Higher physical activity during pregnancy was associated with selected brain-structure measures and fewer parent-reported executive-function challenges in children at age 8. The study did not show that active mothers “raise smarter children,” and it did not test an exercise prescription.
The difference between association and causation is central: families were observed, not randomly assigned to activity levels, and eight years of other influences separated the pregnancy measurements from the childhood assessment.
The Study Followed 69 Mother-Child Pairs
The prospective study included 69 women and their children, comprising 38 boys and 31 girls. During pregnancy, participants wore an accelerometer on the ankle for three to seven consecutive days in each trimester.
Researchers calculated steps, activity counts and time spent sedentary or at light, moderate and vigorous intensities. At age 8, the children underwent MRI scans measuring cortical thickness, surface area, volume and local gyrification.
Parents or guardians also completed the 86-item Behavior Rating Inventory of Executive Function, known as BRIEF. Higher scores on that questionnaire indicate more reported difficulty; they are not IQ scores, school grades or a direct test of intelligence.
Some Brain and Behavior Measures Were Correlated
The researchers found positive correlations between maternal activity measures and cortical surface area, volume or gyrification in selected brain regions. Reported correlation coefficients for those structural findings ranged from 0.35 to 0.54 after correction across brain regions.
Activity in mid-to-late pregnancy also correlated with lower scores for some BRIEF domains, including inhibition, shifting and emotional control. The paper notes that most children in the cohort did not reach a clinical level of executive dysfunction.
These results do not mean that more steps caused a particular brain structure or behavior. Brain measurements are intermediate outcomes, and a statistical difference in a region does not by itself demonstrate better learning, health or daily functioning.
The Design Leaves Major Alternative Explanations
The authors explicitly described their findings as associations rather than causal relationships. The sample was small, limiting further analysis of whether cortical differences could mediate behavioral results.
The researchers adjusted for the child's age, sex, race, body-mass index and childhood trauma, along with parental socioeconomic status. They did not account for every plausible influence, including maternal mental health and the child's own physical activity across the years before assessment.
The study corrected multiple comparisons across brain regions but not across all time points or physical-activity parameters. The ankle monitor could also miss isolated upper-body movement. These limitations increase the possibility that unmeasured factors or repeated testing contributed to the pattern.
A larger study with more complete prenatal and postnatal measurements would be needed to reproduce the findings. A randomized exercise trial would answer a different and stronger question about whether changing activity produces a child-development benefit. It would also need prespecified outcomes and enough participants to distinguish a reproducible effect from chance.
Pregnancy Guidance Comes From Broader Evidence
The American College of Obstetricians and Gynecologists says physical activity is safe and desirable for most people without obstetric or medical complications. It recommends clinical evaluation before an exercise program and careful individual assessment when complications exist.
For healthy pregnancies, ACOG and CDC guidance points to at least 150 minutes of moderate-intensity aerobic activity per week. That recommendation is based on established maternal-health evidence, not on this 69-family imaging study and not on a promise of higher childhood cognitive scores.
The study's useful finding is a signal worth reproducing, not a new standard by which to judge pregnant people or their children. Turning a small observational correlation into a smarter-child promise would distort both the science and the clinical guidance. The next evidentiary step is replication with richer confounder data and outcomes children actually experience, not a more commanding headline.