In The Brain, A Fraction Of A Second Separates Learning From Movement

One hundred milliseconds. That is roughly the time it takes to blink. It is also, according to a new study, the difference between a dopamine signal linked to learning-related changes in the brain and one that drives a faster movement instead. The margin barely registers on a clock, yet the brain appears to track it precisely, and the consequences reach into some of the most debilitating neurological conditions known to medicine.

Researchers at New York University found that a chemical called acetylcholine acts as a timing gatekeeper inside the striatum, a brain region central to learning, decision-making, and movement control. Published in Nature Neuroscience, the study reveals that the sequence in which dopamine and acetylcholine arrive at brain cells, not just the size of either signal, determines what those cells actually do with the information.

“Our findings suggest that cholinergic dynamics determine whether dopamine promotes vigor or learning, depending on the instantaneous behavioral context,” the authors wrote. Cholinergic is simply the scientific term for anything involving acetylcholine.

How Does the Brain Sort Dopamine Signals for Learning and Movement?

Dopamine does double duty in the brain. It signals whether a reward was better or worse than expected, a process central to how people and animals learn, and it also helps control the speed and vigor of physical movement. Parkinson’s disease, caused by the loss of dopamine-producing neurons, illustrates how essential that second function is. Without sufficient dopamine, movement becomes slow, rigid, and difficult to initiate.

Sorting those two types of signals has always been the harder problem. Brain cells in the striatum receive dopamine continuously, yet somehow distinguish between “update what you expect from this situation” and “move now, move fast.” How that sorting happens had remained one of the more stubborn unanswered questions in neuroscience.

Acetylcholine, produced by a small population of neurons embedded in the striatum, had long been suspected as the sorting mechanism. Earlier lab work suggested it could gate whether dopamine produced lasting changes in neural connections, but the idea had never been tested in animals actively making decisions. This study was built to do exactly that.

Rats, Rewards, and the Milliseconds That Matter for Dopamine Learning

To test how the two chemicals interact during real behavior, the NYU team trained 79 rats on a decision-making task designed to separate reward-related brain events from movement-related ones. Each trial began with the rat poking its nose into a central port, triggering a tone that signaled how large a water reward was on offer, anywhere from 5 to 80 microliters. A light then showed which side port held the reward. Rats could wait for it or opt out and start a new trial.

Using fiber optic probes implanted in the brain, the team measured dopamine and acetylcholine release in real time inside the dorsomedial striatum, a subregion known to be critical for linking actions to their outcomes.

At the reward-offer tone, acetylcholine dipped first, and dopamine followed roughly 100 milliseconds later. At that moment, the size of the dopamine signal predicted how quickly rats initiated their next trial, a reliable behavioral sign that learning had occurred. Dopamine was functioning as a reward prediction error, updating the brain’s running estimate of how good the current environment was.

Later in the same trial, when the reward became available after an unpredictable delay, the sequence flipped. Dopamine peaked about 50 milliseconds before acetylcholine dipped. That signal looked statistically similar to the first and scaled with how long the rat had waited, which is exactly what a reward prediction error should do. Yet searches for any effect on subsequent behavior, across multiple measures, turned up nothing. Out-of-sequence dopamine left no detectable trace on learning in this task.

During movement-associated moments, a third pattern emerged. When rats physically turned toward the reward port, acetylcholine surged at the same time dopamine spiked, and a larger dopamine signal predicted a faster, more vigorous turn rather than any change in future expectations.

How Dopamine Rewires Brain Cells in Real Time

To see whether the learning-linked dopamine signal was changing the brain at the cellular level, the team used ultra-thin electrode arrays called Neuropixels probes, implanted in the striatum of eight rats, to record the firing activity of individual neurons across hundreds of consecutive trials.

Roughly half of the recorded neurons, 51.6 percent, shifted their activity levels following a large dopamine signal at the reward-offer moment, and those shifts persisted into subsequent trials even when nothing notable happened in between. That persistence points to genuine plasticity, a lasting update to how those cells respond. The effect was specific to the moment when dopamine followed acetylcholine. Where dopamine led, no comparable lasting change appeared.

The authors propose a receptor-level explanation. When acetylcholine drops, certain receptors that normally suppress dopamine’s ability to strengthen neural connections may briefly go offline, opening a window in which dopamine can drive lasting changes in how strongly neurons communicate. When dopamine arrives first, that suppression may still be active, and the signal passes through without leaving a mark. The authors are careful to frame this as a hypothesis, noting that the precise molecular mechanics remain to be confirmed.

What These Findings Mean for Parkinson’s Disease and Addiction

Dopamine has been central to research on Parkinson’s disease, addiction, depression, and schizophrenia for generations, yet treatments targeting it often produce incomplete or unpredictable results. One reason may be that dopamine’s effects are not determined solely by how much is released, but by its timing relative to acetylcholine in the moments surrounding a meaningful event.

Parkinson’s disease involves the progressive loss of dopamine-producing neurons in circuits that depend heavily on the dopamine-acetylcholine relationship this study examined. In addiction, reward prediction errors are thought to drive the learning that locks in drug-seeking behavior. A mechanism in which acetylcholine timing controls whether those prediction errors actually stick opens new ground for understanding why habits form and why they are so hard to break.

A 100-millisecond window is easy to overlook. In the striatum, it may be everything.

---