Simultaneous MEG and intracranial EEG recordings during attentive reading
Introduction
An increasing body of evidence gathered over the last fifteen years suggests that most cognitive functions, from sensory and motor processing to language or memory, involve the formation of local neural assemblies characterized by synchronous oscillations in the gamma band, i.e., at frequencies starting at 40 Hz (for reviews, see Singer, 1999, Tallon-Baudry and Bertrand, 1999). Interestingly, the formation of such assemblies seems to depend upon the degree of attention allocated by subjects or animals to the task at hand (Bauer et al., 2006, Bichot et al., 2005, Fries et al., 2001, Gruber et al., 1999, Steinmetz et al., 2000, Tallon-Baudry et al., 2005), which suggests that monitoring gamma band activity could provide a quantitative measure reflecting attention. Indeed, in a recent intracerebral EEG study, which manipulated explicitly the attention of subjects in a verbal working memory task, we found that gamma band responses were stronger for attended stimuli, later recalled, than for unattended ones (Mainy et al., 2007).
We have all experienced the effect of attention fluctuations on performance in the context of reading, sometimes having the sensation of “flying over” entire pages without truly absorbing their meaning. The previously mentioned studies led us to hypothesize that it may be possible to monitor the degree of attention during reading by measuring the spectral power in the alpha, beta, and gamma bands. Testing this hypothesis in healthy individuals assumes that attention-related power modulations in the lower alpha and beta bands as well as those in the gamma band can be detected with noninvasive recordings such as MEG or scalp level EEG. The present study tackles this issue by investigating the degree and precision by which gamma band power (and to a lesser extent, alpha and beta band) modulations of cortical activity can be recovered at the scalp level in individual subjects during attentive reading. The ability of surface-level electrophysiological measurements to detect gamma band response (GBR) is still a matter of debate. Although it has been more than a decade since the first reports of gamma band power modulations at the scalp level with EEG (Tallon-Baudry et al., 1996Tallon-Baudry and Bertrand, 1999, Rodriguez et al., 1999, Müller et al., 2000), with MEG reports arriving more recently (Kaiser et al., 2002Kaiser et al., 2004, Osipova et al., 2006, Vidal et al., 2006, Guggisberg et al., 2008, Dalal et al., 2008a), those responses have mostly been detected through population-level statistics. The relatively limited number of such reports has so far impeded the use of GBR as a research or clinical index even though increasing results from intracranial EEG suggest that high gamma band modulations could be used as functional markers (Lachaux et al., 2005Lachaux et al., 2006, Sederberg et al., 2003, Szurhaj et al., 2005, Tallon-Baudry et al., 2005, Tanji et al., 2005) tightly related to task-related hemodynamic variations measured by fMRI (Kim et al., 2004, Niessing et al., 2005, Lachaux et al., 2007). In contrast, other important electrophysiological indices such as the N100 evoked potential and mismatch negativity (MMN) are routinely used for both research and clinical purposes. The absence of GBR from clinical use thus cannot be fully explained by the younger age of the field and certainly not to a lack of interest by the EEG/MEG community; more likely, it is due to acknowledged methodological difficulties associated with extracting gamma response components from noninvasive recordings (Trujillo et al., 2005, Hoogenboom et al., 2006).
Moreover, the failure to detect EEG/MEG GBR in a given subject can be a source of confusion. For example, the technical limitations of EEG/MEG could have prevented detection, due to a low signal to noise ratio in the gamma band or insufficient spatial resolution. Alternatively, it could have been the result of individual subject differences. For example, a subject may have employed a different strategy or been less attentive than the others, and therefore employed different cortical mechanisms that may not have involved GBR.
To evaluate our ability to noninvasively recover cortical GBR, and their modulation by attention during reading, we designed an experiment to simultaneously record MEG and intracerebral EEG (iEEG) responses in epileptic patients implanted with depth electrodes and subdural strips for clinical reasons. Such simultaneous recordings provided a unique way to compare the spectral content of the scalp traces of GBR, as well as lower-frequency modulations, to those recorded directly from the cortex via the implanted depth electrodes. It was considered more advantageous to record iEEG simultaneously with MEG as opposed to scalp EEG, due to the difficulty recording from EEG electrode caps in patients with intracerebral electrodes, as well as the likely considerable changes in electric field propagation caused by the implants and craniotomies (Kirchberger et al., 1998).
We simultaneously recorded both MEG and iEEG data in four patients while they performed a simple reading task with variable levels of attention depending on instructions on whether to attend to the narratives formed by the words. In addition to being a proof of concept of simultaneous MEG and iEEG data acquisition, this study presents and discusses the MEG correlates of strong intracerebral alpha, beta, and gamma power modulations generated by word stimuli in temporal, occipital, and parietal lobes. The experimental design is the same used in a comprehensive iEEG study from our group (Jung et al., 2008), and originally implemented by Nobre et al. (1998) for an fMRI experiment.
Previous human studies have described simultaneous acquisition of MEG and iEEG to evaluate high-amplitude epileptiform spikes (Mikuni et al., 1997, Sutherling et al., 2001, Shigeto et al., 2002, Oishi et al., 2002). However, to our knowledge, this experiment comprises the first successful simultaneous recordings to study cognitive function.
Section snippets
Subjects
Simultaneous MEG and iEEG recordings were obtained from four patients with intractable epilepsy (referred to as Pt1–Pt4) who were candidates for resective surgery. Table 1 summarizes their clinical, neuroimaging, and video-EEG characteristics. Intracranial electrodes were implanted in these patients for preresection seizure localization and functional mapping of critical language and motor areas. Electrode implants were guided strictly by clinical indications and research recordings were
Evoked potentials and fields
Fig. 1 displays the event-related fields (MEG ERFs), beamformer source reconstructions, and potentials (iEEG ERPs) obtained by averaging directly the raw MEG and iEEG signals across trials for each condition. At all three levels of recording, evoked responses were clear and reproducible across conditions, with perhaps Pt2's source reconstruction suffering from increased noise. Intracerebral ERPs measured directly in the occipital and temporal lobes are shown together with the corresponding MEG
Discussion
Our primary objective was to test whether attention-related modulation of GBR and ABS observed with intracerebral measurements during a reading task could also be detected noninvasively with MEG. We found that this was indeed the case for the alpha/beta suppression, which was robust and consistent between the two attention conditions in all four patients, and was strengthened by attention in three patients in whom the occipital lobe was properly sampled by the MEG sensors (for practical
Acknowledgments
The authors would like to thank Florence Bouchet, Jean-Claude Bourzeix, and Laurent Hugueville for their technical help. JPL was supported by a grant from the Fondation Fyssen. SSD was supported by a Bourse Chateaubriand from the Government of France and a Marie Curie Fellowship from the European Commission.
References (69)
- et al.
Correlates of eye blinking as determined by synthetic aperture magnetometry
Clin. Neurophysiol.
(2006) - et al.
Event-related beamforming: a robust method for presurgical functional mapping using MEG
Clin. Neurophysiol.
(2007) - et al.
Five-dimensional neuroimaging: localization of the time-frequency dynamics of cortical activity
NeuroImage
(2008) - et al.
A multitrial analysis for revealing significant corticocortical networks in magnetoencephalography and electroencephalography
NeuroImage
(2003) - et al.
A multivariate, spatiotemporal analysis of electromagnetic time-frequency data of recognition memory
NeuroImage
(2003) - et al.
Selective visual–spatial attention alters induced gamma band responses in the human EEG
Clin. Neurophysiol.
(1999) - et al.
The missing link: analogous human and primate cortical gamma oscillations
NeuroImage
(2005) - et al.
Prestimulus oscillations predict visual perception performance between and within subjects
NeuroImage
(2007) - et al.
Localizing human visual gamma-band activity in frequency, time and space
NeuroImage
(2006) - et al.
Magnetoencephalographic gamma-band responses to illusory triangles in humans
NeuroImage
(2004)
Spatial relationship between neuronal activity and BOLD functional MRI
NeuroImage
Estimating the time-course of coherence between single-trial brain signals: an introduction to wavelet coherence
Neurophysiol. Clin.
The many faces of the gamma band response to complex visual stimuli
NeuroImage
Intracerebral dynamics of saccade generation in the human frontal eye field and supplementary eye field
NeuroImage
Simultaneous recording of epileptiform discharges by MEG and subdural electrodes in temporal lobe epilepsy
NeuroImage
Modulation of induced gamma band activity in the human EEG by attention and visual information processing
Int. J. Psychophysiol.
New vistas for alpha-frequency band oscillations
Trends Neurosci.
Frequency dependence of the transmission of the EEG from cortex to scalp
Electroencephalogr. Clin. Neurophysiol.
Neuronal synchrony: a versatile code for the definition of relations?
Neuron
Oscillatory gamma activity in humans and its role in object representation
Trends Cogn. Sci.
EEG phase synchrony differences across visual perception conditions may depend on recording and analysis methods
Clin. Neurophysiol.
Thinking activates EMG in scalp electrical recordings
Clin. Neurophysiol.
Transient induced gamma-band response in EEG as a manifestation of miniature saccades
Neuron
A probabilistic algorithm integrating source localization and noise suppression for MEG and EEG data
NeuroImage
A model for frequency dependence of conductivities of the live human skull
Brain Topogr.
Tactile spatial attention enhances gamma-band activity in somatosensory cortex and reduces low-frequency activity in parieto-occipital areas
J. Neurosci.
Parallel and serial neural mechanisms for visual search in macaque area V4
Science
Spatiotemporal mapping of cortical activity accompanying voluntary movements using an event-related beamforming approach
Hum. Brain Mapp.
Functional mapping of human sensorimotor cortex with electrocorticographic spectral analysis. II. Event-related synchronization in the gamma band
Brain
Functional mapping of human sensorimotor cortex with electrocorticographic spectral analysis. I. Alpha and beta event-related desynchronization
Brain
NUTMEG: a neuromagnetic source reconstruction toolbox
Neurol. Clin. Neurophysiol.
Experimental calculation of magnetic lead fields using MEG simultaneously acquired with intracranial EEG [abstract]
Chronic thalamic stimulation with three-dimensional MR stereotactic guidance
AJNR Am. J. Neuroradiol.
Outcome with respect to epileptic seizures
Cited by (104)
Virtual MEG sensors based on beamformer and independent component analysis can reconstruct epileptic activity as measured on simultaneous intracerebral recordings
2022, NeuroImageCitation Excerpt :However, these studies were not conducted concurrently, so it is hard to compare and interpret the interaction between neural oscillations and spikes recorded in different sessions, at varying vigilance levels and dosages of medication (Dalal et al., 2009). Simultaneously acquired MEG and intracerebral stereotaxic EEG (SEEG) (Talairach and Bancaud, 1973) provides a unique opportunity to determine if the interictal activity detected by MEG and SEEG corresponds to identical underlying epileptic activity and enables us to examine the consistency between the modalities in the exact same brain state (Dalal et al., 2009; Dubarry et al., 2014; He et al., 2019). Our goal was to improve the reconstruction of source-level epileptic traces and to compare the results to simultaneously acquired SEEG data.
Intracranial studies of cognition in humans
2021, Encyclopedia of Behavioral Neuroscience: Second Edition