Newborn face-looking difference needs careful reading
In one study of 130 newborns, girls allocated a larger share of viewing time to a face than boys did, but absolute face-looking time did not differ significantly and broader evidence remains mixed.
The 60-second version
In one experiment, newborn girls allocated a larger share of viewing time to a face than boys did, while absolute face-looking time did not differ significantly.
Key points
- The sample included 130 healthy term newborns tested at an average age of 33 hours.
- Face-viewing share averaged 58.53% in girls and 51.20% in boys, with an effect size of d=0.40.
- The clearest result concerns relative allocation between two videos, not greater absolute face-looking time.
- An independent meta-analysis found no overall sex difference across 20 neonatal face-fixation experiments.
- Hormones, maturation and visual stimulus properties remain untested competing explanations.
Verdict. This is a useful result to replicate, not proof of an inborn social advantage or a forecast for individual children.
A study of 130 healthy newborns found a modest sex difference in how they divided attention between a dynamic face and a mechanical object. Girls devoted a larger percentage of viewing time to the face, but girls and boys did not differ significantly in their absolute face-looking time.
ExperimentWhat the newborns saw
Researchers recruited 67 boys and 63 girls born at term in Cambridge, UK. Their mean age was 33 hours. On a split display, each baby saw a standardized video of a woman's face alongside a moving mechanical mobile; most completed one 70-second trial.
Newborn eye tracking is technically difficult, so three trained observers coded recorded gaze direction. The coders did not know the baby's sex or which side contained each stimulus. Agreement was excellent for the percentage measure and good for absolute looking times.
ResultThe difference appeared in relative allocation
The percentage difference was statistically significant: t(128)=2.31, p=0.023. It remained after adjustment for age at testing, gestational age, delivery mode, birth weight, recent sleep and feeding, trial length, number of trials, and time looking away.
DistinctionPercentage and seconds answer different questions
| Share of viewing time | Girls allocated 58.53% to the face and boys 51.20%; the between-group difference was significant. |
|---|---|
| Absolute face time | Girls averaged 33.20 seconds and boys 30.94 seconds; the between-group comparison was not significant after adjustment. |
| Within each group | Girls looked longer at the face than the object; boys showed no significant face-versus-object preference. |
| Meaning | The strongest evidence concerns the distribution of attention between two competing videos, not greater absolute face attention in girls. |
A relative preference is not the same claim as spending significantly more seconds looking at faces.
ContextThe wider literature is not settled
An independent systematic review covered 31 studies, 40 experiments, and 1,936 newborns. Across 20 experiments on face fixation, it found no significant overall sex difference and concluded that existing evidence did not support a specific social advantage for girls at birth.
The new experiment used a particular contest between natural face motion and predictable mechanical motion. Its result may therefore be task-dependent. It adds evidence worth testing, but does not overturn the broader literature by itself.
LimitsThe design cannot identify the cause
- 1. One task and location: all babies came from one Cambridge hospital, and each contributed only one usable trial.
- 2. Manual gaze coding: blind coders agreed well, but direct newborn eye tracking would provide more detailed measures.
- 3. Stimuli differed in several ways: motion, complexity and other low-level visual features could matter, not only social content.
- 4. Mechanisms were not measured: prenatal hormones were proposed but not tested; maturation and perceptual explanations remain possible.
- 5. No developmental forecast: the study did not follow children or connect the task to later empathy, autism, personality or ability.
NextReplication must separate sex from stimulus
The next step is a preregistered, multisite replication using several faces and objects, repeated trials and improved gaze tracking. Mechanistic claims require direct measurements. For now, the accurate conclusion is that two groups allocated attention differently on average in one experiment, with substantial overlap and no basis for labeling an individual child.