Most of what scientists know about how the brain makes decisions comes from studies built around a single, clean choice: option A or option B, press this button or that one. Real decision-making rarely works that way. A commute home, a stretch of homework interrupted by a text message, a day spent juggling competing responsibilities — these aren’t discrete choices so much as an ongoing balancing act. A new study from Baylor College of Medicine and the Duncan Neurological Research Institute at Texas Children’s Hospital set out to understand how the brain manages exactly that kind of continuous, shifting decision-making, and the answer reassigns a starring role to a brain region better known for an entirely different job.

The study, published in Nature, was led by co-corresponding authors Dr. Benjamin Hayden, professor of neurosurgery and McNair Scholar at Baylor, and Dr. Sameer Sheth, professor of neurosurgery and McNair Scholar at Baylor and director of the Gordon and Mary Cain Pediatric Neurology Research Foundation Laboratories at the Duncan NRI. Baylor graduate student Assia Chericoni was part of the multidisciplinary team behind the work.

Borrowing a framework from robotics

To study decision-making that unfolds continuously rather than in discrete steps, the researchers turned to a concept from control theory and robotics called compositional control. Under this framework, complicated behavior emerges not from switching between simple, mutually exclusive strategies, but from a dynamic blend of several goal-directed strategies running at once, with a higher-level “meta-controller” continuously adjusting how much weight each goal gets as conditions shift.

To test whether the human brain actually operates this way, the team designed a joystick-controlled prey-pursuit game in which 19 participants — all patients already undergoing intracranial monitoring for epilepsy, which allowed direct recording from implanted electrodes — chased virtual targets that varied in speed and reward value. While participants played, the researchers recorded the activity of individual neurons in three brain regions: the hippocampus (HPC), the anterior cingulate cortex (ACC), and the orbitofrontal cortex (OFC). Combining those direct recordings with computational modeling let the team trace how each region contributed to decisions as they unfolded in real time.

A GPS, not a light switch

The central finding was that the brain doesn’t toggle between competing goals the way a simple model of decision-making might predict. Instead, it continuously blends and re-weights multiple goal-directed strategies at once — a process the researchers describe as functioning “much like a GPS system that constantly recalculates a route in response to changing traffic conditions.”

The more unexpected result concerned the hippocampus. Long understood primarily as the brain’s memory center — the structure behind forming and storing new memories and supporting a kind of internal “cognitive map” — the hippocampus turned out to be doing something more actively engaged with the task at hand: tracking possible future strategies and helping to plan upcoming actions in real time.

The data pointed to a three-part division of labor across the regions studied. The hippocampus estimates the current and likely future state of the task and supports planning around it. The anterior cingulate cortex functions as a meta-controller, monitoring conditions and guiding shifts between competing goals or strategies. The orbitofrontal cortex represents the value structure of the environment — supplying a running estimate of how desirable each available option is relative to the others.

Retiring the passive-map view of memory

Taken together, the three-region framework challenges a long-standing view of the hippocampus as a largely passive structure — a kind of internal filing system or cognitive map that gets consulted rather than one that actively participates in shaping behavior. The new results instead position it as an active player in ongoing decision-making, continuously projecting forward to help the brain plan its next move rather than simply logging where it’s already been.

That reframing doesn’t diminish the hippocampus’s established role in memory; rather, it suggests memory and planning are more intertwined functions of the same structure than previously appreciated. The anterior cingulate cortex’s role as an arbiter of when priorities should shift, and the orbitofrontal cortex’s role in tracking the relative value of options, complete a coordinated network that the researchers argue underlies the constant, low-grade balancing act of everyday decision-making — the sort that never announces itself as a decision at all, because it’s simply how attention and effort get allocated moment to moment.

The practical implications reach beyond basic neuroscience. Because the recordings came from patients undergoing treatment for epilepsy, the findings sit at the intersection of fundamental brain science and clinical neurosurgery — the kind of direct human neural recording that’s rarely available outside a clinical context, and one that gives researchers a much finer-grained view of real-time decision-making than external brain imaging alone can provide.

The project was supported by National Institutes of Health grants R01 DA038615, R01 MH125377, U01 NS121472, and R01 MH129439, along with funding from the Cain Foundation and the McNair Foundation.

---

Endnotes

1. Baylor College of Medicine, “Beyond yes or no: How the brain navigates between competing goals,” news release, August 25, 2026, https://www.eurekalert.org/news-releases/1141464.
2. Assia Chericoni et al., “Neural basis of compositional control,” Nature, 2026, https://doi.org/10.1038/s41586-026-10896-8.