Sleep & RecoveryResearch PaperPaywall

Infant Sleep Cycles Lengthen Dramatically in the First Year of Life

A large-scale actigraphy study of 152 infants reveals sleep cycles grow ~10 minutes longer between 3 and 12 months, with breastfeeding linked to longer cycles.

Sunday, June 14, 2026 2 views
Published in Sleep
A sleeping baby wearing a small wrist actigraphy device on a white crib sheet, soft natural light from a nearby window

Summary

Researchers tracked sleep cycles in 152 infants over their first year of life using wrist actigraphy, accumulating more than 35,000 hours of data. They found that infants have shorter sleep cycles than adults — roughly 62 minutes versus 81 minutes in parents — and that these cycles lengthen by about 10 minutes between 3 and 12 months of age. Breastfeeding was associated with longer cycles in both infants and their mothers. The study also found that the physical pattern of limb inactivity during sleep matures significantly within just the first year. These findings open the door to using wearable actigraphy at scale to track brain and sleep development in infants, potentially flagging health or developmental issues early.

Detailed Summary

Sleep cycles — the rhythmic alternation between REM and non-REM sleep — are a fundamental feature of human sleep from infancy through old age, yet how they develop and what drives their timing remains poorly understood. Studying this in infants at scale has historically been difficult, requiring expensive lab-based sleep monitoring. This new research offers a practical, large-scale solution using wrist actigraphy.

The study analyzed over 35,000 hours of actigraphy data from 152 infants measured at 3, 6, and 12 months of age, alongside data from their parents. Researchers applied a metric called Locomotor Inactivity During Sleep (LIDS), previously validated in adults as a proxy for sleep cycle detection, to infant data for the first time at this scale.

Key results showed infant sleep cycles averaged approximately 62 minutes — notably shorter than the parental average of 81 minutes. Between 3 and 12 months, infant cycles lengthened by roughly 10 minutes. This maturation was partially explained by longer continuous sleep bouts, suggesting that consolidated sleep drives cycle elongation. Breastfeeding was associated with longer cycles in both infants (+2.5 min) and their mothers (+6.7 min), a novel and intriguing finding.

Beyond cycle length, the shape of the inactivity waveform also matured — infants showed sharper declines and greater amplitude compared to parents, but these parameters began converging toward adult patterns within the first year, indicating rapid neurological maturation.

These findings have potential clinical implications. Deviations from typical sleep cycle development could serve as early biomarkers for neurodevelopmental conditions. The scalability of actigraphy means this approach could feasibly be deployed in large pediatric cohorts. Caveats include the reliance on movement as an indirect proxy for REM/NREM cycles and the absence of concurrent polysomnography validation in this dataset.

Key Findings

  • Infant sleep cycles average ~62 min at 3 months, about 19 minutes shorter than adult parental cycles of ~81 min.
  • Sleep cycles lengthen by ~10 minutes between 3 and 12 months, tracking with longer continuous sleep bouts.
  • Breastfeeding is associated with longer sleep cycles in both infants (+2.5 min) and their mothers (+6.7 min).
  • LIDS waveform shape — onset, decline rate, and amplitude — begins maturing toward adult patterns within the first year.
  • Actigraphy can detect infant sleep cycle rhythms at scale, enabling large-cohort developmental sleep research.

Methodology

Longitudinal actigraphy study of 152 infants measured at 3, 6, and 12 months of age, generating over 35,000 hours of sleep data. The Locomotor Inactivity During Sleep (LIDS) metric and complementary signal processing techniques were used to identify ultradian cycle patterns. Parent actigraphy was also collected for comparison, providing a within-family reference group.

Study Limitations

This summary is based on the abstract only, as the full paper is not open access. The LIDS metric is an indirect proxy for REM/NREM cycling and has not been concurrently validated against polysomnography in this infant sample. Actigraphy cannot distinguish specific sleep stages, which limits mechanistic interpretation of cycle length changes.

Enjoyed this summary?

Get the latest longevity research delivered to your inbox every week.