Late-talking toddlers often show wide variation in language development, but most will catch up without intervention. The study followed late-talking toddlers at 20 months and 4 years, finding that while most resolved without issues, some had persistent language problems. Key predictors of later problems included a family history of language/literacy difficulties, poorer language skills like comprehension at 20 months, and parental language abilities like nonword repetition. For late-talkers without risk factors and good comprehension, a watchful waiting approach is appropriate to avoid over-identifying children likely to catch up on their own.

1. When should we be concerned about
late-talking toddlers?
Dorothy V. M. Bishop
1

2. Wide variation in age at first words
Language milestones
100
90
80
70 Boys, sentences
Percentile
% children
60
with words Girls sentences
or sentences 50 Boys, single words
at given age 40
30 Girls, single words
20
10 Single words-”Three or four
0 different words for people or
objects, correctly used.”
0 12 24 36 Sentences-”Three or more
words, strung together to
Age in months make some sort of sense.”
Epidemiological sample of children born in Newcastle-upon-Tyne
Neligan, G. A., & Prudham, D. (1969). Norms for four standard developmental
milestones by sex, social class and place in family. Developmental Medicine and Child
Neurology, 11, 413-422. 2

3. Most children with long-term language
problems were late talkers
Language milestones
100
90
80
70
Percentile
60 Boys, single words
50 Girls, single words
40
SLI
30
20
10
0
0 12 24 36 48 60
Age in months
Red shows Neligan/Prudham norms
Blue shows children from Dawn House School- NB severe SLI
3
Only 12% of SLI had first words before 18 months

4. http://www.johnbercow.co.uk/
23012007_disabled_children
• I believe that we suffer continually in this country from
a problem of late assessment, late identification and
late diagnosis. ….
• I believe that sometimes those operating the system
play precisely upon the hopes, as well as the more
exaggerated fears, of some parents by saying, “Don’t
worry. There’s plenty of time.” Of course, those who
are articulating that message very often have a
reason of financial self-interest for holding back the
provision of what is needed.
4

5. Most children who have SLI were late talkers,
≠ most late talkers will have SLI
Key issue of ‘base rates’
Consider population of 1000 children
• In whole population, 10% are late talkers = 100
cases
• Suppose 3% of children have severe SLI
In population of 1000, will be 30 cases of severe
SLI
88% of severe SLI were late talkers (i.e. first
words after 18 months), = 26 cases
• So only 26/100 late talkers will have severe SLI
5

6. Butphenomenon ofdo ‘grow out of it’:
The many children the ‘late bloomer’
Late bloomers
Fischel et al (1989)
•26 2-year-olds who “understood complete sentences
but who could say only a few words.”
•Followed after 5 months
Approx 1/3 no improvement
1/3 mild improvement
1/3 in the normal range
Fischel, J. E., Whitehurst, G. J., Caulfield, M. B., & Debaryshe, B. (1989). Language
growth in children with expressive language delay. Pediatrics, 83, 218-227. 6

7. General agreement: early
intervention better than late
• Failure to intervene may lead to
development of secondary problems
• Also, notion of brain plasticity
• Easier to modify brain development before
neural pathways are committed
7

8. Brain plasticity example:
eye-patching (occlusion) for ‘lazy eye’
Effectiveness of occlusion by age
100
% change eye acuity
In vision,
80
much better
difference
60 outcome of
treatment if
40 done early in
life
20
0
24 36 48 60 72 84 96 108 120 132 144
Age (months)
Epelbaum, M. et al (1993). The sensitive period for strabismic amblyopia in humans. 8
Ophthalmology, 100(3), 323-327.

9. Is there a “sensitive period” for language
Evidence from ‘natural experiments’, e.g. ultimate skill of second language
learners much better for those who came to new country before 10 yr old
Adults tested on grammaticality judgement task, e.g. judge correct/incorrect:
The man allows his son to watch TV
The man allows his son watch TV
The man lets his son to watch TV
The man lets his son watch TV
Grammar test score
280
260
240
220
200
native 3 to 7 8 to 10 11 to 15 17 to 39
Age at arriving in US
Johnson, J. S., & Newport, E. L. (1989). Critical period effects in second language learning: the influence of
maturational state on the acquisition of English as a second language. Cognitive Psychology, 21, 60-99 9 .

10. The quandary
• Late intervention
– Children develop secondary problems
– May be too late to be effective
• Early intervention:
– Risk of intervening with children who are just
‘late bloomers’ – wasted resources
– May create problems in the child
10

11. What is needed
• Some way of distinguishing ‘late bloomers’
from those with more persistent problems
11

12. How to find out?
• Need studies that start with late talkers
and follow them up
• At follow-up, divide into those who have
resolved and those with persistent
problems
• Then look at time 1 data and see if we
could have distinguished them
12

13. What factors likely to affect
outcome?
• Home environment*
• Birth risk*
• Genetic risk*
• Severity of language delay
• Profile of language difficulties
* Predictors of late talking in study by Zubrick, S. R., et al (2007). Late language emergence at
24 months: An epidemiological study of prevalence, predictors, and covariates. Journal of
Speech, Language and Hearing Research, 50(6), 1562-1592. doi: 13
10.1044/1092-4388(2007/106)

14. Our study: the children
• Recruited from a database of families from local
maternity ward and local toddler groups.
• Mothers filled in Oxford University
Communicative Development Inventory (OCDI)
when child 18-19 mo old: long list of words –
parent indicates if child says word and
understands it, or just understands, or neither
Bishop, D. V. M., Holt, G., Line, E., McDonald, D., McDonald, S., & Watt, H. (2012).
Parental phonological memory contributes to prediction of outcome of late talkers from 20
months to 4 years: a longitudinal study of precursors of specific language impairment.
Journal of Neurodevelopmental Disorders, 4(3). doi: 10.1186/1866-1955-4-3 14

15. Section of OCDI understands
understands
and says
Total OCDI has 416 words; At 18 months, bottom 15% say <= 10 words
15

16. The children
• 26 late talkers: expressive OCDI more than 1 SD below
the mean (<= 10 words)
• 70 average talkers: OCDI from 20th to 75th centile
• Aimed to include children with a family history of
speech/language problems
– defined as having 1st degree relative with history of receiving
intervention for language or reading problems
• Child seen for a language and cognitive assessment
• 24 late talkers and 59 average talkers available for
follow-up at 4 yr. 16

17. Assessment at 20 months
• Vineland Adaptive Behavior Scales, 2nd edition
(VABS)
– Parental interview used to evaluate the child's
development in the areas of Communication,
Socialisation, Daily Living, and Motor Skills.
• Mullen Scales of Early Learning
– Child-based assessment that provides scores for
Gross Motor, Visual Reception, Fine Motor,
Expressive Language, and Receptive Language
skills.
17

18. Assessment at 20 months
• Mullen Scales of
Early Learning
– Child-based
assessment that
provides scores
for Gross Motor,
Visual Reception,
Fine Motor,
Expressive
Language, and
Receptive
Language skills.
18

19. Assessment at 4 years
• Nonverbal Ability
• Expressive and receptive language
• Parental report of communication skills
Also brief assessment of parental language (done when
child 20 months and 4 years – just one parent, usually
the mother)
19

20. Classification at 4 yrs
N
• Typical development (TD): Normal range
nonverbal ability and no more than one
impaired language measure 60
• Specific language impairment (SLI): impaired
on at least two language measures, with normal 15
nonverbal ability
• In addition, 7 children with below average
nonverbal ability – not considered further
20

21. Predicting outcome for individual children
Best prediction of 4 yr outcome from:
• Child’s language at 20 months + family history +
parent nonword repetition;
Prediction was better than chance, but far from
perfect
21

22. Clinical prediction for individual children
Simple rules of thumb from our data:
High risk of persistent language problems if:
+ Family history and mother scores more than 1
SD below average on nonword repetition test.
OR
No family history but Vineland communication is
below 85 (1 SD below mean)
22

23. Other factors to consider
Thal et al (1991)
• Poor outcome specifically in those with poor
comprehension/gesture
Our study
• Comprehension measures did not add much to
prediction but our sample had few children with
poor comprehension at 18 months
• Have not yet analysed measures of gesture
Thal, D. J., Tobias, S., & Morrison, D. (1991). Language and gesture in late talkers: a
one-year follow-up. Journal of Speech and Hearing Research, 34, 604-612.
23

24. Summary
• Majority of late talkers did not have later
language difficulties
• Having a family history of language/literacy
problems is a risk factor for persisting problems
• Poor comprehension or lack of gesture may also
be important predictors
• Parental report on Vineland can also help
identify those likely to need help
• Appropriate to adopt ‘watchful waiting’ strategy
with late talkers who have good comprehension
and don’t have family history of language
problems 24