"What was surprising is that metabolism isn't just a passive thing that happens in the background," said Bhaduri, who is a member of both the UCLA Broad Stem Cell Research Center and the UCLA Health Jonsson Comprehensive Cancer Center. "It can really control how stem cells make decisions."

The results, while preliminary, could eventually help scientists examine how maternal nutrition, metabolic disorders, and other environmental influences affect the developing brain. The metabolic atlas itself also stands as one of the most detailed resources yet for studying human brain development.

Why do thalamic fibers arrive so early?

The second study, published in Science and led by first author Claudia Nguyen, examined a completely different source of developmental information: the thalamus, a structure deep inside the brain that relays information throughout the nervous system.

Scientists have known for years that thalamic neurons send long, wire-like projections toward the cortex, where they eventually connect with specific neurons. But anatomical studies show that in humans, these fibers reach the cortex long before those final connections form. That timing raised a question: why arrive so early?

Working with human stem cell-derived brain "assembloids" — lab models that combine multiple brain-region-like structures — the UCLA team found part of the answer. The thalamic projections physically touch radial glia while the brain is still developing.

That contact changed the stem cells' behavior. It pushed them to produce more excitatory neurons, the primary signal-carrying cells of the cortex. The effect was strongest for upper-layer neurons, which are particularly expanded in the human brain.

"We already knew that these projections influence how the cortex develops," Bhaduri said. "What we specifically found is that this influence comes through an actual physical connection between the projections and the radial glia — a point of contact that just hasn't been identified before, and one that very likely does not exist in rodents."

Where does an autism-linked gene fit in?

The researchers tied this physical interaction to NRXN1, a gene already known for helping neurons form connections with one another. Mutations in NRXN1 have previously been linked to autism spectrum disorder.

To investigate its role, the team built assembloids from patient-derived cells carrying an NRXN1 mutation. In these models, the altered thalamic signals behaved differently from signals produced by unaffected cells. Those changes shifted the balance between the number of stem cells and the neurons they generated, offering a possible avenue for studying how early developmental disturbances shape the cortex.

What do the two studies have in common?

One study examined metabolism; the other explored neural connections. Yet both arrive at the same broader conclusion: radial glia do not make their developmental decisions in isolation. Their behavior is continuously shaped by signals from the environment around them.

The studies also highlight how far organoid technology has come. Roughly a decade ago, scientists had few practical ways to directly investigate how uniquely human neural stem cells behave. Today, brain organoids and assembloids allow researchers to recreate key features of human brain development in the laboratory and test questions that animal models alone cannot answer.

Bhaduri hopes the findings will push scientists to treat metabolism and physical cell-to-cell connections as active drivers of development rather than background processes.

"Ultimately, these studies give us a glimpse under the hood of how these cells make decisions," she said. "Understanding those decisions is a first step toward understanding normal brain development, disease vulnerability and, potentially, how similar stem-cell programs operate in brain cancer."

The research drew support from the National Institutes of Health, the National Science Foundation, the Simons Foundation, the Chan Zuckerberg Initiative, and more than a dozen other funders. As with any laboratory-based work, the findings come from organoid and donated-tissue models, and their implications for human health will require further study.