Personality alone does not appear to be a
  predictor of the developmental pathway of
  stuttering (persistency or recovery). Personality
  traits are useful in combination with other
  factors

62
Personality - PERCEPTION

• CWS are generally stereotyped as more nervous,
  shy, withdrawn, tense and anxious
• Unlike intelligence these negative perceptions
  are continued into adulthood.
• Held by a wide range of groups, from members of
  the general public, including college professors
  and teachers and even clinicians

63
How people who stutter are perceived

• By parents of children who stutter
• A high proportion of parents described their
  children variously as - nervous, highly strung,
  anxious, fearful, clingy or tense. (Starkweather,
  1997)
• By Clinicians
• 75 described children who stutter as nervous
  and fearful and 64 as shy and insecure (Wood
  Williams, 1971)

64
How people who stutter are perceived

• By academics and students
• increased social distance
• rated as lower academic ability
• highly anxious (McKinnon, Hess Landry, 1986)
• Rated more negatively on 20 personality traits
• Academics rated people who stutter more
  negatively than did the students
• (Dorsey Guenther, 2000)

65
Does research support the concept that people who
stutter are more anxious than non-stutterers

• Yes
• self report (Gabel et al 2002)
• saliva cortisol levels (Blood et al 1994)
• No
• self report (Peters Hulstijn, 1984, Blood et
  al, 1994)
• state/trait anxiety inventories (Miller Watson,
  1992)
• Skin conduction response (Dietrich and
  Rohman,2001)

66
Anxiety

• We examined for differences between the recovered
  and persistent groups (and between each of these
  groups and fluent controls) in trait and state
  anxiety at teenage.
• Trait anxiety is a personality characteristic
  (how anxious a person is).
• State anxiety is a transitory state that varies
  and fluctuates over time and situation.

67

• Three groups persistent (n18), recovered
  (n17), control (n19)
• 12 19 years
• Groups age and gender matched
• Recovery / persistence established as previously
  indicated
• STAIC (Speilberger et al, 1970)
• Looked at feelings in 4 different speaking
  situations (states) plus trait anxiety

68

• Significant effect of language group across the
  four state conditions
• Post hoc tests show persistent group to be
  significantly more anxious than
• controls in the shop, phone and class conditions
• No significant differences between groups in the
  trait condition.

69
Conclusions

• No significant differences between groups in
  underlying trait anxiety
• Persistent stutterers more anxious in
  communication situations than recovered and
  control participants
• Is this an effect of continued stuttering?
• Or have children recovered from stuttering
  because anxiety has been reduced (therapy)?
• Latter is unlikely as therapy that participants
  experienced (MPC, City) does not focus on anxiety
  reduction
• It would seem that anxiety is not a causal factor
  in stuttering but measurements of anxiety could
  be useful way of tracking recovery

70
Adults

• Occupation and career

71
Occupations - PWS self-report

• 70 of the participants reported that they
  believed they could have had a better job if they
  did not stutter
• 56 reported deciding on their career because it
  required less speaking
• 35 of the participants responded that they
  believed that their career progression had been
  affected because they stuttered.
• 32 felt discriminated against in the employment
  process

72
Occupations - Employers perceptions

• PWS viewed more positively following successful
  speech therapy
• positive change in career
• more success at interviews
• promotion and upgrades

73
and finally (1)

• CWS receiving treatment score above norms on
  intelligence tests - this may be an effect of
  social status.
• People who stutter are generally perceived as
  more anxious, nervous, shy and insecure. Also
  rated as less academically able.
• Little evidence that parental speaking styles
  affect childrens fluency.
• Retrospective reports indicate that bullying
  and/or rejection at school can affect long term
  fluency.

74
and finally (2)

• Sociometric studies show a relationship between
  language groups and bullying and/or rejection.
• No evidence that social isolation causes fluency
  breakdown or leads to persistent stuttering.
• The self-esteem of PWS becomes more negative as
  they become older

75
and finally (3)

• Negative communication attitudes are present in
  children who stutter close to onset of the
  disorder.
• Children who stutter present with a different
  temperamental profile compared to children who do
  not stutter.
• PWS are widely perceived to be anxious and
  nervous although empirical evidence is
  contradictory.
• Employment opportunities for PWS are restricted
  but improve if fluency improves

76

• Social and emotional factors are critical in the
  onset and maintenance of stuttering
• Most models explaining the onset and development
  of the disorder include a social component
• There are methods and issues in social psychology
  that can be used to investigate this area

77
Furnham, A., Davis, S. (2004). Involvement of
social factors in stuttering A review and
assessment of current methodology. Stammering
Research, 1, 112-122. Howell, P., Davis, S.
Williams, R. (submitted). Late childhood
stuttering. Journal of Speech, Language and
Hearing Research. Wingate, M. (1988). The
structure of stuttering A psycholinguistic
study. New York Springer-Verlag. Wingate, M. E.
(2002). Foundations of stuttering. San Diego
Academic Press. Yairi, E., Ambrose, N. G.
(2005). Early childhood stuttering. Austin TX
Pro Ed. Speech Group website http//www.speech.p
sychol.ucl.ac.uk/