Music and Premature Infants: Evidence and Limits

When a baby is born early, families may encounter many messages about what they should do to support development. Some promote songs that supposedly stimulate the brain or guarantee better abilities. These promises deserve scrutiny: studying music in a neonatal unit does not establish that a recording at home will increase intelligence.

The question calls for care because it involves hope, uncertainty and decisions belonging to the clinical team. Here we compare research on functional brain connections with research on development assessed later. Understanding the difference helps us value interesting findings without assigning families another task or responsibility for outcomes beyond their control.

A neonatal unit is a specific setting

A hospital study has eligibility criteria, care conditions and people responsible for deciding when an intervention is appropriate. It does not necessarily include every premature baby or those needing the greatest medical support. Before extending a finding to another situation, we need to understand those criteria. Preterm describes a broad population, not a single health condition.

What researchers called music also matters. A recording selected by a team, a familiar voice and professionally supported music therapy are not interchangeable procedures. They may share sound as a medium while differing in goals, relationship, adaptation and follow-up. NCCIH's general information about music and health distinguishes these types of intervention.

This article therefore provides no volume levels, schedules, durations or instructions for putting headphones on a newborn. Describing an experiment does not turn its procedure into a home recommendation. Decisions about stimulation and family participation should be discussed with the team that knows the baby and their progress.

What the brain images showed

Lordier and colleagues compared premature infants exposed to selected music in hospital with preterm controls and infants born at term. Final analyses included 14 preterm infants receiving music, 15 preterm controls and 16 term-born infants. They investigated connectivity using resting-state functional MRI around term-equivalent age. Certain networks showed greater coupling in the music group. This small sample provided neither an intelligence test nor a study of later school performance. Lordier et al., 2019.

Functional connectivity describes relationships between signals measured in different regions; it is not a count of abilities a baby has acquired. The finding supports questions about brain development. Claims about improved language, movement or daily life require assessments of those abilities, appropriate comparisons and follow-up. Greater resemblance to another group's images is insufficient.

The same caution applies when reading a publication's title. A positive-sounding technical term may refer to a marker rather than improvement observed by families. Asking what was measured does not diminish the study. It places the work within a larger investigation in which mechanisms and clinical outcomes need to be connected through evidence.

LongSTEP followed development at two years

The LongSTEP secondary analysis evaluated parent-led, infant-directed singing supported by music therapists during hospitalisation and/or after discharge, alongside standard care. Of 206 initial participants, 112 remained at the 24-month corrected-age follow-up. No significant differences in language, cognition or motor development were demonstrated between the conditions studied. Bieleninik et al., 2024.

Wide confidence intervals do not exclude meaningful effects in either direction. Participant loss and pandemic-related follow-up difficulties limit precision. Typographical errors in the limitations and supplementary material were corrected on July 12, 2024; this article uses the corrected version. Corrected article and correction notice.

Corrected age places an assessment in context by accounting for birth before the expected due date. It should not simply be confused with the number of months since birth. Preserving that distinction when describing follow-up helps prevent treating assessments at different developmental stages as equivalent.

Interpreting a finding that does not confirm benefit

A nonsignificant result means the analysis did not establish a difference with sufficient statistical clarity. It means neither that a benefit is proven but hidden nor that any effect is forever impossible. Both exaggerations prematurely close an open question. The practical conclusion is to preserve uncertainty and avoid promising an outcome the study did not confirm.

The initial sample is also not necessarily the number available for every final comparison. Missing assessments can reduce precision and raise questions about whether those who remained adequately represent the original group. Recognising this requires no advanced statistics: look for initial participants, available follow-up and reasons for absence before repeating a number.

The two studies do not cancel each other out because they ask different questions. One observes brain signals; the other assesses developmental abilities, and their interventions also differ. It would be misleading to use an imaging finding to avoid mentioning null follow-up results, or to use the latter to declare every musical experience meaningless.

Supporting families without adding pressure

A family interested in singing or music can ask the neonatal team about its purpose, whether it is appropriate at that time, and who would guide participation. They can also ask how rest, tolerance and the baby's needs would be respected. These are questions for a clinical conversation, not a protocol to implement independently.

Not singing, not knowing lullabies or being unable to stay at the hospital all the time does not make anyone responsible for a possible developmental delay. An intervention paper does not establish parental duties. Sensitive presentation of options includes recognising fatigue, schedules, cultural preferences and a family's choice not to add another activity to hospital life.

These studies also provide no basis for recommending a superior composer, promoting a supposed Mozart effect or extending neonatal findings to all pregnancies. When a professional intervention is discussed, distinguish supportive goals from expected clinical outcomes. Our article on what music therapy is and is not explores that distinction further.

A separately researched extension of the series

DW's documentary about music's influence, published in 2022, is the cultural starting point for this series. This article is a researched extension about the lifespan; we do not attribute unverified neonatal scenes to the video. The original Spanish-language documentary appears in music, the brain and emotions. English audio or subtitles have not been verified.

The next article discusses music, memory and dementia, with the same attention to outcomes and limitations. To learn about the Centro's approach, use CMM's contact page. Neonatal decisions belong to the treating team, and this article does not advertise a neonatal service at the Centro.

General educational information. This article does not replace neonatal care, developmental assessment or individual medical guidance.

Music and Premature Babies: What We Know and What Still Needs Research