New research from the Washington University School of Medicine in St. Louis suggests that as children mature, the way their brains represent the emotions of other people becomes more similar to other children their age.

Published this spring in Nature Neuroscience, this study marks the largest effort to date to examine how children respond to emotional cues presented in videos. The researchers argue that video content provides more context than static images and serves as better proxy for cues a child may experience during real-world interactions.

Minions and moods

The team sought to distinguish whether the concepts, described by the authors as “computational shortcuts”, that help a child interpret and respond to another person’s emotions become more alike as they stabilize across development or whether they diverge based on an individual’s personal experiences,

To this end, they characterized all the emotional content present at each timestamp of a short animated clip called The Present as well as in scenes from the film Despicable Me. Over 500 children between the ages of five and fifteen watched these clips while their brain activation was measured by functional magnetic resonance imaging (fMRI), which quantifies blood flow in different regions of the brain.

The authors used artificial intelligence to distinguish activation patterns associated with emotional content at each time stamp. Researchers found that data from the the temporo-ventromedial prefrontal cortex, the cerebellum, and the auditory, visual, cingulo-opercular, dorsal attention, default mode, frontoparietal, medial parietal, somatomotor-mouth, and salience networks were most useful in helping their model classify emotional activation patterns, meaning that these structures and networks active as children processed the emotions portrayed in the video clips.

While there was only a modest change in activation patterns elicited by different emotional cues between participants of different ages, activation patterns among older children were more similar to each other than those among younger children. The researchers also concluded that brain activation patterns were most alike in response to more intense emotional cues.

Older children had most similar brain activation patterns in response to scenes that depicted negative emotions. Unlike the younger children, this was true for scenes in which the emotional content relied on contextual information about the story. The authors attribute this synchronicity to convergent involvement of the default mode network, a set of structures believed to facilitate resting state awareness, in emotional processing at later stages of development,

Happiness is the truth?

Over the years, scientists have developed several theories about how humans understand emotions. The Washington University team used these theories to contextualize their findings. They note that their results could support cognitive models of emotional processing, in which emotions are believed to be encoded as broad concepts rather than relegated to a specific part of the brain. The authors discuss their findings through the lens of one particular cognitive model, the social information processing model, which holds that many different domains of cognition are involved in interpreting emotions, from detecting cues to placing those cues and information within one’s own experience. They argue that the widespread activation seen in the brain in response to each emotion lends credence to this integrated model.

Another implication of this research is the likely existence of discrete “basic” emotions shared by all people that are represented in the brain in similar ways. The authors suggest that the relatively meager change in activation patterns in response to emotional cues across different age groups could indicate that these emotional concepts are innate and are only refined during development. The researchers attribute this refinement to the experiences a person acquires during childhood rather than biologically programmed brain changes due to puberty because compared to studies in which children were presented with static images, younger children interpreted emotional cues more like older children.

The authors note that takeaways from this study are limited because the experiment is cross-sectional meaning kids from all age groups were studied at one time, rather than a longitudinal experiment in which a cohort of children is studied multiple times over many years. The latter study design would allow researchers to characterize changes in an individual’s emotional processing as they age rather than drawing inferences based on different age groups.

While the conclusions that can be drawn from fMRI research are a subject of ongoing scientific debate, these findings cast a novel light on the influences that shape a person’s response to the emotions of others.