Information from Lay-Language Summaries is Embargoed Until the Conclusion of the Scientific Presentation
555—Extrastriate Cortex: Neural Coding
Tuesday, November 12, 2013, 8:00 am - 12:00 noon
555.08: The amplitude and phase of EEG oscillations index the spiking correlation of underlying brain areas
Location: Halls B-H
*A. C. SNYDER1,2, C. M. WILLIS1, M. A. SMITH1,2,3; 1Dept. of Ophthalmology, 2Ctr. for the Neural Basis of Cognition, 3Dept. of Bioengineering, Univ. of Pittsburgh, Pittsburgh, PA
Abstract Body: An understanding of neuronal communication, both within and between regions of the brain, is fundamental to explaining perception and behavior. Communication among neurons has been studied at a variety of scales, from the relatively coarse resolution afforded by scalp electroencephalography (EEG) to the fine resolution achieved in extracellular microelectrode recordings of pairs of neurons. Measures of functional connectivity such as synchronization of EEG oscillations on a large scale, and correlation of spiking activity among small groups of neurons on a small scale, have each been separately linked to a variety of cognitive and perceptual processes. Although the knowledge gained to date by these parallel efforts has been invaluable, a detailed understanding of neuronal communication will ultimately need to bridge these scales so that the relationships between individual neurons can be understood in the context of the interactions of large-scale brain regions, and vice versa. Here we take the first steps toward this goal with the use of simultaneous EEG and single unit recordings in alert macaque monkeys. Since EEG signals are generally considered to be manifestations of coherent local field potentials (LFPs) generated in cortex, and coherent LFPs would reflect common inputs to that region of cortex, we reasoned that EEG oscillations would be related to the correlation in spiking activity between pairs of neurons in the underlying cortical area. To test this prediction, we recorded population spiking activity from microelectrode arrays chronically implanted in visual area V4 of awake, behaving macaques simultaneously with signals from EEG electrodes on the scalps of the animals. Consistent with our predictions, we found that the amplitude and phase of EEG oscillations indexed the correlation of the underlying spiking activity. Interestingly, we found this relationship between EEG oscillations and spiking correlation to be frequency-specific, which may be related to important functional distinctions between different EEG frequency bands. This is a crucial development that directly links EEG signals to physiological processes with clear computational consequences, and opens an avenue for understanding spiking correlations in terms of the interaction among large-scale brain networks.
Lay Language Summary: Our research links brain activity recorded non-invasively at the scalp to the electrical signals generated within the brain by individual neurons. In humans, the most common and most accessible method for measuring brain activity involves recording electrical potentials from the scalp, also known as electroencephalography or EEG. It is known from research with animals that brief electrical impulses, or “spikes”, of individual neurons form the basic currency of the nervous system. Although EEG has long been known to reflect the activity of underlying neurons, an explanatory gap has persisted. Changes in EEG activity have been linked to perception and cognition in healthy individuals, and EEG is also used to study and diagnose diseases and disorders. However, the nature of the distinct neural events, in terms of the activity of groups of individual neurons, that give rise to both the EEG signals and the behaviors themselves have remained elusive. Our research is designed to translate between these two important measures of brain activity. Our results show that the details of the scalp signals can predict the way in which underlying groups of neurons are communicating, and the extent to which they can process information. This is a key step towards closing the explanatory gap between EEG and the neural circuits that give rise to mental behaviors. To make this development, we performed the first simultaneous recordings of EEG signals at the scalp along with the activity of dozens of individual neurons in the brains of monkeys. We estimated the extent to which the neurons “talked” with each other using a statistical measure of their interactions, the extent to which one neuron’s activity influences another’s. We tested whether this communication among neurons was related to distinctive features of the EEG. We found that the magnitude and progression of waves in the EEG on the scalp reflect the amount of communication among neurons in the underlying brain regions, and in turn their capacity for processing sensory information from the outside world. Uncovering this relationship between features of brain waves that are recorded non-invasively and the activity of underlying neurons in unison is extremely valuable. Existing findings based on EEG research can be reinterpreted in the context of neuronal processing. Future research with humans can be more focused because of this emphasis on neuronal processes, and will be easier to relate to the important research performed at the cellular and molecular levels, which can only be performed with animals. Our current findings were derived from animals performing very simple tasks, an important first step to establish baseline metrics. Our future research is aimed at conducting our simultaneous recordings while animals perform more complex tasks, designed to parallel experiments performed with humans.
Neuroscience 2013 (43rd annual meeting of the Society for Neuroscience)Exit