Babies Rarely See Exaggerated Facial Expressions Used in Emotion Research, Study in Communications Psychology Finds

Babies aged 6 to 17 months rarely see the exaggerated facial expressions — the toothy grin, the deep scowl — that scientists have long used to test emotional understanding, a study published in September in Communications Psychology found, a result that questions how infant emotion research is designed.

The study, led by University of Kansas researchers Katie Hoemann and Jeffrey M. Girard with collaborators at Rutgers University, Northeastern University, Massachusetts General Hospital and the University of California, Davis, analyzed monthly, hourlong home recordings of infants, categorizing every face and expression the babies saw.

What the team found was stark. Exact matches to the facial configurations hypothesized to be evolutionarily conserved, universal expressions of particular emotions — what researchers call “prototypes” — were genuinely rare in everyday life. Even happiness, the most common, appeared in only a small fraction of the time babies spent looking at full faces. And full faces themselves were scarce: babies saw one just 12% of the time the researchers measured.

That matters because decades of developmental research — and the emotion-literacy posters hanging in pediatric waiting rooms — rest on the assumption that these exaggerated expressions are reliable, frequent signals in an infant’s world. If they are not, the study suggests, babies must be learning emotion some other way.

Babies learn emotions from everyday interactions

  • The exaggerated “happy,” “angry” and “sad” faces used in lab studies of babies almost never occur in real homes.
  • Babies spend most of their time seeing partial faces, varied expressions, or no faces at all — not the clean, textbook versions.
  • Emotional learning appears to draw on words, gestures, touch and context, not facial configurations alone.
  • Caregivers need not perform exaggerated expressions; everyday interaction is doing the developmental work.

What the home recordings actually showed

The researchers did not collect new footage. They used an existing data set of monthly, hourlong recordings made in infants’ homes between 6 and 17 months of age, then coded what babies were exposed to — full faces, partial faces, and facial movements.

“When we looked for exact matches to facial configurations that have been hypothesized to be evolutionarily conserved, universal expressions of particular emotions, we found that it was really rare,” Hoemann, assistant professor of psychology at KU, told Medical Xpress.

“Happiness was relatively the most common,” she said. “But even if we look at the share of all full-face time — any moment an infant was seeing a full face — it’s still only 12% of the time.”

In other words, even within the narrow slice of an infant’s day spent looking at a whole face, the standardized expressions that populate lab stimuli and textbook diagrams were the exception, not the rule.

Why “basic emotion theory” is being tested

Much of the field’s work on infant emotion rests on what researchers call basic emotion theory — the idea, as Hoemann explained it, “that specific facial configurations were evolutionarily conserved as ways of communicating particular emotion-relevant signals among humans.”

From that assumption follows a familiar research toolkit: happiness means a big toothy smile, anger a big scowl, sadness a super-exaggerated frown. “And these are the kinds of stimuli that are used in tests of emotion perception for people of all age groups, for the most part,” she told Medical Xpress.

The new findings challenge the prevalence half of that assumption. If infants rarely encounter these configurations, then their eventual emotional competence — becoming, in Hoemann’s phrase, “emotionally functional culture members” — cannot be explained by repeated exposure to them alone.

What infants actually see What the study found
Full faces, any expression Only 12% of measured time
Exact “prototype” expressions (e.g., big toothy smile) Really rare; happiness the most common
Everyday facial movements Highly varied; same configuration can mean different things in different contexts

The same face can mean different things

A second problem compounds the first. The facial movements babies do see are not tidy one-to-one signals. “The same collection of facial movements that you would make can mean different things in different contexts,” Hoemann said. “But also, in relation to the same context — like being happy or angry — you can make lots of different combinations of facial movements.”

Hoemann’s framing draws on theoretical frameworks in which infants learn from their environment broadly — what she calls their “information ecology.” Facial expressions arrive bundled with words, gestures, movement, touch and other sensory information, in emotional contexts that change moment to moment.

“It’s not that they never happen, but they’re not capturing all of this beautiful heterogeneity that is really out there,” she said. “Instead, we should think about the process of emotion and the process of learning emotion as deeply contextually embedded.”

Caregiver language as the next clue

Hoemann’s current work, led by Shannon M. Brady of Rutgers University, examines whether infants latch onto emotion words from caregivers — labels or clear references such as “gross” or “yucky” — as salient cues that tag emotional moments.

“These kinds of properties of associative and integrative learning have been widely shown to drive learning in general, and also for infants,” Hoemann said. “We’re really applying these principles now to looking at emotion-relevant information.”

The open question she poses is which pieces of a messy environment infants actually use. “Clearly, infants and children are doing it, but the question is: What pieces are they leveraging to be able to do that?” Hoemann said.

What this changes for research and for parents

For researchers, the implication is methodological: studies of emotion perception should account for a wider range of information than facial expressions alone, and stimuli should reflect the variability of real faces rather than distilled prototypes. For clinicians who use emotion-recognition materials in developmental assessment, the findings are a prompt to treat simplified face sets as tools with known limits, not windows onto everyday experience.

For parents, the message is quietly reassuring. Mugging for the baby is fun, Hoemann noted, but caregivers should recognize that emotional learning happens in a much wider context. “When we see these emotion-literacy posters, where people are smiling, and people are scowling, realize that those aren’t representative of what’s going on in our everyday lives,” she said.

Talk to a pediatrician about developmental concerns

This study concerns developmental research methods, not a treatment, and carries no medical contraindications. It gives no basis for alarm about a baby’s emotional development: the researchers state plainly that infants and children do become emotionally competent, and the study addresses how, not whether.

Caregivers who have specific concerns about an infant’s social or emotional development — for example, limited eye contact or reduced responsiveness to faces — should raise them with a pediatrician, who can use validated developmental screening. That guidance comes from standard pediatric practice, not from this study, which did not evaluate individual children.

Where the work goes from here

The paper, “Infants’ facial input from 6 to 17 months rarely includes hypothesized expressive emotion prototypes,” appeared in Communications Psychology on September 24, alongside coauthors Marissa Ogren, Lauren Leotti and Vanessa LoBue of Rutgers; Lisa Feldman Barrett of Northeastern University and Massachusetts General Hospital; and Lisa M. Oakes of the University of California, Davis.

The next step, already underway, is to pair facial exposure data with the language infants hear — testing whether emotion words serve as the tags around which babies organize what they see. As Hoemann put it, the goal is to ask how infants “might leverage that to learn what they ultimately do.”

References

  • Hoemann, K., Girard, J. M., Ogren, M., et al. “Infants’ facial input from 6 to 17 months rarely includes hypothesized expressive emotion prototypes.” Communications Psychology (2026). https://doi.org/10.1038/s44271-026-00539-z
  • Medical Xpress. “Babies rarely see exaggerated facial expressions used in emotion research, study finds.”
Photo of author

Health Desk

Health Desk at Archyde.

John Egan Replaces Injured Nathan Collins for Ireland vs Kosovo

Sydney Swans AFLW Set Record Attendance as Fans Support Club