For years, neuroscientists operated under a tidy assumption: high-gamma activity - that buzzy, high-frequency electrical chatter (70-300 Hz) picked up by electrodes in the brain - was basically the sound of nearby neurons firing. Local neurons spike, local field potential goes up, case closed. It was so intuitive, so elegant, that thousands of studies ran with it. Then a team at Northwestern University handed monkeys a neurological steering wheel and watched the whole theory quietly fall apart.

A Signal with an Identity Crisis

Here's the backstory. When researchers stick electrodes into brain tissue, they pick up two main things: spikes (individual neurons firing, like hearing one person clap) and local field potentials, or LFPs (the roar of the whole crowd). Within that crowd noise, there's a frequency band called high-gamma activity (HGA) that scientists have leaned on heavily - for brain-computer interfaces, for mapping brain function, for understanding how neural circuits process information.

The million-dollar question was: where does high-gamma actually come from? Two competing hypotheses have been duking it out. Team A said HGA is just the sum of all those local spikes - neurons near the electrode firing and their electrical blips adding up into that high-frequency hum. Team B argued it's actually summed postsynaptic potentials - incoming messages from other neurons landing on local cells, like a mailbox overflowing with letters from across the neighborhood (Buzsáki et al., 2012).

Most of the field had its money on Team A. A landmark 2011 study showed high-gamma tracked tightly with spiking in visual cortex, and that became the default assumption (Ray & Maunsell, 2011).

Teaching Monkeys to Split the Unsplittable

Enter Tianhao Lei, Michael Scheid, and colleagues from Marc Slutzky's lab at Northwestern. They devised a brilliantly sneaky experiment, published in Nature (Lei et al., 2026). They built a brain-machine interface where monkeys controlled a cursor on a screen using signals from a single electrode in their motor cortex. But here's the twist: the cursor's horizontal movement was driven by spike rates, while vertical movement was driven by high-gamma activity.

If HGA really is just summed spikes from the same neurons, this should be impossible. You can't separately control two knobs that are wired to the same dial. It would be like trying to turn up the volume on your TV without changing the channel when they're controlled by the same button.

The monkeys nailed it within a few sessions. They learned to move the cursor in any direction, independently controlling their spike rates and high-gamma levels on the same electrode. That's the neuroscience equivalent of watching someone pat their head and rub their stomach, except the head and stomach were supposed to be the same body part.

So Where Does the Signal Actually Come From?

With the "summed local spikes" hypothesis looking shaky, the team dug deeper. When monkeys increased their high-gamma activity without increasing local spiking, something interesting happened across the electrode array: neurons distributed across millimeters of cortex started firing in synchrony. The high-gamma signal wasn't reflecting what the nearest neurons were doing - it was picking up the incoming mail from a much wider postal district.

The timing clinched it. The high-gamma bumps on one electrode lined up with spikes arriving from other electrodes at just the right delay, consistent with signals traveling along axons and generating postsynaptic potentials at their destination. High-gamma activity, it turns out, is more like an inbox than a microphone - it captures what's being received, not what's being sent.

Why Your Brain-Computer Interface Should Care

This isn't just academic navel-gazing. High-gamma signals have been a workhorse in brain-computer interfaces partly because they're remarkably stable over time - way more reliable than tracking individual neuron spikes, which tend to drift and vanish as electrodes shift (Flint et al., 2013). Slutzky's finding finally explains why: if high-gamma reflects the pooled input from many distant neurons rather than a handful of local ones, losing a few nearby cells to electrode migration barely matters. The mail still gets delivered even if a few houses on the street move away.

But there's a flip side. Every study that interpreted high-gamma as a readout of local neural processing might need a second look. That signal you thought was telling you what neurons right here were doing? It was actually a bulletin from the whole cortical neighborhood (Pesaran et al., 2018). For neuroscientists mapping brain function with high-gamma, this is like discovering your "local news" channel was actually broadcasting from three counties over.

The Bottom Line

Sometimes the most useful signals are the ones we understood the least. High-gamma activity spent decades being treated as a proxy for nearby neuron firing, and it worked well enough that nobody questioned the fine print. It took a clever BMI experiment - and some very cooperative monkeys - to reveal that this workhorse signal has been reporting on a much larger story all along. The brain, as usual, is weirder and more interconnected than we gave it credit for.

References

1. Lei, T., Scheid, M.R., Flint, R.D., Glaser, J.I., & Slutzky, M.W. (2026). Active dissociation of intracortical spiking and high gamma activity. Nature. DOI: 10.1038/s41586-026-10331-y

2. Ray, S. & Maunsell, J.H.R. (2011). Different origins of gamma rhythm and high-gamma activity in macaque visual cortex. PLOS Biology, 9(4), e1000610. DOI: 10.1371/journal.pbio.1000610. PMID: 21532743

3. Buzsáki, G., Anastassiou, C.A., & Koch, C. (2012). The origin of extracellular fields and currents - EEG, ECoG, LFP and spikes. Nature Reviews Neuroscience, 13, 407-420. DOI: 10.1038/nrn3241. PMID: 22595786

4. Flint, R.D., Wright, Z.A., Scheid, M.R., & Slutzky, M.W. (2013). Long term, stable brain machine interface performance using local field potentials and multiunit spikes. Journal of Neural Engineering, 10(5), 056005. DOI: 10.1088/1741-2560/10/5/056005. PMID: 23918061

5. Pesaran, B., Vinck, M., Einevoll, G.T., et al. (2018). Investigating large-scale brain dynamics using field potential recordings: analysis and interpretation. Nature Neuroscience, 21(7), 903-919. DOI: 10.1038/s41593-018-0171-8. PMID: 29942039

Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.