The dominance of large-scale phase dynamics in human cortex, from delta to gamma
David M Alexander and Laura Dugué. eLife 2026.

The brain is still largely a mystery. We know a lot of facts about the details, such as kinds of cells, how they communicate, and what they respond to out in the world. But exactly which aspects are important is still an open question.  Neuroscience often paints a picture of parts and regions, for example, with vision being processed at the back of the brain, decision-making in the front, and cells firing to preferred stimulus orientations or movements of the body. This is at odds with how we experience our own cognition, which seems a more unified phenomenon.

A long-held view is that the cortex (the big folded structure that takes most of the space in our skulls) is best understood as cells sending impulses to each other, and the rest of the measurable activity is largely just a side-effect of adding these inputs together. This is much the same idea that transistors in a computer are important, and the heat and noise that a computer gives out is not. Another view is that the patterns of brain activity over space and time, and including whole cortex patterns, are key features to understanding. These patterns arise from the biochemistry of the cell walls and are detectable as electrical activity.

We wanted to know which features of the cortex’s activity are important, by considering the different sizes of patterns of activity. We assessed the strength of activity over different sized regions (like finding the dominant note in a chord, or the brightest colour in the rainbow). Knowing the strongest patterns, and how big they are, has not been assessed before for direct measurements of the cortex’s electrical activity that include widespread regions.

We found that the strongest patterns occurred over the largest regions of cortex that we could measure. This was found by looking at brain recordings from epilepsy patients. Some patients have many recording sites placed in their brain to find out where the problem is. A key part of our research was to use specialized mathematics to take into account the irregular placement of the recording sites, which are positioned for medical reasons. We found that the large, strongest patterns occurred at many rhythms: from as slow as a clock tick to faster than a humming bird’s wings.

With this study we wanted to improve our basic understanding of how the brain works. The findings add evidence to the view that the cortex works as a coordinated whole, consistent with recent results using fMRI that show the global geometry of the cortex is critical to its function. In particular, the present results show that if you want to understand activity at a local site in the cortex, you also need to measure whole cortex activity. This result adds to our previous findings that large-scale traveling waves in the brain can be used to predict local activity in the near future.