Traveling waves in the cortex form spirals and sinks that vary with the task

A spiral is harder to write off as engine noise than a plane is. A wave that converges on one spot, or rotates around it, has a geometry, and in the April 2026 study the geometry differed from task to task ^[1][3]. Joshua Jacobs, one of the University of Chicago researchers on the team, makes the case without needing the stronger claim ^[2]. "Even if the traveling wave was just the result of neurons firing, like the sound of the engine, it still tells us something very interesting about the brain based on what the person is doing," he said ^[8]. "If you see a sound moving in a direction across your engine, it means something in your engine is directionally organized." ^[9]

Seeing that geometry at all depends on where the electrodes sit. EEG has measured the brain's oscillations from the scalp since the 1920s, and recording from outside the skull limits how much detail comes through ^[10]. Jacobs' lab records from inside the brain instead. People with severe epilepsy sometimes have electrodes implanted to find where their seizures start. With roughly 100 electrodes in one region, and the patients' permission, the team captures activity at high resolution in space and time while participants do thinking tasks ^[11].

Direction came first. In a 2024 study in Nature Human Behavior, Jacobs, Uma Mohan (now a neuroengineer at the National Institutes of Health) and colleagues reported waves crossing the cortex in opposite directions during memory tasks ^[12]. "They either go back to front or front to back across the brain," Jacobs said ^[13]. He takes the direction to be the direction information travels. Visual regions sit at the back and the prefrontal cortex at the front, so when you attend to what you see, signals run from the eyes back to visual cortex and then forward into the rest of the brain ^[14].

The study does not show whether the waves do anything. As Quanta describes the Nature Communications result, different tasks were associated with distinct wave patterns ^[3]. The association fits waves that coordinate processing. It also fits waves that are a byproduct of whatever does the coordinating, and Jacobs' engine analogy leaves that possibility open ^[8]. Quanta's account does not give participant numbers or effect sizes. The population is narrow as well: people with severe epilepsy, with electrodes placed where clinicians need to look for seizures ^[11].

The case for the stronger reading rests on timing. Neurons change their connections with their neighbours over days or months, while behavior has to adapt within seconds ^[6]. A single day is 86,400 seconds, so rewiring is tens of thousands of times too slow for a second-by-second adjustment ^[1]. According to Quanta, a significant number of neuroscientists now believe large-scale wave patterns may reorganize the brain in real time and fill that gap ^[7]. Earl K. Miller, a cognitive neuroscientist at MIT, describes how the question has moved. "The work coming out is moving this from 'Are they relevant?' to 'This is a major motif of how the cortex processes information,'" he said ^[5].

In my view the structured-signal claim holds as a description. Source, sink and spiral patterns that change with the task are more organized than a featureless background hum would be ^[1][3]. That the waves are how the cortex routes information is still a hypothesis. Its support so far is association, in humans and animals, plus the timing argument ^[15][6].