Sentence memory of individuals with Down's syndrome and typically developing children.
Language production skill, not talking speed, sets the limit on how much verbal information clients with Down syndrome can hold.
01Research in Context
What this study did
The team tested 24 people with Down syndrome and 24 typically developing kids. Both groups had the same mental age of about five years.
Each person repeated sentences that got longer each time. The researchers also timed how fast each person spoke.
What they found
The Down syndrome group remembered fewer words than the control group. Surprisingly, they talked faster, not slower.
Speaking speed did not predict memory span. Instead, better language production skills meant longer memory spans.
How this fits with other research
Martínez-Castilla et al. (2024) later showed teens with Down syndrome also struggle with auditory stress and receptive vocabulary. Together, the two studies build a chain: poor sound parsing leads to weak sentence memory.
Whitehouse et al. (2014) used brain scans during story listening. They found weaker and oddly placed activation in the same population. The 2004 memory gaps now have a neural picture.
Myers et al. (2018) tracked infants with Down syndrome. Early joint attention predicted later language, not speech segmentation. The 2004 paper shows the gap lasts: production skill still drives verbal memory years later.
Why it matters
Stop drilling rapid articulation. Work on building longer, meaningful utterances instead. Ask clients to repeat full sentences about their interests, then slowly add more words. Track how many words they hold, not how fast they say them.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick a favorite topic, have the client repeat a four-word sentence, then add one word each turn until they top out—note the final count as the baseline span.
02At a glance
03Original abstract
BACKGROUND: Individuals with Down's syndrome (DS) have an auditory short-term memory span disproportionately shorter than the non-verbal mental age (MA). This study evaluated the Baddeley model's claim that verbal short-term memory deficits might arise from slower speaking rates (and thus less material rehearsed in a 2 s passive store) by using the sentence memory subtest of the Stanford-Binet. Previous work had shown digit span recall speaking rate to be comparable to the examiner's slow rate (one syllable per second) for both DS and language-matched participants. METHOD: Thirty individuals with DS were compared to two control groups [non-verbal MA-matched and mean length of utterance (MLU)-matched] on the sentence span and speaking rate for the longest verbatim recalled sentence. Sentence stimuli were presented at a normal speaking rate. RESULTS: The DS group had shorter sentence memory span than the MA-matched group and a faster, rather than slower, speaking rate (syllables per second) than the MLU-matched controls. CONCLUSIONS: Language production level accounted for a substantial portion of the variance in the sentence memory span in the DS group. Thus, language production skill, rather than speaking rate, predicts variability in verbal memory span.
Journal of intellectual disability research : JIDR, 2004 · doi:10.1111/j.1365-2788.2004.00526.x