How Babies Spend Their Day Shapes Motor Skills and Growth, Study Finds

The first years of life are a whirlwind of transformation. In the span of just forty-eight months, a human being goes from being unable to lift their own head to running, jumping, negotiating, and asserting preferences with startling determination. During this window, a significant portion of the brain’s structure and capacity is established, and the foundations of lifelong health are laid. A new study from the Netherlands, published in the Journal of Activity, Sedentary and Sleep Behaviors, now offers one of the most detailed looks yet at how the way toddlers and infants divide their twenty-four hours between movement, stillness, and sleep relates to their physical growth and developmental progress. The findings are nuanced, occasionally counterintuitive, and they expose just how much uncertainty still surrounds the official guidelines that parents are told to follow.

The research, led by Jelle Arts of Amsterdam UMC and colleagues, forms part of the My Little Moves study, a longitudinal observational cohort conducted across the Netherlands. Rather than treating physical activity, sedentary behavior, and sleep as isolated variables, the team embraced a concept that has been gaining momentum in the field: the twenty-four-hour movement composition. Because a child’s day is finite, every minute spent in one behavior is a minute taken from another. Analyzing these behaviors separately, the researchers argue, misses the essential trade-off structure of a young child’s day. Their solution was compositional data analysis, a statistical framework that treats the full daily allocation of time as a single, interdependent whole.

Data collection relied on an unusual tool: a smartphone application called the My Little Moves app, designed specifically for this cohort. Parents recorded their child’s daily activities for seven consecutive days using a time-use diary format, selecting from categories such as playing, sleeping, screen use, passive transport, and personal care. Depending on the activity chosen, the app prompted parents for additional details about intensity, posture, and context. Crucially, the app adapted its content to the child’s developmental stage, asking about motor milestones like crawling, sitting without support, and walking. Parents typically spent between ten and thirty minutes per day completing the diary, and preliminary validation against accelerometer data suggested the method yields reliable results when activities are reported for at least two full days.

The study drew on two sub-cohorts, each followed at two timepoints roughly nine months apart. One sub-cohort focused on social-emotional development, providing complete data for 101 children at baseline, whose average age was just under twenty-one months, and 62 children at follow-up. The other focused on gross motor development and growth, with 60 children at baseline and 46 at the second timepoint. Development was assessed with the Dutch version of the Bayley Scales of Infant and Toddler Development, Third Edition, a widely validated instrument. Researchers also measured children’s weight and height directly, using calibrated scales and stadiometers, to compute body mass index z-scores relative to World Health Organization growth standards.

To represent the movement behavior composition statistically, the team calculated isometric log-ratios, transformed coordinates that capture how much time one behavior occupies relative to the others. These coordinates were then entered into linear regression models for the cross-sectional analysis and linear mixed-effects models for the longitudinal analysis, with the child’s age and sex included as covariates. The mixed-model approach, which uses maximum likelihood estimation, allowed the researchers to include children with data from at least one timepoint without resorting to imputation of missing values. Outliers were screened and removed, and sensitivity analyses confirmed that the handling of influential measurements did not drive the main findings.

The headline result concerns gross motor development. The composition of the twenty-four hours was significantly associated with raw gross motor scores both cross-sectionally and longitudinally, with p-values below .001, although the additional variance explained by the movement composition was modest, at roughly four percent. When the researchers examined which behavior mattered most, a clear pattern emerged: more sedentary time, relative to physical activity and sleep, was linked to lower motor scores, while more sleep, relative to the other behaviors, was linked to higher scores. Physical activity itself, relative to the rest of the day, did not show a significant independent association, a finding that distinguishes this study from earlier work in preschoolers where moderate-to-vigorous activity often took center stage.

The most practically striking numbers came from the compositional isotemporal reallocation analysis, a technique that estimates what would happen to developmental scores if fixed slices of time were shifted between behaviors. Reallocating ten minutes from sedentary behavior to physical activity increased predicted raw gross motor scores by 0.22 points, with a 95 percent confidence interval of 0.11 to 0.33. Shifting the same ten minutes from sedentary behavior to sleep produced an even larger gain of 0.27 points. Conversely, sacrificing ten minutes of sleep in favor of sedentary time cost 0.26 points. In a sample where children spent roughly 55 percent of their day asleep, 30 percent in sedentary pursuits, and only about 14 percent in physical activity, these small per-minute effects accumulate meaningfully across a day, a week, and a childhood.

The growth findings were more surprising. The movement composition was associated with BMI z-scores in the cross-sectional analysis, explaining a medium-sized thirteen percent of additional variance, but the direction of the effects ran counter to expectations. Higher relative sleep was associated with higher BMI z-scores, and higher relative sedentary behavior with lower ones. Shifting ten minutes from sleep to sedentary behavior lowered predicted BMI z-scores by 0.04, while the reverse shift raised them by the same amount. The authors caution that these changes are small and may not be clinically relevant, and they note that the sample’s mean BMI z-scores of 0.7 to 1.0 were likely inflated by the light clothing children wore during measurements, for which no standardized correction exists in this age group. The apparent paradox may also reflect the fact that in very young children, greater physical activity can build fat-free mass, complicating any simple reading of BMI.

Perhaps the most sobering result was the null finding. The composition of movement behaviors showed no significant association with social-emotional development, whether measured as raw or norm-referenced scores, in either the cross-sectional or longitudinal analyses. This mirrors earlier compositional studies by Kuzik and colleagues and St. Laurent and colleagues, which also failed to detect links. The researchers suggest that the social context of activities may matter more than the broad behavioral category: being read to by an adult, for instance, is technically sedentary but may nurture emotional development, whereas unmonitored screen time, also sedentary, may do the opposite. Disentangling these qualitative differences is a task for future research.

The study’s limitations are candidly acknowledged. Dropout was considerable, with complete data available for only about half of the social-emotional sub-cohort at baseline, and the majority of participating parents were highly educated, limiting generalizability. The reliance on parent-reported diaries introduces potential recall and social desirability bias, and parents cannot report activities that occur at daycare. The nine-month follow-up interval may have been too short to capture meaningful developmental change, and limited statistical power meant the models could adjust only for age and sex, not for ethnicity, parental education, or income. Still, the consistency of the motor development findings across both cross-sectional and longitudinal analyses lends them weight. The authors conclude that evidence on the optimal distribution of movement behaviors in early childhood remains inconclusive, and that larger longitudinal studies, ideally incorporating objective measurement and longer follow-up, are essential before the WHO’s twenty-four-hour guidelines for the youngest children can rest on firmer empirical ground. For now, the message for parents is a modest but actionable one: in the economy of a toddler’s day, minutes traded away from stillness toward active play and sleep appear to buy measurable gains in motor development.

Subject of Research: Associations between 24-hour movement behaviors and growth, motor, and social-emotional development in children aged 0 to 4 years

Article Title: Cross-sectional and longitudinal associations between 24-hour movement behaviors and growth, motor, and social-emotional development in early childhood

Article References: Arts, J., Altenburg, T. M., Lettink, A., Verhoeff, A. P., Gubbels, J. S., & Chinapaw, M. J. M. (2025). Cross-sectional and longitudinal associations between 24-hour movement behaviors and growth, motor, and social-emotional development in early childhood. Journal of Activity, Sedentary and Sleep Behaviors, 4(1), Article 14. https://doi.org/10.1186/s44167-025-00085-9

Image Credits: AI Generated

DOI: 10.1186/s44167-025-00085-9

Keywords: early childhood, physical activity, sedentary behavior, sleep, motor development, social-emotional development, BMI, compositional data analysis, Bayley Scales, 24-hour movement guidelines, longitudinal study, My Little Moves