DATE: November 13, 2026 at 07:00AM
SOURCE: PSYPOST.ORG
** Research quality varies widely from fantastic to small exploratory studies. Please check research methods when conclusions are very important to you. **
-------------------------------------------------
TITLE: Neuroscientists just found a hidden โfingerprintโ in human memory
URL: https://www.psypost.org/neuroscientists-just-found-a-hidden-fingerprint-in-human-memory/
Electrical rhythms sweeping across the surface of the human brain do not just travel in straight lines when people form and recall memories, according to new research published in Nature Communications. Using direct brain recordings from neurosurgical patients, scientists found that these rhythms also organize into swirling spirals, expanding or contracting rings, and other intricate shapes that shift with both the stage of memory and the specific letter a person is viewing. Machine-learning models trained on these geometric patterns were able to tell apart different memory states and individual remembered items above chance levels.
The cerebral cortex, the folded outer layer of neural tissue covering the brain, plays a central role in memory, language, and perception. Within this tissue, groups of brain cells generate rhythmic electrical fluctuations called neural oscillations, which cycle several times per second. When the peak of one of these electrical cycles shifts progressively from one patch of cortex to neighboring patches over thousandths of a second, much like a ripple moving across water, neuroscientists call it a traveling wave.
Individual brain cells in the cortex are physically wired to many distant regions, and those connections cannot rewire themselves from second to second as a personโs task changes. The authors hypothesize that by sweeping across the brainโs surface in organized patterns, traveling waves may temporarily align the timing of specific neuron groups, allowing separated regions to communicate flexibly as mental demands shift.
โIn this study we took an unusual approach to studying large-scale brain activity and its ability to reorganize over time,โ senior author Joshua Jacobs, a professor of neurology at the University of Chicago, and lead author Anup Das, a postdoctoral researcher at the university, told PsyPost. โWe probed the spatial organization of signals across multiple brain regions together, rather than the more common approach of studying individual sites separately.โ
โTo explain our approach, an analogy we often find useful is trying to understand an ocean by measuring the water at individual locations without ever studying the waves traveling across its surface,โ Jacobs and Das explained. โYou might learn a great deal about what is happening at each location, but you would miss an important dimension of how the system organizes itself as a whole. Similarly, understanding the brain requires us to look beyond how much activity occurs at individual locations and investigate how traveling waves of neural activity are spatially organized and propagate across the cortex.โ
Studies of traveling waves during waking human thought previously focused on planar waves, which sweep in one straight direction across a broad stretch of cortex. In a 2024 direct brain-recording study by two of the same authors, straight-line waves changed their direction of travel between storing and recalling memories. Animal work hinted at richer shapes: a 2022 working-memory study of monkeys in PLOS Computational Biology found that frontal-lobe waves rotated in circles more often than they moved in straight planes.
Earlier human work also indicated that brain rhythms reflect individual items held in mind. A 2009 study of surgical patients co-authored by Jacobs found that localized fast electrical signals varied with the letter being viewed; that studyโs verbal-memory data were reanalyzed in the new research. Building on those findings, Das and Jacobs, alongside mathematics professor Bard Ermentrout of the University of Pittsburgh and biomedical engineering researcher Erfan Zabeh of Columbia University, tested whether traveling waves across the awake human cortex form spirals, rings, and other geometries that distinguish both broad memory stages and specific remembered items.
The researchers analyzed publicly available recordings from 24 patients with drug-resistant epilepsy hospitalized for seizure monitoring before surgery, and posted their custom analysis code online. Each patient had flat sensor grids placed directly on the cortical surface beneath the skull, a clinical procedure called electrocorticography, with electrodes spaced 10 millimeters apart. Seven additional patients from the original archives were excluded because fewer than two-thirds of their sensors displayed a shared rhythm.
Patients typically had 100 to 150 sensors implanted, each roughly one square centimeter in size and capable of picking up electrical signals from up to a million brain cells. The data came from what Jacobs and Das described as โthe gold standard of measuring electrical activity in the human brain: people preparing for surgery to manage severe, drug-resistant epilepsy.โ
Nine patients (four women and five men, ages 20 to 57) played a laptop spatial memory game, steering across a virtual beach to open treasure chests and memorize object locations. Each round progressed through navigation, encoding an uncovered object for 1.5 seconds, a shell-game distractor, confidence ratings, location recall (averaging 41% accuracy), and score feedback. The other 15 patients (seven women and eight men, ages 20 to 58) completed a verbal working-memory task, viewing lists of one to three English letters for one second each before indicating whether a test letter matched the list (averaging 92% accuracy).
The team first identified adjacent electrode clusters pulsing at the same rhythm between 3 and 40 cycles per second, or hertz, finding oscillations on about 86% of sensors across slower theta and alpha bands (4 to 12 hertz) and faster beta bands (12 to 30 hertz). This yielded 39 clusters across the 24 patients: 13 in the spatial task and 26 in the verbal task. Comparing each sensorโs cycle timing with neighbors within 25 millimeters mapped local wave flow across stable bursts lasting at least about 50 milliseconds (62.5 in the verbal task).
To track shifting geometries, the scientists used independent component analysis, a mathematical technique that separates overlapping signals into recurring spatial patterns and scores each patternโs strength per burst. Computer simulations and shuffled-data tests confirmed that this method extracted genuine wave shapes without inventing patterns from noise. Each pattern was classified as planar (moving in one straight direction), rotational (spiraling clockwise or counterclockwise), concentric (expanding outward as a source or contracting inward as a sink), or complex (patterns fitting none of the other categories, often mixing several local flows). Cross-validated neural networks were then trained on most of the data and tested on held-out portions to classify pairs of task stages or viewed letters from these scores.
Stable wave bursts lasted roughly 80 to 180 milliseconds on average before dissolving and reorganizing, with higher-frequency beta waves holding steady for slightly briefer windows than slower theta and alpha waves. These waves crossed the cortex at about 0.3 to 3.3 meters per second, traveling roughly twice as fast at higher frequencies than at lower ones.
Individual cortical regions did not stick to one wave geometry. Each electrode cluster cycled through an average of six recurring spatial patterns, which together accounted for more than 80% of the variation in wave movement.
Planar waves were the strongest category, with a median wave strength of 59% at lower frequencies and 43% at higher frequencies across tasks. (Wave strength reflects how closely the electrical timing across neighboring sensors fits an organized wave.) At higher beta frequencies, rotational spiral waves rose to a median strength of 37%, particularly during the spatial memory game.
Wave geometries also differed between hemispheres. Planar waves were stronger on the right side of the brain (about 64% strength) than on the left (about 42%), while left-hemisphere non-planar waves favored spirals during spatial memory and concentric rings during verbal memory.
โOne of the most exciting implications of this work is that the brain may communicate through much more elaborate spatiotemporal patterns than we traditionally appreciate,โ Jacobs and Das noted, referring to patterns that unfold across both space and time. โWe often think about brain activity in terms of which regions become active. But our findings suggest that we also need to pay attention to the geometry of neural activity itself: how it travels, rotates, expands, and contracts across the cortex.โ
โWhat surprised me most was the sheer diversity of traveling wave shapes that we found in the human cortex, and how meaningfully these different configurations related to behavior,โ Jacobs told PsyPost. โAnd these patterns were observed across widespread cortical areas, including frontal, temporal, and parietal regions.โ
When the researchers checked whether spiral waves spun preferentially in one direction across the temporal, parietal, and frontal lobes, any preference for one rotational direction over the opposite direction was not statistically significant in either task. Concentric ring waves, by contrast, reversed their predominant flow between tasks: 7 of 9 concentric patterns (78%) radiated outward as source waves during spatial memory, whereas 15 of 22 (68%) converged inward as sink waves during verbal memory.
All four wave types could change direction, strength, or both across behaviors, and every cluster recorded during the spatial game showed such shifts. In one patient, for instance, a planar wave nearly vanished during navigation and the distractor game, strengthened during encoding, and shifted direction during recall. Across the group, both rotational and complex waves grew stronger during the distractor phase.
Similar shifts appeared at the level of individual items in the verbal working-memory task. Half of the 26 clusters altered wave direction or strength depending on the letter viewed, such as flowing toward the upper front of a grid for โGโ and โQโ and reversing for โDโ and โJ.โ
โWe saw that there is a specific shape associated with a specific type of behavior,โ Das said in a University of Chicago news release. โThese shapes are not random. I like to think of it as a fingerprint for each person.โ Regarding the shifts between individual memorized letters, Das added, โThis suggests that brain waves can also encode item specific, fine-grained details in addition to broad events.โ
โAnother big surprise was that there was no single wave pattern that appeared in exactly the same way in every individual,โ the researchers told PsyPost. โThe repertoire and prevalence of these patterns varied considerably across patients. For example, the act of remembering a location in the treasure hunt task might generate a spiral wave for one person, while it produced a radiating source for another.โ
โWe find this heterogeneity particularly interesting,โ Jacobs and Das noted. โIt suggests that the human brain may not rely on one universal spatial pattern to support a particular cognitive operation. Different individuals may organize their neural activity through different dynamical configurations, even while performing similar memory tasks. Yet, within individuals, these patterns were sufficiently structured and reliable to distinguish behavioral states and, in some cases, specific memory representations.โ
Because wave patterns varied so much between people, the decoders were trained separately for each personโs electrode clusters rather than across patients. Neural networks trained on wave-pattern scores distinguished both spatial-task stages and remembered letters reliably above the 50% chance level. Decoding behavior from wave shape alone succeeded about 70% of the time, according to the universityโs news release.
Pairs with more distinct wave geometries yielded higher accuracy: navigation and distractor periods stood out from encoding and retrieval (but were hard to tell apart from each other), and โH,โ โJ,โ and โQโ were the most distinguishable letters. In control tests, peak frequency differences across stages or letters were not statistically significant, and decoders trained solely on frequency or rhythm strength relative to background noise failed to beat chance.
The researchers see possible applications for brain-computer interfaces, devices that translate brain signals into commands for computers or prosthetics. โBrainโcomputer interfaces, for example, depend on extracting meaningful information from neural signals to infer a personโs intentions or cognitive states,โ Jacobs and Das explained. โMost existing approaches focus on features such as neuronal firing rates or signal amplitudes. Our findings suggest that the spatial organization and propagation of neural activity could provide another dimension of information that could potentially complement these established approaches and could influence how we design future brainโcomputer interfaces.โ
These findings align with an earlier direct brain-recording study of sleeping clinical patients, which examined rotating cortical waves during sleep rather than active waking memory tasks and suggested that circular traveling waves may coordinate brain regions to support memory. In that sleep study, brief bursts of sleep-related rhythms called sleep spindles swept in circular loops across the cortex hundreds of times nightly rather than peaking simultaneously, though waking spirals in the new study spun in both directions without a single preferred loop.
To explain why outward source waves appeared more often during spatial memory while inward sink waves dominated verbal memory, the authors propose that spatial navigation requires local hubs to broadcast signals across widespread networks, whereas holding a few letters in mind relies on dampening competing activity and funneling signals inward. Based on mathematical models of connected rhythmic cells, the researchers also suggest that local differences in oscillation speed shape these geometries, as waves tend to travel from faster-pulsing cortical patches toward slower-pulsing ones. More broadly, the researchers see traveling waves as one way the brain might coordinate billions of neurons spread across distinct regions, and they propose that spirals may help strengthen connections between neurons, a process called synaptic plasticity.
โWhat makes traveling waves particularly interesting is that they provide a potential mechanism for this coordination,โ Jacobs and Das said. โFor example, rotational waves such as spirals revisit the same brain areas in multiple cycles, therefore sequentially scanning distributed brain areas to help enhance synaptic plasticity for memory consolidation. Similarly, source waves can take information from one brain area and broadcast that information outward across multiple brain areas. In this sense, traveling waves offer more than another way to visualize brain activity; they may help us understand the dynamical principles through which the brain coordinates information processing.โ
There are a few caveats to keep in mind regarding what these recordings can establish. Tracking and decoding wave shapes shows that cortical geometries shift alongside mental states, but this observational design cannot show that a spiral or concentric wave directly causes a memory to form or be retrieved, nor did the surface sensors record the firing of individual neurons beneath each grid. In addition, electrode grids covered only cortical regions selected for each patientโs clinical care rather than the entire brain.
Because several grids covered compact patches of the brainโs outer surface, segments of wide spirals may have resembled planar waves, potentially undercounting spirals, and surface sensors could not detect three-dimensional waves moving through deeper central relay structures such as the thalamus. Visual input, eye movements, and joystick control also accompanied memory processing. Future experiments combining single-neuron recordings, eye-tracking, and comparisons between memory and non-memory tasks with identical stimuli could help separate memory-specific wave geometries from sensory or motor activity.
Looking ahead, the team hopes to test whether targeted electrical or magnetic stimulation, such as transcranial magnetic stimulation applied from outside the skull, could nudge faltering rhythms to support memory rehabilitation in people experiencing cognitive decline. โIf a traveling wave is relevant for a certain kind of memory encoding, then if you want to help a person remember things better, maybe you can apply brain stimulation that strengthens that kind of traveling wave,โ Jacobs said in the news release. โHopefully, by characterizing patterns like this, then we can go to the next step and use mathematical models of each of these waves to understand where they are coming from. And then if we understand where theyโre coming from mechanistically, then we can potentially apply brain stimulation in a way to strengthen them.โ
โWe think we are only beginning to appreciate the computational possibilities offered by these different wave geometries,โ the researchers added. โThere is still a great deal to discover about how planar, spiral, and concentric waves interact, why the brain switches between these configurations, and whether these patterns provide distinct mechanisms for coordinating distributed neuronal populations. Ultimately, we hope this line of research will help establish a more comprehensive understanding of how the brain uses rich spatiotemporal dynamics to support cognition.โ
The study, โPlanar, spiral, and concentric traveling waves distinguish behavioral states in human memory,โ was authored by Anup Das, Erfan Zabeh, Bard Ermentrout, and Joshua Jacobs.
URL: https://www.psypost.org/neuroscientists-just-found-a-hidden-fingerprint-in-human-memory/
-------------------------------------------------
Private, vetted email list for mental health professionals: https://www.clinicians-exchange.org
Unofficial Psychology Today Xitter to toot feed at Psych Today Unofficial Bot @PTUnofficialBot
-------------------------------------------------
#psychology #counseling #socialwork #psychotherapy @psychotherapist @psychotherapists @psychology @socialpsych @socialwork @psychiatry #mentalhealth #psychiatry #healthcare #depression #psychotherapist #Neuroscience #Memory #BrainWaves #BrainResearch #CognitiveScience #HumanBrain #NeuralOscillations #Neurotechnology #BrainComputerInterface #ScienceNews
