Cohesion in the discourse interaction of autistic, specifically language-impaired, and normal children.
Autistic kids own the same cohesion toolbox as peers; they just need more accurate practice.
01Research in Context
What this study did
The team compared how autistic, language-impaired, and typical kids hold a conversation.
All children were matched for language age, not birthday age.
The researchers counted every way the children tied their sentences together—pronouns, repeated words, and connectors like "so" and "because."
They looked at how often each tie was used correctly and how often it broke down.
What they found
All three groups used the same kinds of ties in the same order.
Autistic children simply used fewer correct ties and made more errors.
The pattern was the same; the accuracy was not.
How this fits with other research
Baixauli et al. (2016) pooled 24 later studies and still found weaker narratives in high-functioning autism.
That meta-analysis includes the 1992 data you just read, so the story holds across decades.
Favart et al. (2016) show the gap can close by adolescence in writing when the task tells the child to "help the reader."
Together the papers say: cohesion tools are present early, but extra cues and practice are needed for flawless use.
Why it matters
You do not need to teach new cohesion types—autistic kids already have them.
Your job is to give more reps with clear feedback so the errors drop and frequency rises.
During story-retell or conversation practice, mark every correct tie with praise or a token and gently remodel the misses.
A short daily drill can turn the same-old patterns into accurate, natural speech.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick one narrative activity, tally each correct pronoun or connector, and deliver immediate praise for every right tie.
02At a glance
03Original abstract
This study examines discourse cohesion in young normal, specifically language-impaired, and autistic children. The Halliday and Hasan (1976) approach was applied to investigate cohesion strategies used in interactive discourse by 10 autistic, 8 specifically language-impaired, and 8 normal children matched for language age (MLU, receptive vocabulary and syntax). The following questions were investigated: (a) Do all three groups use the same cohesion strategies in their discourse, and if so, do they use these strategies with the same rate and with the same patterning? (b) What can be said with respect to the errors the individual groups make? Results showed that all three groups used the same cohesion strategies. All groups were similar in patterning. Significant group differences were found in the overall rate of correct use and in the use of individual cohesive ties. The groups also differed significantly in overall error rates, and error rates related to individual tie categories. Results are discussed from the perspective of possible variables contributing to group differences. Frequency of occurrence in natural discourse is discussed in terms of specific cohesive strategies used by the children.
Journal of autism and developmental disorders, 1992 · doi:10.1007/BF01046399