Pregnancy and child

Babies 'detect emotion in voices'

“Babies can tell sad voices at 3 months,” according to The Daily Telegraph. The newspaper reported that brain scans have shown that parts of the brain “light up more when babies hear sad voices”.

A sample of 21 babies, aged three to seven months old, was given a special type of MRI scan to measure the activity in different regions of the brain. While asleep, they listened to human speech and also to various “non-speech vocalisations”, which were each made to sound emotionally neutral (such as coughing), happy (laughing) or sad (crying). The scanner revealed that, in babies, an area of the brain called the temporal cortex is very sensitive to voices, much as it is in the developed brains of adults. The researchers also noticed that, while there was little difference in brain activation caused by neutral and happy sounds, sad emotions activated slightly different areas of the brain. This suggested that the ability of the brain to process human voices and negative emotions happens very early on in life.

This is an interesting biological study of brain activation in young babies in response to human sounds, but limited conclusions can be drawn from this study alone. It is not known from this study whether the brain is activated differently when the baby is awake or asleep, when the vocalisations come from different people (for example, whether the baby’s brain responds differently to crying from a stranger or a parent), or when listening to more complex, emotionally charged speech (such as an argument). Also, the study cannot tell us whether exposure to different emotional sounds has any influence on a baby’s development or personality.

Where did the story come from?

The study was carried out by researchers from King’s College, University College London and Birkbeck College. It was funded by a number of UK institutions, including the Medical Research Council, the National Institute for Health Research, Maudsley NHS Foundation Trust and the Institute of Psychiatry, King’s College London. The study was published in the peer-reviewed scientific journal Current Biology.

The Daily Mail and The Daily Telegraph both reflected the findings of this scientific research.

What kind of research was this?

This study used brain scans to identify which areas of the brain were activated in babies when listening to various adult vocalisations, such as speech, laughter and crying noises.

The researchers say that human voices play a principal role in social communication and that various specific regions of the brain are involved in processing the emotional content of voices. However, it is still unknown at what stage of growth a person will develop this specialist ability. For example, previous brain imaging studies in babies have suggested that, unlike adults, the infant temporal cortex (an area at the side of the brain responsible for processing sounds) is not able to differentiate speech from music. However, other studies have suggested that the temporal cortex in babies is able to identify speech but that the area of the temporal cortex that carries out this function is in a slightly different location during infancy. Another area of uncertainty is which specific areas of the brain are involved in processing non-speech human sounds and vocalisations such as laughter - a subject that is the focus of this experimental research.

The research had two specific aims: to determine whether the temporal cortex of babies shows specialisation for human voices, and to determine which brain areas are activated when babies process non-speech vocalisations, indicating their ability to perceive emotions in these sounds.

What did the research involve?

The study involved 21 babies aged three to seven months. While they were sleeping naturally, the babies were placed in a functional magnetic resonance imaging (fMRI) scanner. This is a special type of MRI scan which is able to detect the extent of blood flow in different areas of the brain. It is based on the principal that increased nerve cell activity is associated with increased blood flow, and can identify activity in specific regions of the brain. The babies were presented with speech and three types of non-speech vocalisation: emotionally neutral (such as coughing or sneezing), emotionally positive (laughing) and emotionally negative (crying). They were also assessed when listening to a series of non-voice environmental sounds that they may have been familiar with (such as water splashing and toy sounds).

The researchers looked at the patterns of brain activation on fMRI imaging when the babies listened to speech and non-speech vocalisations, as well as whether there were any differences between the different emotionally charged vocalisations.

What were the basic results?

Imaging revealed that, compared to hearing no sound at all, any sound produced significant activation in the five brain regions at the side, front and back of the brain. This was consistent with the findings of hearing studies in other babies, children and adults. The researchers found that, compared to environmental sounds, neutral emotional vocalisations caused greater activation in the temporal cortex on the right side of the brain and at a region at the front of the brain. Environmental sounds tended to cause more activation of a region of the temporal cortex on the left side of the brain.

When comparing human voices to environmental sounds, the researchers noticed a greater contrast in brain activation caused by these two types of sounds in older babies. This suggests that the ability of this area of the brain to perceive different types of speech increases with age. No difference was observed between brain activation for happy vocalisations compared to neutral vocalisations, but sad vocalisations were observed to cause greater activation of two different regions towards the front of the brain (the insula and gyrus rectus).

How did the researchers interpret the results?

The researchers concluded that the temporal cortex is a strong voice-sensitive region of the brain in young babies. They say that the babies show activation in the regions at the front of the temporal cortex in response to vocalisation, which is similar to adults. However, sad vocalisations cause activation of different regions of the brain, suggesting that the ability to process human voices and negative emotions develops very early in life.


This study furthers our understanding of the brain’s biology and which areas of a baby’s brain are activated by speech and by different emotionally charged vocalisations. It will be of interest to the scientific and medical community and will contribute to the findings of similar studies conducted in this area. As expected for this type of experimental study, the sample of babies was small and, therefore, the study is more likely to produce chance results than studies in more participants.

All babies were sleeping during the study, which is understandably more feasible and ethical when placing young babies in a scanner, and removes the effect that a baby’s own crying and distress would have on brain activity. It is not known whether there is a difference in brain response when the baby is awake compared to asleep, particularly because a baby would also respond to visual stimuli, such as a happy or sad face, when awake.

Additionally, the study does not specifically state who made the vocalisations, but it is assumed that this would be researchers or study volunteers. Therefore, it is not known whether vocalisations from different people would produce different results, such as whether the baby’s brain responds differently to crying from a stranger or a parent. Also, it is not known whether babies’ brains would respond differently to differences in pitch and tone, such as male and female voices, or to differences in volume or frequency, such as differences in the sound quality of sneezing compared to crying or laughing.

Further to this, all the emotional sounds were basic non-speech vocalisations such as coughing, laughing or crying, and whether there is any difference in brain response when listening to more complex emotionally charged speech (e.g. an argument) cannot be determined from this study.

While this interesting, small study suggests that the brains of young babies can distinguish between different non-speech vocal cues, it looked at immediate responses within the brain. It cannot tell us whether exposure to different emotional sounds could influence a babies’ development or personality. Therefore, we do not know what long-term effect, if any, regular exposure to certain emotions may have.

NHS Attribution