Assessment & Research

Effect of maturation on suprasegmental speech processing in full- and preterm infants: a mismatch negativity study.

Ragó et al. (2014) · Research in developmental disabilities 2014
★ The Verdict

Preterm infants show normal phoneme detection but delayed stress-pattern detection at 6–10 months—an early marker you can catch with MMN before language scores drop.

✓ Read this if Early-intervention BCBAs working with preterm infants in NICU follow-up or birth-to-three programs.
✗ Skip if Clinicians who only see verbal school-age clients.

01Research in Context

01

What this study did

Kretschmer et al. (2014) recorded brain waves while 6- to 10-month-old babies listened to sounds. Some babies were born early (preterm). Others were born on time (full-term).

The team watched for a tiny brain spike called MMN. MMN pops up when the brain notices a sound change. They tested two changes: a new phoneme and a new word stress pattern.

02

What they found

Both groups gave the same MMN to phoneme swaps. Preterm babies, however, showed a weaker or late MMN when the beat (stress) of a word changed.

In plain words: preterm infants could hear that "ba" became "da," but they missed the difference between BA-ba and ba-BA.

03

How this fits with other research

Yaari et al. (2018) followed the same preterm/full-term split over the study period. They saw growing gaps on Mullen language and cognitive scales. The MMN stress gap is the first warning sign; the Mullen gap shows where it leads.

Velikos et al. (2015) gave Bayley-III scales at 12 months corrected age. Preterm babies scored below average across the board. Together, the three papers form a timeline: odd MMN at 6–10 m → low Bayley at 12 m → lower Mullen at 18 m.

Kuang et al. (2025) built a model from 6- to 24-month BSID-III scores. It spots very preterm infants who will later dodge ASD with 94 % certainty. Their model uses the same milestones that look shaky in Anett’s preterm group, tying early auditory signs to later social risk.

04

Why it matters

If you serve preterm babies, add a quick MMN or other prosody probe to your intake. A flat response to stress patterns tells you to start language-rich, rhythm-heavy interventions months before standard tests turn red.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

During your next assessment, clap or drum two-syllable words with different stress (BA-by, ba-NAN-a) and watch for orienting—if the baby ignores the shift, flag for speech-language referral.

02At a glance

Intervention
not applicable
Design
other
Sample size
46
Population
developmental delay
Finding
mixed

03Original abstract

Infants born prematurely are at higher risk for later linguistic deficits present in delayed or atypical processing of phonetic and prosodic information. In order to be able to specify the nature of this atypical development, it is important to investigate the role of early experience in language perception. According to the concept of Gonzalez-Gomez and Nazzi (2012) there is a special intrauterine sensitivity to the prosodic features of languages that should have a special role in language acquisition. Therefore, we may also assume that pre- and full-term infants having months difference in intrauterine experience show different maturation patterns of processing prosodic and phonetic information present at word level. The aim of our study was to investigate the effect of these differences on word stress pattern vs. phoneme information processing. Two age groups of infants (6 and 10 month-olds) were included in our study. 21 of 46 of the total of infants investigated were prematurely born with low birth weight. We used the mismatch negativity (MMN) event related brain potential (ERP) component, a widely used electrophysiological correlate of acoustic change detection, for testing the assumed developmental changes of phoneme and word stress discrimination. In a passive oddball paradigm we used a word as standard, a pseudo-word as phoneme deviant, and an illegally uttered word as stress deviant. Our results showed no differences in MMN responses in the phoneme deviant condition between the groups, meaning a relatively intact maturation of phoneme processing of preterm infants as compared to their contemporaries. However, the mismatch responses measured in the stress condition revealed significant between-group differences. These results strengthen the view that the total length of intrauterine experience influences the time of emergence of prosodic processing.

Research in developmental disabilities, 2014 · doi:10.1016/j.ridd.2013.10.006