Depression during pregnancy is common. Somewhere between one in five and two in five pregnant people report elevated symptoms.

For decades, studies have linked it to differences in how children develop, from emotional difficulties to a higher risk of mental health problems later. But almost all of that evidence is correlational. Mothers who are depressed often also face poverty, stress or other hardships, so it is hard to tell what is actually doing the damage.

A new study takes a rare step closer to cause and effect. It looked at the brains of babies whose mothers had been randomly assigned to depression treatment during pregnancy.

The babies of treated mothers had a smaller right hippocampus. And in these babies, smaller looked like a good sign.

The study followed babies from a randomized trial of therapy in pregnancy

The study builds on the Care Project, a randomized trial in the Denver area led by researchers at the University of Denver, the University of Colorado Anschutz and colleagues.

The original trial enrolled 234 pregnant participants with elevated depression symptoms. They were randomly assigned to one of two groups:

  • Brief interpersonal psychotherapy: eight weekly 50-minute sessions focused on relationships, social support and conflicts in daily life
  • Enhanced usual care: normal prenatal care plus a one-hour education session about depression and information on mental health resources

The therapy worked. In the therapy group, the share of participants with major depression fell from about 37% to 6% by the end of pregnancy. In the comparison group, it fell from about 37% to 26%.

For the new follow-up, the researchers studied 109 of the babies: 52 from the therapy group and 57 from the comparison group.

  • At about six weeks old, each baby had a brain MRI while naturally asleep.
  • At six months, babies watched a moving, noisy toy robot in the lab, and researchers rated how upset they became: facial distress and crying. This measures negative emotionality, an early temperament trait linked to later risk of anxiety and depression.

Babies in the therapy group had a smaller right hippocampus

Right hippocampus: smaller in the therapy group. The difference was moderate in size (Cohen’s d = 0.62, p = .002). It held after accounting for overall brain size, age at scan, sex, birth order, maternal health, medication use and the mother’s depression after birth.

Left hippocampus and amygdala: no clear difference. The left hippocampus pointed the same way but wasn’t statistically significant. The amygdala, a region involved in fear and emotion, showed no difference.

Smaller hippocampi went with calmer babies. Across all infants, a smaller hippocampus at six weeks was linked to less distress at six months. The link was modest for the right side (r = 0.19) and stronger for the left side (r = 0.32).

To put those correlations in everyday terms: the right hippocampus explained only about 4% of the variation in how upset babies got, the left about 10%. Most of what makes one baby fussier than another lies elsewhere.

In babies, smaller isn’t bad news

This is the surprising part. In adults, a smaller hippocampus is often bad news, linked to depression, chronic stress and PTSD.

But a baby’s brain is not an adult brain. In infancy, the hippocampus is growing fast. The question isn’t just how big it is, but how fast it is getting there.

One leading idea is the stress acceleration hypothesis. When a developing brain senses a stressful environment, it may grow up faster, like a child forced to take on adult responsibilities early. That might help in the short run. But racing ahead may shorten the window when the brain is most flexible and able to learn, and it has been linked to more negative emotionality and later depression.

Some earlier studies found that babies exposed to prenatal depression had larger hippocampi in early infancy. Seen this way, the smaller volumes in the therapy group may mean the brain was developing at a more typical pace, not that it was underdeveloped.

The authors present this as one possible interpretation, not an established fact. The infant research is inconsistent, and studies have found larger, smaller and unchanged volumes.

The hippocampus is packed with stress-hormone receptors

The hippocampus is best known for memory, but it also helps regulate emotion and the body’s stress response.

It is also packed with receptors for stress hormones such as cortisol. During pregnancy, a mother’s stress system and the placenta both produce stress-related hormones, some of which reach the developing baby. That makes the hippocampus one of the regions most likely to “notice” what the prenatal environment is like.

How the therapy reached the baby is still a puzzle

You might expect the mothers whose depression improved most to have babies with the biggest brain differences. That’s not what the researchers found. How much a mother’s depression, anxiety or stress changed during pregnancy didn’t predict her baby’s brain volume.

So the pathway is unclear. The authors suggest several possibilities:

  • the study may have been too small to detect that link
  • therapy may have changed stress hormones or inflammation in ways questionnaires don’t capture
  • it may have improved sleep, diet or activity
  • it may have strengthened social support, reducing everyday stress

The study couldn’t test any of these directly.

What the study doesn’t show

This is a peer-reviewed study with an unusually strong design for this question, but the findings are preliminary:

  • The brain results were exploratory. The trial was designed to test whether therapy reduced depression, not to measure babies’ brains.
  • It’s a relatively small sample, 109 babies, fewer than half of the original trial. Some were lost to follow-up, partly because of COVID-19 shutdowns. The researchers found no demographic or clinical differences between babies who were scanned and the full trial.
  • One side only. The difference showed up in the right hippocampus but not the left, so the authors caution against reading much into which side.
  • One scan, at six weeks. It doesn’t show whether the difference lasts, grows or disappears.
  • Six months is very early. The study doesn’t show that these babies will have better mental health as children or adults.
  • One US city, English speakers only.

Treating depression in pregnancy may help the baby too

The case for treating depression during pregnancy already rests on the mother’s own wellbeing. This study adds a possible, still unproven, benefit for the baby.

As a reasonable inference:

  • If you’re pregnant and have felt low, hopeless or unable to enjoy things for more than a couple of weeks, tell your midwife, obstetrician or doctor. Depression in pregnancy is common and treatable.
  • Brief talk therapy, like the eight-session programme in this trial, is one effective option.
  • This study is not a reason to choose one treatment over another based on what it might do to a baby’s brain volume. That decision belongs with your care team.

A randomized trial moves the evidence beyond correlation

Most of what we know about mothers’ mental health and babies’ development comes from watching what happens. This study, by building on a randomized trial, moves a step closer to testing whether helping the mother changes the baby’s development.

The answer, for now, is a cautious “possibly”. The hippocampus at six weeks isn’t a verdict on a child’s future, but it hints that pregnancy may be one of the earliest moments to protect two people’s mental health at once.