Surprising new research reveals how fetal brain complexity declines before and after birth

Fetal Brain Complexity: New Research Challenges Expectations

Even as birth nears, the complexity of brain activity decreases with fetal maturation and continues to decline after birth, according to results recently published in Nature Mental Health by a research team with key contributors from the University of Tuebingen, Germany and led by Dr. Joel Frohlich, who also serves as a research consultant for the Institute for Advanced Consciousness Studies in addition to his postdoctoral position in Tuebingen.

Sex Differences in Fetal Brain Development: A Surprising Finding

More surprising still, male and female fetuses show different changes in complexity, with boys declining faster than girls. According to the authors, the study’s findings carry important implications for future efforts to develop predictive biomarkers of psychiatric disorders based on the complexity of brain activity recorded before birth.

Investigating Fetal Brain Activity with MEG

The published manuscript, titled “Sex differences in prenatal development of neural complexity in the human brain”, was sparked by a two-year interdisciplinary collaboration between the laboratories of University of Tuebingen professors Alireza Gharabaghi and Hubert Preissl, with Frohlich serving liaison between the respective labs. Prof. Gharabaghi leads Tuebingen’s Institute for Neuromodulation and Neurotechnology, whose mission to develop new brain stimulation techniques led the institute to consider how sensory stimulation could be safely applied to investigate neural integrity in fetuses and infants, much in the same manner as electromagnetic brain stimulation is currently used in adults. When the collaboration began, Prof. Preissl, director of Tuebingen’s fMEG Center, had already supervised relevant experiments conducted by Dr. Julia Moser (now at the University of Minnesota) and Dr. Katrin Sippel, which gave the team the data they were looking for.

Fetal brain activity has already revealed evidence of learning before birth

The team used magnetoencephalography or MEG (a technology for detecting weak magnetic activity in the brain) to record neural signals non-invasively from third trimester fetuses and newborns. These signals were responses to sequences of auditory tones, which included patterns that were occasionally broken to test if fetuses (and, later, newborn infants) had successfully learned the original sequence. Two earlier research manuscripts, led by Moser, examined these data in the context of fetal and newborn learning and revealed evidence that fetuses respond to pattern violations as early as 35 weeks gestation, late in the third trimester of pregnancy. “Sensory stimulation provides us with a unique opportunity to observe how young brains process information from the outside. And all in a completely safe way,” explained Prof. Preissl in a recent press release.

In the current manuscript, the research team then applied several different algorithms to estimate the complexity of the MEG signals using entropy, or the number of possible ways in which states of the signal can be arranged. Some of these algorithms work by determining how difficult it is to compress the signal, as more complex data are harder to compress. To understand this better, consider the compressed size of two image files on your computer, one depicting a Vermeer painting and the other depicting a Rothko painting. The more complex image (Vermeer) will be harder to compress into a small file. Similarly, more complex brain signals are harder to compress, allowing scientists to estimate their entropy.

Surprising results from fetal brain activity

The researchers originally hypothesized that fetal brain activity would grow more complex with maturation. To their surprise, the opposite occurred. “Intuitively, I had thought that as the brain matures, its activity should grow more complex just as its anatomy and function grows more complex,” said Frohlich. “In hindsight though, it makes a decent amount of sense, especially considering the fact that we recorded brain activity evoked by sensory signals, rather than spontaneous activity.” As the brain develops, it moves away from random patterns toward more ordered modes of activity sculpted by emerging synaptic connections. These connections constrain the number of ways in which the brain can respond to stimuli such as the auditory patterns in the experiment. This is likely why more mature fetal brains showed less complex activity in their responses: these brains had fewer ways of responding to the same stimulus, and thus lower complexity. According to Frohlich, if the experiment had instead looked at spontaneous brain activity in the absence of stimuli, the results might have been different.

Crucial to the team’s conclusions were contributions from Imperial College London professor Pedro Mediano, who shared a sophisticated mathematical algorithm which allowed the team to determine which properties of the fetal brain signals were driving the decrease in their complexity. Using Mediano’s approach, the team found that changes in the amplitude or “strength” of the signal were related to the decreasing complexity. In fact, the effect of amplitude appeared to mask changes caused by another property of the signal, phase, which opposed complexity by driving increases in complexity with maturation. The presence of two opposing processes might partially explain the team’s surprising results with respect to maturation.

However, the effects of fetal sex are still leaving the team slightly puzzled. “I didn’t expect fetal sex to have any impact on neural complexity,” said Frohlich, “but it’s possible that this relates to the greater vulnerability of the male brain during gestation, as many neurodevelopmental disorders, like autism and ADHD, are diagnosed more frequently in boys.” Frohlich is inspired by earlier studies that have linked the complexity of neural activity to brain health, including one previous study which predicted the onset of autism years later from brain activity recorded in young infants. Neurodevelopmental disorders such as autism are best prevented very early in life while the brain is still highly plastic, which creates a need for early detection of risk. “The earlier we identify the risk of developing neuropsychiatric and metabolic disorders, the more effectively we can support brain development to prevent serious illness,” explained Prof. Gharabaghi in a press release.

Looking toward the future

According to Frohlich, whose 2018 PhD dissertation focused on biomarkers of neurodevelopmental disorders, the next step in this line of work will be to follow fetuses several years after birth to see whether the complexity of their fetal brain activity predicts later outcomes such as autism or ADHD. “For example, you could recruit pregnant women who already have a child with autism, which means that the fetus has some familial risk of also developing it. By recording brain activity from the fetus and then following up with the family three years later, you could see if the complexity of prenatal brain activity is predictive of an eventual diagnosis.”