High-frequency (600 Hz) population spikes in human EEG delineate thalamic and cortical fMRI activation sites
Introduction
fMRI delineates human brain activity noninvasively at high spatial resolution and has become a cornerstone of cognitive neuroscience (Belliveau et al., 1991, Bandettini et al., 1992, Frahm et al., 1992, Kwong et al., 1992, Ogawa et al., 1992). A limitation of fMRI, however, arises from its dependence on activity-coupled changes in blood flow or hemoglobin oxygenation (blood oxygenation level dependent–BOLD–contrast). All underlying neurophysiological events, e.g., excitatory as well as inhibitory synaptic activity and action potentials (APs) which occur on a time scale of milliseconds, are represented by a ‘one-dimensional’ vascular response (i.e., the BOLD signal) which develops on a time scale of seconds. Despite considerable recent progress in understanding the relationship between neuronal activity and associated changes in metabolism and blood flow (Logothetis et al., 2001, Attwell and Iadecola, 2002, Logothetis, 2002, Caesar et al., 2003, Logothetis, 2003, Raichle, 2003, Iadecola, 2004, Logothetis and Wandell, 2004, Lauritzen, 2005, Mukamel et al., 2005, Raichle and Mintun, 2006, Viswanathan and Freeman, 2007), inferences from the BOLD signal on the underlying neurophysiological events have remained largely elusive.
One way to circumvent this ‘inverse problem of fMRI’ (Buckner, 2003) is to combine fMRI with more direct measurements of neuronal activity. This approach has been employed in animal studies by incorporating simultaneous invasive electrode-based recordings of neuronal low-frequency local field potentials and high-frequency APs into the fMRI study (Logothetis et al., 2001, Hyder et al., 2002, Smith et al., 2002, Schwindt et al., 2004, Van der Linden et al., 2007). By analogy, we are pursuing a novel approach to combine fMRI with a noninvasive recording of APs so that it can be employed in human subjects.
Assessing APs noninvasively in humans, however, is challenging as the conventional scalp EEG is dominated by low-frequency signals that mainly reflect synchronized excitatory and inhibitory postsynaptic potentials. Therefore, the contribution of APs to the EEG is minor owing to their temporal characteristics of short duration and high-frequency content, small generator volumes and tissue low-pass filtering (Nunez and Silberstein, 2000). As a unique proxy option for obtaining a noninvasive AP measure during fMRI, we capitalize here on HFBs acquired by surface EEG: somatosensory evoked potentials (SEPs) contain at least two-minute HFB components that represent scalp projections of population spikes of subcortical as well as cortical origin (Curio 2000). By combining single unit recordings and epidural EEG in behaving monkeys, it has recently been demonstrated that the late HFB component reflects spiking activity in primary somatosensory cortex (S1) (Baker et al., 2003). For the early HFB component, EEG source localization studies (Gobbele et al., 1998, Restuccia et al., 2002) and invasive depth recordings (Klostermann et al., 2003, Curio, 2005) point to thalamic sources.
Here, we evaluate the feasibility of incorporating recordings of these HFBs into fMRI in three steps: (i) we first investigate whether HFBs with amplitudes in the nanovolt range can be recorded with conventional surface EEG during fMRI data acquisition, (ii) we then test the sensitivity of this approach to detect stimulus-rate induced HFB amplitude modifications, and (iii) we finally utilize spontaneous modulation of rapidly succeeding population spikes to investigate corresponding fMRI activation sites along the somatosensory thalamocortical pathway.
Section snippets
Subjects
49 subjects participated in the study, (14 male, mean age 26, range 20–31 years). All subjects were healthy, right-handed volunteers and gave written informed consent prior to the investigation. The study was approved by the Ethics Committee of the Charité, University Medicine Berlin.
Experimental design
During simultaneous EEG-fMRI acquisition, subjects were lying relaxed in a supine position in the dimly lit bore of the MR-scanner. They were instructed to stay awake with open eyes and to pay attention to the
Step 1: feasibility
In order to explore whether ultrafast EEG signals such as HFBs can be recovered from simultaneous EEG/fMRI recordings, 49 healthy subjects were electrically stimulated at the median nerve during simultaneous acquisition of EEG and fMRI under three different stimulus rates. In all subjects stimulus-locked EEG averaging resulted in a typical SEP over the contralateral somatosensory cortex. Individual unfiltered and bandpass-filtered SEPs are shown in Supplementary Fig. 1, respectively.
Discussion
Although fMRI has developed into a cornerstone of modern cognitive neuroscience, a huge gap still exists to the immense knowledge on brain function which is based on invasive electrophysiological recordings. In order to bridge this gap it will be crucial to relate the fMRI signal to different aspects of the underlying neural events, particularly spiking activity. Here, we introduce a new strategy for identifying measures of neuronal spiking during fMRI and relating them to the fMRI signal
Acknowledgments
This work was supported by the German Federal Ministry of Education and Research BMBF (Berlin Neuroimaging Center; Bernstein Center for Computational Neuroscience) and the German Research Foundation DFG (Berlin School of Mind and Brain; SFB 618-B4).
The authors declare that they have no competing financial interests.
References (78)
- et al.
Identification of EEG events in the MR scanner: the problem of pulse artifact and a method for its subtraction
NeuroImage
(1998) - et al.
A method for removing imaging artifact from continuous EEG recorded during functional MRI
NeuroImage
(2000) - et al.
Processing of location and pattern changes of natural sounds in the human auditory cortex
NeuroImage
(2007) - et al.
Stepping stone sampling for retrieving artifact-free electroencephalogram during functional magnetic resonance imaging
NeuroImage
(2003) - et al.
The neural basis of functional brain imaging signals
Trends Neurosci
(2002) - et al.
Combining electrophysiology and functional imaging — different methods for different questions
Trends Cogn. Sci.
(2007) - et al.
Single-trial EEG-fMRI reveals the dynamics of cognitive function
Trends Cogn. Sci.
(2006) - et al.
Endogenous cortical rhythms determine cerebral specialization for speech perception and production
Neuron
(2007) - et al.
High-frequency, (600 Hz) SEP activities originating in the subcortical and cortical human somatosensory system
Electroencephalogr. Clin. Neurophysiol.
(1998) - et al.
Functional dissociation of a subcortical and cortical component of high-frequency oscillations in human somatosensory evoked potentials by motor interference
Neurosci. Lett.
(2003)
Different origins of low- and high-frequency components, (600 Hz) of human somatosensory evoked potentials
Clin. Neurophysiol.
Somatic evoked high-frequency magnetic oscillations reflect activity of inhibitory interneurons in the human somatosensory cortex
Electroencephalogr. Clin. Neurophysiol.
High-frequency oscillations in human visual cortex do not mirror retinal frequencies
Neurosci. Lett.
EEG-correlated fMRI of human alpha activity
NeuroImage.
Recent advances in recording electrophysiological data simultaneously with magnetic resonance imaging
NeuroImage
Event-related fMRI with simultaneous and continuous EEG: description of the method and initial case report
NeuroImage
Correlates of alpha rhythm in functional magnetic resonance imaging and near infrared spectroscopy
NeuroImage
Integration of fMRI and simultaneous EEG: towards a comprehensive understanding of localization and time-course of brain activity in target detection
NeuroImage.
Dissociated changes of somatosensory evoked low-frequency scalp responses and 600 Hz bursts after single-dose administration of lorazepam
Brain Res.
Influence of cholinergic circuitries in generation of high-frequency somatosensory evoked potentials
Clin. Neurophysiol.
Increase of brain-stem high-frequency SEP subcomponents during light sleep in seizure-free epileptic patients
Clin. Neurophysiol.
Simultaneous EEG-fMRI
Neurosci. Biobehav. Rev.
Evaluating gradient artifact correction of EEG data acquired simultaneously with fMRI
Magn. Reson. Imaging
Studying spontaneous EEG activity with fMRI
Brain Res. Brain Res. Rev.
Cortical responses to pain in healthy individuals depends on pain catastrophizing
Pain
Dipole source localization and fMRI of simultaneously recorded data applied to somatosensory categorization
NeuroImage.
The effects of stimulus rates on high frequency oscillations of median nerve somatosensory-evoked potentials—direct recording study from the human cerebral cortex
Clin. Neurophysiol.
Decomposition of working memory-related scalp ERPs: crossvalidation of fMRI-constrained source analysis and ICA
Int. J. Psychophysiol.
Changes of short latency somatosensory evoked potential in sleep
Electroencephalogr. Clin. Neurophysiol.
EEG oscillations at 600 Hz are macroscopic markers for cortical spike bursts
J. Physiol.
Time course EPI of human brain function during task activation
Magn. Reson. Med.
Visual evoked potentials recovered from fMRI scan periods
Hum. Brain Mapp.
Functional mapping of the human visual cortex by magnetic resonance imaging
Science
Localizing P300 generators in visual target and distractor processing: a combined event-related potential and functional magnetic resonance imaging study
J. Neurosci.
Mental chronometry of working memory retrieval: a combined functional magnetic resonance imaging and event-related potentials approach
J. Neurosci.
Visual evoked potential, (VEP) measured by simultaneous 64-channel EEG and 3 T fMRI
NeuroReport
The hemodynamic inverse problem: making inferences about neural activity from measured MRI signals
Proc. Natl. Acad. Sci. U. S. A.
Dissociation of spikes, synaptic activity, and activity-dependent increments in rat cerebellar blood flow by tonic synaptic inhibition
Proc. Natl. Acad. Sci. U. S. A.
Refractory periods observed by intrinsic signal and fluorescent dye imaging
J. Neurophysiol.
Cited by (35)
Schrödinger filtering: a precise EEG despiking technique for EEG-fMRI gradient artifact
2021, NeuroImageCitation Excerpt :Removal of the gradient artifact is challenging because it overlaps in time and frequency with the EEG bands, limiting the effectiveness of basic time- or frequency-based processing methods like bandpass filtering. Several techniques to reduce the gradient artifact have been developed, including (1) gradient artifact censoring (Goldman et al., 2002; Kruggel et al., 2000; Pan et al., 2010; Ritter et al., 2008); (2) low-pass filtering (Allen et al., 2000; Anami et al., 2003; Mantini et al., 2007; Niazy et al., 2005); (3) notch filtering (Grouiller et al., 2007; Hoffmann et al., 2000); (4) reference layer artifact subtraction (Chowdhury et al., 2014; Steyrl et al., 2017); (5) adaptive noise cancelation (Allen et al., 2000; Niazy et al., 2005); (6) principal component analysis and singular value decomposition (Liu et al., 2012; Negishi et al., 2004; Niazy et al., 2005); (7) independent component analysis (Acharjee et al., 2015; Grouiller et al., 2007; Mantini et al., 2007; Mayeli et al., 2019 ; Piorecky et al., 2019; Ryali et al., 2009); (8) motion parameter-driven correction (LeVan et al., 2016; Maziero et al., 2016; Sun and Hinrichs, 2009); and (9) average artifact subtraction (AAS) (Allen et al., 2000; Anami et al., 2003; Ferreira et al., 2016; Mandelkow et al., 2006; Moosmann et al., 2009; Niazy et al., 2005). AAS, which forms a rough template of the periodic gradient artifact via moving average, has proven to be the simplest and most effective for simultaneous signal preservation and artifact reduction.
The Cerebellar Mossy Fiber Synapse as a Model for High-Frequency Transmission in the Mammalian CNS
2016, Trends in Neurosciences