Northwestern Medicine scientists have discovered new insights into how high gamma activity — an informative, widely studied brain signal — is generated, findings that can impact how past and future neurological studies using this signal are interpreted, according to a recent study published in Nature.
When neuroscientists record electrical activity in the brain, they frequently use electrodes placed on either the surface of the brain or inside the cerebral cortex. These electrodes are often used to record local field potentials, which are combined signals from multiple neurons.
One component of the local field potential, known as high gamma band activity with a frequency range of 70 to 300 hertz, is used to investigate many brain functions, including sensory processing, motor control and brain-machine interfaces — direct communication links between the brain’s electrical activity and an external device, or brain computer interfaces — as well as cognitive processing like attention and memory.
How exactly high gamma activity is generated in the brain, however, has remained elusive, said Marc Slutzky, ‘02 MD, ‘00 PhD, ‘06 GME, professor in the Ken and Ruth Davee Department of Neurology’s Division of Comprehensive Neurology and senior author of the study.
“Many studies have assumed that it is just summed action potentials — a.k.a. spikes, the ‘all-or-nothing’ signal that individual neurons produce — that are produced near the electrode. Some others suggest that high gamma activity is actually from the post-synaptic potentials, which are smaller potentials produced in the postsynaptic neuron by presynaptic spikes,” said Slutzky, who is also a professor of Neuroscience and of Biomedical Engineering in the McCormick School of Engineering.
To test these hypotheses more rigorously, Slutzky’s team developed a brain-machine interface (BMI) to test whether animal subjects could dissociate spiking rates from high gamma activity recorded on the same electrode. If high gamma activity were due to summed spikes, the nearest spike would contribute the most to the high gamma, and therefore they would not be able to be dissociated, according to Slutzky.
The BMI recorded both spikes and high gamma activity on single electrodes and mapped them to perpendicular components of cursor movement in the BMI, allowing subjects to move the cursor to different targets on a monitor.
After just minutes of training, the subjects could differentiate local spiking from high gamma activity on a single electrode, suggesting that high gamma activity is generated by summed postsynaptic potentials near the electrode.
Further analyses showed that the high gamma activity was correlated with the activity of spiking that was distributed across the entire recorded area of the cortex. This, plus the timing of the high gamma activity relative to the spikes on other electrodes, indicated that these postsynaptic potentials are generated by spiking in neurons that are widespread across the cortex.
Slutzky said these findings have implications for future brain-machine interfaces and how many studies using high gamma activity are interpreted.
“This might explain why high gamma signals are more stable than spikes: because high gamma represents the summed postsynaptic potentials of many neurons. It also has implications in how we interpret the findings of many fundamental neuroscience studies that have used high gamma, because it may not represent activity that is as localized as we previously thought,” Slutzky said.
Tianhao Le, a student in the Northwestern University Interdepartmental Neuroscience (NUIN) program, and Michael Scheid, PhD, a former graduate student in the Slutzky laboratory, were co-lead authors of the study.
Co-authors include Robert Flint, PhD, research assistant professor of Neurology in the Division of Comprehensive Neurology, and Joshua Glaser, ‘18 PhD, assistant professor of Neurology in the Division of Comprehensive Neurology.
This study was supported in part by National Institute of Health grants K08NS060223, R01NS094748, R01NS112942, R01NS099210, RF1NS125026, T32EB009406, R00NS119787 and T32NS047987; the National Institute for Theory and Mathematics in Biology through the National Science Foundation (DMS-2235451); and the Simons Foundation (MPTMPS-545 00005320).
