Interpretation of high-resolution current source density profiles: a simulation of sublaminar contributions to the visual evoked potential

Volumetric mesoscopic electrophysiology: a new imaging modality for the nonhuman primate

The primate brain is a densely interconnected organ whose function is best understood by recording from the entire structure in parallel, rather than parts of it in sequence. However, available methods either have limited temporal resolution (functional magnetic resonance imaging; fMRI), limited spatial resolution (macroscopic electroencephalography), or a limited field of view (microscopic electrophysiology). To address this need, we developed a volumetric, mesoscopic recording approach (MePhys) by tessellating the volume of a monkey hemisphere with 992 electrode contacts that were distributed across 62 chronically implanted multielectrode shafts. We showcase the scientific promise of MePhys by describing the functional interactions of local field potentials between the more than 300,000 simultaneously recorded pairs of electrodes. We find that a subanesthetic dose of ketamine-believed to mimic certain aspects of psychosis-can create a pronounced state of functional disconnection and prevent the formation of stable large-scale intrinsic states. We conclude that MePhys provides a new and fundamentally distinct window into brain function whose unique profile of strengths and weaknesses complements existing approaches in synergistic ways.NEW & NOTEWORTHY We created a new imaging modality for the nonhuman primate, mesoscopic electrophysiology, or MePhys by sampling local field potentials (LFPs) in a dense three-dimensional grid from across the volume of one entire hemisphere. MePhys combines the millisecond temporal resolution of electrophysiology with the large field of view and millimeter spatial resolution of functional magnetic resonance imaging (fMRI). MePhys' unique profile of strengths and limitations makes it an ideal imaging method for the nonhuman primate brain observatories of the future.

Contributions of action potentials to scalp EEG: Theory and biophysical simulations

Differences in the apparent 1/f component of neural power spectra require correction depending on the underlying neural mechanisms, which remain incompletely understood. Past studies suggest that neuronal spiking produces broadband signals and shapes the spectral trend of invasive macroscopic recordings, but it is unclear to what extent action potentials (APs) influence scalp EEG. Here, we combined biophysical simulations with statistical modelling to examine the amplitude and spectral content of scalp potentials generated by the electric fields from spiking activity. In physiological parameter regimes, we found that APs contribute negligibly to the EEG spectral trend. Consistent with this, comparing our biophysical simulations with previously published data from pharmacologically paralyzed subjects suggested that the EEG spectral trend can be explained by a combination of synaptic timescales and electromyogram contamination. We also modelled rhythmic EEG generation, finding that APs can generate detectable narrowband power between approximately 60 and 1000 Hz, reaching frequencies much faster than would be possible from synaptic currents. Finally, we show that different spectral detrending strategies are required for AP generated oscillations compared to synaptically generated oscillations, suggesting that existing detrending methods for EEG spectra need to be modified for high frequency signals.

Circuit dynamics of superficial and deep CA1 pyramidal cells and inhibitory cells in freely moving macaques

Diverse neuron classes in hippocampal CA1 have been identified through the heterogeneity of their cellular/molecular composition. How these classes relate to hippocampal function and the network dynamics that support cognition in primates remains unclear. Here, we report inhibitory functional cell groups in CA1 of freely moving macaques whose diverse response profiles to network states and each other suggest distinct and specific roles in the functional microcircuit of CA1. In addition, pyramidal cells that were grouped by their superficial or deep layer position differed in firing rate, burstiness, and sharp-wave ripple-associated firing. They also showed strata-specific spike-timing interactions with inhibitory cell groups, suggestive of segregated neural populations. Furthermore, ensemble recordings revealed that cell assemblies were preferentially organized according to these strata. These results suggest that hippocampal CA1 in freely moving macaques bears a sublayer-specific circuit organization that may shape its role in cognition.

Circuit dynamics of superficial and deep CA1 pyramidal cells and inhibitory cells in freely-moving macaques

Diverse neuron classes in hippocampal CA1 have been identified through the heterogeneity of their cellular/molecular composition. How these classes relate to hippocampal function and the network dynamics that support cognition in primates remains unclear. Here we report inhibitory functional cell groups in CA1 of freely-moving macaques whose diverse response profiles to network states and each other suggest distinct and specific roles in the functional microcircuit of CA1. In addition, pyramidal cells that were segregated into superficial and deep layers differed in firing rate, burstiness, and sharp-wave ripple-associated firing. They also showed strata-specific spike-timing interactions with inhibitory cell groups, suggestive of segregated neural populations. Furthermore, ensemble recordings revealed that cell assemblies were preferentially organized according to these strata. These results suggest sublayer-specific circuit organization in hippocampal CA1 of the freely-moving macaques that may underlie its role in cognition.

Volumetric mesoscopic electrophysiology: a new imaging modality for the non-human primate

The primate brain is a densely interconnected organ whose function is best understood by recording from the entire structure in parallel, rather than parts of it in sequence. However, available methods either have limited temporal resolution (functional magnetic resonance imaging), limited spatial resolution (macroscopic electroencephalography), or a limited field of view (microscopic electrophysiology). To address this need, we developed a volumetric, mesoscopic recording approach ( MePhys ) by tessellating the volume of a monkey hemisphere with 992 electrode contacts that were distributed across 62 chronically implanted multi-electrode shafts. We showcase the scientific promise of MePhys by describing the functional interactions of local field potentials between the more than 300,000 simultaneously recorded pairs of electrodes. We find that a subanesthetic dose of ketamine -believed to mimic certain aspects of psychosis- can create a pronounced state of functional disconnection and prevent the formation of stable large-scale intrinsic states. We conclude that MePhys provides a new and fundamentally distinct window into brain function whose unique profile of strengths and weaknesses complements existing approaches in synergistic ways.

Pre-Stimulus Alpha-Band Phase Gates Early Visual Cortex Responses

Alpha-band (8-13 Hz) oscillations have been shown to phasically inhibit perceptual reports in human observers, yet the underlying physiological mechanism of this effect is debated. According to contrasting models, based primarily on animal experiments, alpha activity is thought to either originate from specialized cells in the visual thalamus and periodically inhibit the relay of visual information to the primary visual cortex (V1) in a feedforward manner, or to propagate from higher visual areas back to V1 in a feedback manner. Human neurophysiological evidence in favor of either hypothesis, both, or neither, has been limited. To help address this issue, we explored the link between pre-stimulus alpha phase and visual electroencephalography (EEG) responses thought to arise from afferent input onto human V1. Specially-designed visual stimuli were used to elicit large amplitude C1 event-related potentials (ERP), with polarity, topography, and timing indicative of striate genesis. Single-trial circular-linear associations between pre-stimulus phase and post-stimulus global field power (GFP) during the C1 time window revealed significant effects peaking in the alpha frequency band. Control analyses ruling out the potential confound of post-stimulus data bleeding into the pre-stimulus window demonstrated that GFP amplitude decreases as pre-stimulus alpha phase deviates from an individual's preferred phase. These findings demonstrate an early locus - suggesting that the phase of pre-stimulus alpha oscillations could modulate visual processing by gating the feedforward flow of sensory input between the thalamus and V1, although other models are potentially compatible.

Scalp-recorded N40 visual evoked potential: Sensory and attentional properties

N40 is a well-known component of evoked potentials with respect to the auditory and somatosensory modality but not much recognized with regard to the visual modality. To be detected with event-related potentials (ERPs), it requires an optimal signal-to-noise ratio. To investigate the nature of visual N40, we recorded EEG/ERP signals from 20 participants. Each of them was presented with 1800 spatial frequency gratings of 0.75, 1.5, 3 and 6 c/deg. Data were collected from 128 sites while participants were engaged in both passive viewing and attention conditions. N40 (30-55 ms) was modulated by alertness and selective attention; in fact, it was larger to targets than irrelevant and passively viewed spatial frequency gratings. Its strongest intracranial sources were the bilateral thalamic nuclei of pulvinar, according to swLORETA. The active network included precuneus, insula and inferior parietal lobule. An N80 component (60-90 ms) was also identified, which was larger to targets than irrelevant/passive stimuli and more negative to high than low spatial frequencies. In contrast, N40 was not sensitive to spatial frequency per se, nor did it show a polarity inversion as a function of spatial frequency. Attention, alertness and spatial frequency effects were also found for the later components P1, N2 and P300. The attentional effects increased in magnitude over time. The data showed that ERPs can pick up the earliest synchronized activity, deriving in part from thalamic nuclei, before the visual information has actually reached the occipital cortex.

The Hippocampus and Amygdala Are Integrators of Neocortical Influence: A CorticoCortical Evoked Potential Study

Brain stimulation is increasingly viewed as an effective approach to treat neuropsychiatric disease. The brain's organization in distributed networks suggests that the activity of a remote brain structure could be modulated by stimulating cortical areas that strongly connect to the target. Most connections between cerebral areas are asymmetric, and a better understanding of the relative direction of information flow along connections could improve the targeting of stimulation to influence deep brain structures. The hippocampus and amygdala, two deep-situated structures that are crucial to memory and emotions, respectively, have been implicated in multiple neurological and psychiatric disorders. We explored the directed connectivity between the hippocampus and amygdala and the cerebral cortex in patients implanted with intracranial electrodes using corticocortical evoked potentials (CCEPs) evoked by single-pulse electrical stimulation. The hippocampus and amygdala were connected with most of the cortical mantle, either directly or indirectly, with the inferior temporal cortex being most directly connected. Because CCEPs assess the directionality of connections, we could determine that incoming connections from cortex to hippocampus were more direct than outgoing connections from hippocampus to cortex. We found a similar, although smaller, tendency for connections between the amygdala and cortex. Our results support the roles of the hippocampus and amygdala to be integrators of widespread cortical influence. These results can inform the targeting of noninvasive neurostimulation to influence hippocampus and amygdala function.

Distinct Laminar Processing of Local and Global Context in Primate Primary Visual Cortex

Visual perception is affected by spatial context. In visual cortex, neuronal responses to stimuli inside the receptive field (RF) are suppressed by stimuli in the RF surround. To understand the circuits and cortical layers processing spatial context, we simultaneously recorded across all layers of macaque primary visual cortex while presenting stimuli at increasing distances from the recorded cells' RF. We find that near versus far-surround stimuli activate distinct layers, thus revealing unique laminar contributions to the processing of local and global spatial context. Stimuli in the near-surround evoke the earliest subthreshold responses in superficial and upper-deep layers, and earliest suppression of spiking responses in superficial layers. Conversely, far-surround stimuli evoke the earliest subthreshold responses in feedback-recipient layer 1 and lower-deep layers, and earliest suppression of spiking responses almost simultaneously in all layers, except 4C, where suppression emerges last. Our results suggest distinct circuits for local and global signal integration.

Investigating large-scale brain dynamics using field potential recordings: analysis and interpretation

New technologies to record electrical activity from the brain on a massive scale offer tremendous opportunities for discovery. Electrical measurements of large-scale brain dynamics, termed field potentials, are especially important to understanding and treating the human brain. Here, our goal is to provide best practices on how field potential recordings (electroencephalograms, magnetoencephalograms, electrocorticograms and local field potentials) can be analyzed to identify large-scale brain dynamics, and to highlight critical issues and limitations of interpretation in current work. We focus our discussion of analyses around the broad themes of activation, correlation, communication and coding. We provide recommendations for interpreting the data using forward and inverse models. The forward model describes how field potentials are generated by the activity of populations of neurons. The inverse model describes how to infer the activity of populations of neurons from field potential recordings. A recurring theme is the challenge of understanding how field potentials reflect neuronal population activity given the complexity of the underlying brain systems.

Ongoing Alpha Activity in V1 Regulates Visually Driven Spiking Responses

The interlaminar connections in the primate primary visual cortex (V1) are well described, as is the presence of ongoing alpha-range (7-14 Hz) fluctuations in this area. Less well understood is how these interlaminar connections and ongoing fluctuations contribute to the regulation of visual spiking responses. Here, we investigate the relationship between alpha fluctuations and spiking responses to visual stimuli across cortical layers. Using laminar probes in macaque V1, we show that neural firing couples with the phase of alpha fluctuations, and that magnitude of this coupling is particularly pronounced during visual stimulation. The strongest modulation of spiking activity was observed in layers 2/3. Alpha-spike coupling and current source density analysis pointed to an infragranular origin of the alpha fluctuations. Taken together, these results indicate that ongoing infragranular alpha-range fluctuations in V1 play a role in regulating columnar visual activity.

Compensating Level-Dependent Frequency Representation in Auditory Cortex by Synaptic Integration of Corticocortical Input

Robust perception of auditory objects over a large range of sound intensities is a fundamental feature of the auditory system. However, firing characteristics of single neurons across the entire auditory system, like the frequency tuning, can change significantly with stimulus intensity. Physiological correlates of level-constancy of auditory representations hence should be manifested on the level of larger neuronal assemblies or population patterns. In this study we have investigated how information of frequency and sound level is integrated on the circuit-level in the primary auditory cortex (AI) of the Mongolian gerbil. We used a combination of pharmacological silencing of corticocortically relayed activity and laminar current source density (CSD) analysis. Our data demonstrate that with increasing stimulus intensities progressively lower frequencies lead to the maximal impulse response within cortical input layers at a given cortical site inherited from thalamocortical synaptic inputs. We further identified a temporally precise intercolumnar synaptic convergence of early thalamocortical and horizontal corticocortical inputs. Later tone-evoked activity in upper layers showed a preservation of broad tonotopic tuning across sound levels without shifts towards lower frequencies. Synaptic integration within corticocortical circuits may hence contribute to a level-robust representation of auditory information on a neuronal population level in the auditory cortex.

Dipolar extracellular potentials generated by axonal projections

Extracellular field potentials (EFPs) are an important source of information in neuroscience, but their physiological basis is in many cases still a matter of debate. Axonal sources are typically discounted in modeling and data analysis because their contributions are assumed to be negligible. Here, we established experimentally and theoretically that contributions of axons to EFPs can be significant. Modeling action potentials propagating along axons, we showed that EFPs were prominent in the presence of terminal zones where axons branch and terminate in close succession, as found in many brain regions. Our models predicted a dipolar far field and a polarity reversal at the center of the terminal zone. We confirmed these predictions using EFPs from the barn owl auditory brainstem where we recorded in nucleus laminaris using a multielectrode array. These results demonstrate that axonal terminal zones can produce EFPs with considerable amplitude and spatial reach.

1D Current Source Density (CSD) Estimation in Inverse Theory: A Unified Framework for Higher-Order Spectral Regularization of Quadrature and Expansion-Type CSD Methods

Estimation of current source density (CSD) from the low-frequency part of extracellular electric potential recordings is an unstable linear inverse problem. To make the estimation possible in an experimental setting where recordings are contaminated with noise, it is necessary to stabilize the inversion. Here we present a unified framework for zero- and higher-order singular-value-decomposition (SVD)-based spectral regularization of 1D (linear) CSD estimation from local field potentials. The framework is based on two general approaches commonly employed for solving inverse problems: quadrature and basis function expansion. We first show that both inverse CSD (iCSD) and kernel CSD (kCSD) fall into the category of basis function expansion methods. We then use these general categories to introduce two new estimation methods, quadrature CSD (qCSD), based on discretizing the CSD integral equation with a chosen quadrature rule, and representer CSD (rCSD), an even-determined basis function expansion method that uses the problem's data kernels (representers) as basis functions. To determine the best candidate methods to use in the analysis of experimental data, we compared the different methods on simulations under three regularization schemes (Tikhonov, tSVD, and dSVD), three regularization parameter selection methods (NCP, L-curve, and GCV), and seven different a priori spatial smoothness constraints on the CSD distribution. This resulted in a comparison of 531 estimation schemes. We evaluated the estimation schemes according to their source reconstruction accuracy by testing them using different simulated noise levels, lateral source diameters, and CSD depth profiles. We found that ranking schemes according to the average error over all tested conditions results in a reproducible ranking, where the top schemes are found to perform well in the majority of tested conditions. However, there is no single best estimation scheme that outperforms all others under all tested conditions. The unified framework we propose expands the set of available estimation methods, provides increased flexibility for 1D CSD estimation in noisy experimental conditions, and allows for a meaningful comparison between estimation schemes.

Quantification of mid and late evoked sinks in laminar current source density profiles of columns in the primary auditory cortex

Current source density (CSD) analysis assesses spatiotemporal synaptic activations at somatic and/or dendritic levels in the form of depolarizing current sinks. Whereas many studies have focused on the short (<50 ms) latency sinks, associated with thalamocortical projections, sinks with longer latencies have received less attention. Here, we analyzed laminar CSD patterns for the first 600 ms after stimulus onset in the primary auditory cortex of Mongolian gerbils. By applying an algorithm for contour calculation, three distinct mid and four late evoked sinks were identified in layers I, III, Va, VIa, and VIb. Our results further showed that the patterns of intracortical information-flow remained qualitatively similar for low and for high sound pressure level stimuli at the characteristic frequency (CF) as well as for stimuli ± 1 octave from CF. There were, however, differences associated with the strength, vertical extent, onset latency, and duration of the sinks for the four stimulation paradigms used. Stimuli one octave above the most sensitive frequency evoked a new, and quite reliable, sink in layer Va whereas low level stimulation led to the disappearance of the layer VIb sink. These data indicate the presence of input sources specifically activated in response to level and/or frequency parameters. Furthermore, spectral integration above vs. below the CF of neurons is asymmetric as illustrated by CSD profiles. These results are important because synaptic feedback associated with mid and late sinks—beginning at 50 ms post stimulus latency—is likely crucial for response modulation resulting from higher order processes like memory, learning or cognitive control.

Surface Laplacians (SL) and phase properties of EEG rhythms: Simulated generators in a volume-conduction model

Surface Laplacian (SL) methods offer advantages in spectral analysis owing to the well-known implications of volume conduction. Although recognition of the superiority of SL over reference-dependent measures is widespread, well-reasoned cautions have precluded their universal adoption. Notably, the expected selectivity of SL for superficial rather than deep generators has relegated SL to the role of an add-on to conventional analyses, rather than as an independent area of inquiry, despite empirical findings supporting the consistency and replicability of physiological effects of interest. It has also been reasoned that the contrast-enhancing effects of SL necessarily make it insensitive to broadly distributed generators, including those suspected for oscillatory rhythms such as EEG alpha. These concerns are further exacerbated for phase-sensitive measures (e.g., phase-locking, coherence), where key features of physiological generators have yet to be evaluated. While the neuronal generators of empirically-derived EEG measures cannot be precisely known due to the inverse problem, simple dipole generator configurations can be simulated using a 4-sphere head model and linearly combined. We simulated subdural and deep generators and distributed dipole layers using sine and cosine waveforms, quantified at 67-scalp sites corresponding to those used in previous research. Reference-dependent (nose, average, mastoids reference) EEG and corresponding SL topographies were used to probe signal fidelity in the topography of the measured amplitude spectra, phase and coherence of sinusoidal stimuli at and between "active" recording sites. SL consistently outperformed the conventional EEG measures, indicating that continued reluctance by the research community is unfounded.

The advantages of the surface Laplacian in brain-computer interface research

Brain-computer interface (BCI) systems frequently use signal processing methods, such as spatial filtering, to enhance performance. The surface Laplacian can reduce spatial noise and aid in identification of sources. In BCI research, these two functions of the surface Laplacian correspond to prediction accuracy and signal orthogonality. In the present study, an off-line analysis of data from a sensorimotor rhythm-based BCI task dissociated these functions of the surface Laplacian by comparing nearest-neighbor and next-nearest neighbor Laplacian algorithms. The nearest-neighbor Laplacian produced signals that were more orthogonal while the next-nearest Laplacian produced signals that resulted in better accuracy. Both prediction and signal identification are important for BCI research. Better prediction of user's intent produces increased speed and accuracy of communication and control. Signal identification is important for ruling out the possibility of control by artifacts. Identifying the nature of the control signal is relevant both to understanding exactly what is being studied and in terms of usability for individuals with limited motor control.

Issues and considerations for using the scalp surface Laplacian in EEG/ERP research: A tutorial review

Despite the recognition that the surface Laplacian may counteract adverse effects of volume conduction and recording reference for surface potential data, electrophysiology as a discipline has been reluctant to embrace this approach for data analysis. The reasons for such hesitation are manifold but often involve unfamiliarity with the nature of the underlying transformation, as well as intimidation by a perceived mathematical complexity, and concerns of signal loss, dense electrode array requirements, or susceptibility to noise. We revisit the pitfalls arising from volume conduction and the mandated arbitrary choice of EEG reference, describe the basic principle of the surface Laplacian transform in an intuitive fashion, and exemplify the differences between common reference schemes (nose, linked mastoids, average) and the surface Laplacian for frequently-measured EEG spectra (theta, alpha) and standard event-related potential (ERP) components, such as N1 or P3. We specifically review common reservations against the universal use of the surface Laplacian, which can be effectively addressed by employing spherical spline interpolations with an appropriate selection of the spline flexibility parameter and regularization constant. We argue from a pragmatic perspective that not only are these reservations unfounded but that the continued predominant use of surface potentials poses a considerable impediment on the progress of EEG and ERP research.

Microcircuitry of agranular frontal cortex: contrasting laminar connectivity between occipital and frontal areas

Neocortex is striking in its laminar architecture. Tracer studies have uncovered anatomical connectivity among laminae, but the functional connectivity between laminar compartments is still largely unknown. Such functional connectivity can be discerned through spontaneous neural correlations during rest. Previous work demonstrated a robust pattern of mesoscopic resting-state connectivity in macaque primary visual cortex (V1) through interlaminar cross-frequency coupling. Here we investigated whether this pattern generalizes to other cortical areas by comparing resting-state laminar connectivity between V1 and the supplementary eye field (SEF), a frontal area lacking a granular layer 4 (L4). Local field potentials (LFPs) were recorded with linear microelectrode arrays from all laminae of granular V1 and agranular SEF while monkeys rested in darkness. We found substantial differences in the relationship between the amplitude of gamma-band (>30 Hz) LFP and the phase of alpha-band (7-14 Hz) LFP between these areas. In V1, gamma amplitudes in L2/3 and L5 were coupled with alpha-band LFP phase in L5, as previously described. In contrast, in SEF phase-amplitude coupling was prominent within L3 and much weaker across layers. These results suggest that laminar interactions in agranular SEF are unlike those in granular V1. Thus the intrinsic functional connectivity of the cortical microcircuit does not seem to generalize across cortical areas.

Convergent approaches toward the study of multisensory perception

Classical analytical approaches for examining multisensory processing in individual neurons have relied heavily on changes in mean firing rate to assess the presence and magnitude of multisensory interaction. However, neurophysiological studies within individual sensory systems have illustrated that important sensory and perceptual information is encoded in forms that go beyond these traditional spike-based measures. Here we review analytical tools as they are used within individual sensory systems (auditory, somatosensory, and visual) to advance our understanding of how sensory cues are effectively integrated across modalities (e.g., audiovisual cues facilitating speech processing). Specifically, we discuss how methods used to assess response variability (Fano factor, or FF), local field potentials (LFPs), current source density (CSD), oscillatory coherence, spike synchrony, and receiver operating characteristics (ROC) represent particularly promising tools for understanding the neural encoding of multisensory stimulus features. The utility of each approach and how it might optimally be applied toward understanding multisensory processing is placed within the context of exciting new data that is just beginning to be generated. Finally, we address how underlying encoding mechanisms might shape-and be tested alongside with-the known behavioral and perceptual benefits that accompany multisensory processing.

Ultra high-resolution fMRI and electrophysiology of the rat primary somatosensory cortex

High-resolution functional-magnetic-resonance-imaging (fMRI) has been used to study brain functions at increasingly finer scale, but whether fMRI can accurately reflect layer-specific neuronal activities is less well understood. The present study investigated layer-specific cerebral-blood-volume (CBV) fMRI and electrophysiological responses in the rat cortex. CBV fMRI at 40×40 μm in-plane resolution was performed on an 11.7-T scanner. Electrophysiology used a 32-channel electrode array that spanned the entire cortical depth. Graded electrical stimulation was used to study activations in different cortical layers, exploiting the notion that most of the sensory-specific neurons are in layers II-V and most of the nociceptive-specific neurons are in layers V-VI. CBV response was strongest in layer IV of all stimulus amplitudes. Current source density analysis showed strong sink currents at cortical layers IV and VI. Multi-unit activities mainly appeared at layers IV-VI and peaked at layer V. Although our measures showed scaled activation profiles during modulation of stimulus amplitude and failed to detect specific recruitment at layers V and VI during noxious electrical stimuli, there appears to be discordance between CBV fMRI and electrophysiological peak responses, suggesting neurovascular uncoupling at laminar resolution. The technique implemented in the present study offers a means to investigate intracortical neurovascular function in the normal and diseased animal models at laminar resolution.

Effects of neural synchrony on surface EEG

It has long been assumed that the surface electroencephalography (EEG) signal depends on both the amplitude and spatial synchronization of underlying neural activity, though isolating their respective contribution remains elusive. To address this, we made simultaneous surface EEG measurements along with intracortical recordings of local field potentials (LFPs) in the primary visual cortex of behaving nonhuman primates. We found that trial-by-trial fluctuations in EEG power could be explained by a linear combination of LFP power and interelectrode temporal synchrony. This effect was observed in both stimulus and stimulus-free conditions and was particularly strong in the gamma range (30-100 Hz). Subsequently, we used pharmacological manipulations to show that neural synchrony can produce a positively modulated EEG signal even when the LFP signal is negatively modulated. Taken together, our results demonstrate that neural synchrony can modulate EEG signals independently of amplitude changes in neural activity. This finding has strong implications for the interpretation of EEG in basic and clinical research, and helps reconcile EEG response discrepancies observed in different modalities (e.g., EEG vs. functional magnetic resonance imaging) and different spatial scales (e.g., EEG vs. intracranial EEG).

Generator localization by current source density (CSD): implications of volume conduction and field closure at intracranial and scalp resolutions

The topographic ambiguity and reference-dependency that has plagued EEG/ERP research throughout its history are largely attributable to volume conduction, which may be concisely described by a vector form of Ohm's Law. This biophysical relationship is common to popular algorithms that infer neuronal generators via inverse solutions. It may be further simplified as Poisson's source equation, which identifies underlying current generators from estimates of the second spatial derivative of the field potential (Laplacian transformation). Intracranial current source density (CSD) studies have dissected the "cortical dipole" into intracortical sources and sinks, corresponding to physiologically-meaningful patterns of neuronal activity at a sublaminar resolution, much of which is locally cancelled (i.e., closed field). By virtue of the macroscopic scale of the scalp-recorded EEG, a surface Laplacian reflects the radial projections of these underlying currents, representing a unique, unambiguous measure of neuronal activity at scalp. Although the surface Laplacian requires minimal assumptions compared to complex, model-sensitive inverses, the resulting waveform topographies faithfully summarize and simplify essential constraints that must be placed on putative generators of a scalp potential topography, even if they arise from deep or partially-closed fields. CSD methods thereby provide a global empirical and biophysical context for generator localization, spanning scales from intracortical to scalp recordings.

How local is the local field potential?

Local field potentials (LFPs) are of growing importance in neurophysiological investigations. LFPs supplement action potential recordings by indexing activity relevant to EEG, magnetoencephalographic, and hemodynamic (fMRI) signals. Recent reports suggest that LFPs reflect activity within very small domains of several hundred micrometers. We examined this conclusion by comparing LFP, current source density (CSD), and multiunit activity (MUA) signals in macaque auditory cortex. Estimated by frequency tuning bandwidths, these signals' "listening areas" differ systematically with an order of MUA < CSD < LFP. Computational analyses confirm that observed LFPs receive local contributions. Direct measurements indicate passive spread of LFPs to sites more than a centimeter from their origins. These findings appear to be independent of the frequency content of the LFP. Our results challenge the idea that LFP recordings typically integrate over extremely circumscribed local domains. Rather, LFPs appear as a mixture of local potentials with "volume conducted" potentials from distant sites.

Temperature- and concentration-dependence of kainate-induced γ oscillation in rat hippocampal slices under submerged condition

AIM

Fast neuronal network oscillation at the γ frequency band (γ oscillation: 30-80 Hz) has been studied extensively in hippocampal slices under interface recording condition. The aim of this study is to establish a method for recording γ oscillation in submerged hippocampal slices that allows simultaneously monitoring γ oscillation and the oscillation-related intracellular events, such as intracellular Ca(2+) concentration or mitochondrial membrane potentials.

METHODS

Horizontal hippocampal slices (thickness: 300 μm) of adult rats were prepared and placed in a submerged or an interface chamber. Extracellular field recordings were made in the CA3c pyramidal layer of the slices. Kainate, an AMPA/kainate receptor agonist, was applied via perfusion. Data analysis was performed off-line.

RESULTS

Addition of kainate (25-1000 nmol/L) induced γ oscillation in both the submerged and interface slices. Kainate increased the γ power in a concentration-dependent manner, but the duration of steady state oscillation was reduced at higher concentrations of kainate. Long-lasting γ oscillation was maintained at the concentrations of 100-300 nmol/L. Under submerged condition, γ oscillation was temperature-dependent, with the maximum power achieved at 29 °C. The induction of γ oscillation under submerged condition also required a fast rate of perfusion (5-7 mL/min) and showed a fast dynamic during development and after the washout.

CONCLUSION

The kainite-induced γ oscillation recorded in submerged rat hippocampal slices is useful for studying the intracellular events related to neuronal network activities and may represent a model to reveal the mechanisms underlying the normal neuronal synchronizations and diseased conditions.

Functional development of large-scale sensorimotor cortical networks in the brain

Large-scale neuronal networks integrating several cortical areas mediate the complex functions of the brain such as sensorimotor integration. Little is known about the functional development of these networks and the maturational processes by which distant networks become functionally connected. We addressed this question in the postnatal rat sensorimotor system. Using epicranial multielectrode grids that span most of the cortical surface and intracortical electrodes, we show that sensory evoked cortical responses continuously maturate throughout the first 3 weeks with the strongest developmental changes occurring in a very short time around postnatal day 13 (P13). Before P13, whisker stimulation evokes slow, initially surface-negative activity restricted mostly to the lateral parietal area of the contralateral hemisphere. In a narrow time window of ∼48 h around P13, a new early, sharp surface-positive component emerges that coincides with subsequent propagation of activity to sensory and motor areas of both hemispheres. Our data show that this new component developing at the end of the second week corresponds principally to functional maturation of the supragranular cortical layers and appears to be crucial for the functional associations in the large-scale sensorimotor cortical network. It goes along with the onset of whisking behavior, as well as major synaptic and functional changes within the S1 cortex that are known to develop during this period.

Neuronal mechanisms and attentional modulation of corticothalamic α oscillations

Field potential oscillations in the ∼10 Hz range are known as the alpha rhythm. The genesis and function of alpha has been the subject of intense investigation for the past 80 years. Whereas early work focused on the thalamus as the pacemaker of alpha rhythm, subsequent slice studies revealed that pyramidal neurons in the deep layers of sensory cortices are capable of oscillating in the alpha frequency range independently. How thalamic and cortical generating mechanisms in the intact brain might interact to shape the organization and function of alpha oscillations remains unclear. We addressed this problem by analyzing laminar profiles of local field potential and multiunit activity (MUA) recorded with linear array multielectrodes from the striate cortex of two macaque monkeys performing an intermodal selective attention task. Current source density (CSD) analysis was combined with CSD-MUA coherence to identify intracortical alpha current generators and assess their potential for pacemaking. Coherence and Granger causality analysis was applied to delineate the patterns of interaction among different alpha current generators. We found that (1) separable alpha current generators are located in superficial, granular, and deep layers, with both layer 4C and deep layers containing primary local pacemaking generators, suggesting the involvement of the thalamocortical network, and (2) visual attention reduces the magnitude of alpha oscillations as well as the level of alpha interactions, consistent with numerous reports of occipital alpha reduction with visual attention in human EEG. There is also indication that alpha oscillations in the lateral geniculate cohere with those in V1.

Infragranular sources of sustained local field potential responses in macaque primary visual cortex

A local field potential (LFP) response can be measured throughout the visual cortex in response to the abrupt appearance of a visual stimulus. Averaging LFP responses to many stimulus presentations isolates transient, phase-locked components of the response that are consistent from trial to trial. However, stimulus responses are also composed of sustained components, which differ in their phase from trial to trial and therefore must be evaluated using other methods, such as computing the power of the response of each trial before averaging. Here, we investigate the basis of phase-locked and non-phase-locked LFP responses in the primary visual cortex of the macaque monkey using a novel variant of current source density (CSD) analysis. We applied a linear array of electrode contacts spanning the thickness of the cortex to measure the LFP and compute band-limited CSD power to identify the laminar sites of persistent current exchange that may be the basis of sustained visual LFP responses. In agreement with previous studies, we found a short-latency phase-locked current sink, thought to correspond to thalamocortical input to layer 4C. In addition, we found a prominent non-phase-locked component of the CSD that persisted as long as the stimulus was physically present. The latter was relatively broadband, lasted throughout the stimulus presentation, and was centered ∼500 μm deeper than the initial current sink. These findings demonstrate a fundamental difference in the neural mechanisms underlying the initial and sustained processing of simple visual stimuli in the V1 microcircuit.

Spectral integration in primary auditory cortex attributable to temporally precise convergence of thalamocortical and intracortical input

Primary sensory cortex integrates sensory information from afferent feedforward thalamocortical projection systems and convergent intracortical microcircuits. Both input systems have been demonstrated to provide different aspects of sensory information. Here we have used high-density recordings of laminar current source density (CSD) distributions in primary auditory cortex of Mongolian gerbils in combination with pharmacological silencing of cortical activity and analysis of the residual CSD, to dissociate the feedforward thalamocortical contribution and the intracortical contribution to spectral integration. We found a temporally highly precise integration of both types of inputs when the stimulation frequency was in close spectral neighborhood of the best frequency of the measurement site, in which the overlap between both inputs is maximal. Local intracortical connections provide both directly feedforward excitatory and modulatory input from adjacent cortical sites, which determine how concurrent afferent inputs are integrated. Through separate excitatory horizontal projections, terminating in cortical layers II/III, information about stimulus energy in greater spectral distance is provided even over long cortical distances. These projections effectively broaden spectral tuning width. Based on these data, we suggest a mechanism of spectral integration in primary auditory cortex that is based on temporally precise interactions of afferent thalamocortical inputs and different short- and long-range intracortical networks. The proposed conceptual framework allows integration of different and partly controversial anatomical and physiological models of spectral integration in the literature.

Laminar analysis of slow wave activity in humans

Brain electrical activity is largely composed of oscillations at characteristic frequencies. These rhythms are hierarchically organized and are thought to perform important pathological and physiological functions. The slow wave is a fundamental cortical rhythm that emerges in deep non-rapid eye movement sleep. In animals, the slow wave modulates delta, theta, spindle, alpha, beta, gamma and ripple oscillations, thus orchestrating brain electrical rhythms in sleep. While slow wave activity can enhance epileptic manifestations, it is also thought to underlie essential restorative processes and facilitate the consolidation of declarative memories. Animal studies show that slow wave activity is composed of rhythmically recurring phases of widespread, increased cortical cellular and synaptic activity, referred to as active- or up-state, followed by cellular and synaptic inactivation, referred to as silent- or down-state. However, its neural mechanisms in humans are poorly understood, since the traditional intracellular techniques used in animals are inappropriate for investigating the cellular and synaptic/transmembrane events in humans. To elucidate the intracortical neuronal mechanisms of slow wave activity in humans, novel, laminar multichannel microelectrodes were chronically implanted into the cortex of patients with drug-resistant focal epilepsy undergoing cortical mapping for seizure focus localization. Intracortical laminar local field potential gradient, multiple-unit and single-unit activities were recorded during slow wave sleep, related to simultaneous electrocorticography, and analysed with current source density and spectral methods. We found that slow wave activity in humans reflects a rhythmic oscillation between widespread cortical activation and silence. Cortical activation was demonstrated as increased wideband (0.3-200 Hz) spectral power including virtually all bands of cortical oscillations, increased multiple- and single-unit activity and powerful inward transmembrane currents, mainly localized to the supragranular layers. Neuronal firing in the up-state was sparse and the average discharge rate of single cells was less than expected from animal studies. Action potentials at up-state onset were synchronized within +/-10 ms across all cortical layers, suggesting that any layer could initiate firing at up-state onset. These findings provide strong direct experimental evidence that slow wave activity in humans is characterized by hyperpolarizing currents associated with suppressed cell firing, alternating with high levels of oscillatory synaptic/transmembrane activity associated with increased cell firing. Our results emphasize the major involvement of supragranular layers in the genesis of slow wave activity.

Novelty P3 reductions in depression: characterization using principal components analysis (PCA) of current source density (CSD) waveforms

We previously reported a novelty P3 reduction in depressed patients compared to healthy controls (n=20 per group) in a novelty oddball task using a 31-channel montage. In an independent replication and extension using a 67-channel montage (n=49 per group), reference-free current source density (CSD) waveforms were simplified and quantified by a temporal, covariance-based principal components analysis (PCA) (unrestricted Varimax rotation), yielding factor solutions consistent with other oddball tasks. A factor with a loadings peak at 343 ms summarized the target P3b source as well as a secondary midline frontocentral source for novels and targets. An earlier novelty vertex source (NVS) at 241 ms was present for novels, but not targets, and was reduced in patients. Compatible CSD-PCA findings were also confirmed for the original low-density sample. Results are consistent with a reduced novelty response in clinical depression, involving the early phase of the frontocentral novelty P3.

Current Source Density Reconstruction from Incomplete Data

We propose two ways of estimating current source density (CSD) from measurements of voltage on a Cartesian grid with missing recording points using the inverse CSD method. The simplest approach is to substitute local averages (LA) in place of missing data. A more elaborate alternative is to estimate a smaller number of CSD parameters than the actual number of recordings and to take the least-squares fit (LS). We compare the two approaches in the three-dimensional case on several sets of surrogate and experimental data, for varying numbers of missing data points, and discuss their advantages and drawbacks. One can construct CSD distributions for which one or the other approach is better. However, in general, the LA method is to be recommended as being more stable and more robust to variations in the recorded fields.

Extracting functional components of neural dynamics with Independent Component Analysis and inverse Current Source Density

Local field potentials have good temporal resolution but are blurred due to the slow spatial decay of the electric field. For simultaneous recordings on regular grids one can reconstruct efficiently the current sources (CSD) using the inverse Current Source Density method (iCSD). It is possible to decompose the resultant spatiotemporal information about the current dynamics into functional components using Independent Component Analysis (ICA). We show on test data modeling recordings of evoked potentials on a grid of 4 × 5 × 7 points that meaningful results are obtained with spatial ICA decomposition of reconstructed CSD. The components obtained through decomposition of CSD are better defined and allow easier physiological interpretation than the results of similar analysis of corresponding evoked potentials in the thalamus. We show that spatiotemporal ICA decompositions can perform better for certain types of sources but it does not seem to be the case for the experimental data studied. Having found the appropriate approach to decomposing neural dynamics into functional components we use the technique to study the somatosensory evoked potentials recorded on a grid spanning a large part of the forebrain. We discuss two example components associated with the first waves of activation of the somatosensory thalamus. We show that the proposed method brings up new, more detailed information on the time and spatial location of specific activity conveyed through various parts of the somatosensory thalamus in the rat.

Classification of cortical microcircuits based on micro-electrode-array data from slices of rat barrel cortex

The bewildering complexity of cortical microcircuits at the single cell level gives rise to surprisingly robust emergent activity patterns at the level of laminar and columnar local field potentials (LFPs) in response to targeted local stimuli. Here we report the results of our multivariate data-analytic approach based on simultaneous multi-site recordings using micro-electrode-array chips for investigation of the microcircuitry of rat somatosensory (barrel) cortex. We find high repeatability of stimulus-induced responses, and typical spatial distributions of LFP responses to stimuli in supragranular, granular, and infragranular layers, where the last form a particularly distinct class. Population spikes appear to travel with about 33 cm/s from granular to infragranular layers. Responses within barrel related columns have different profiles than those in neighbouring columns to the left or interchangeably to the right. Variations between slices occur, but can be minimized by strictly obeying controlled experimental protocols. Cluster analysis on normalized recordings indicates specific spatial distributions of time series reflecting the location of sources and sinks independent of the stimulus layer. Although the precise correspondences between single cell activity and LFPs are still far from clear, a sophisticated neuroinformatics approach in combination with multi-site LFP recordings in the standardized slice preparation is suitable for comparing normal conditions to genetically or pharmacologically altered situations based on real cortical microcircuitry.

Intracranial microprobe for evaluating neuro-hemodynamic coupling in unanesthetized human neocortex

Measurement of the blood-oxygen-level dependent (BOLD) response with fMRI has revolutionized cognitive neuroscience and is increasingly important in clinical care. The BOLD response reflects changes in deoxy-hemoglobin concentration, blood volume, and blood flow. These hemodynamic changes ultimately result from neuronal firing and synaptic activity, but the linkage between these domains is complex, poorly understood, and may differ across species, cortical areas, diseases, and cognitive states. We describe here a technique that can measure neural and hemodynamic changes simultaneously from cortical microdomains in waking humans. We utilize a "laminar optode," a linear array of microelectrodes for electrophysiological measures paired with a micro-optical device for hemodynamic measurements. Optical measurements include laser Doppler to estimate cerebral blood flow as well as point spectroscopy to estimate oxy- and deoxy-hemoglobin concentrations. The microelectrode array records local field potential gradients (PG) and multi-unit activity (MUA) at 24 locations spanning the cortical depth, permitting estimation of population trans-membrane current flows (Current Source Density, CSD) and population cell firing in each cortical lamina. Comparison of the laminar CSD/MUA profile with the origins and terminations of cortical circuits allows activity in specific neuronal circuits to be inferred and then directly compared to hemodynamics. Access is obtained in epileptic patients during diagnostic evaluation for surgical therapy. Validation tests with relatively well-understood manipulations (EKG, breath-holding, cortical electrical stimulation) demonstrate the expected responses. This device can provide a new and robust means for obtaining detailed, quantitative data for defining neurovascular coupling in awake humans.

A mouse model for studying large-scale neuronal networks using EEG mapping techniques

Human functional imaging studies are increasingly focusing on the identification of large-scale neuronal networks, their temporal properties, their development, and their plasticity and recovery after brain lesions. A method targeting large-scale networks in rodents would open the possibility to investigate their neuronal and molecular basis in detail. We here present a method to study such networks in mice with minimal invasiveness, based on the simultaneous recording of epicranial EEG from 32 electrodes regularly distributed over the head surface. Spatiotemporal analysis of the electrical potential maps similar to human EEG imaging studies allows quantifying the dynamics of the global neuronal activation with sub-millisecond resolution. We tested the feasibility, stability and reproducibility of the method by recording the electrical activity evoked by mechanical stimulation of the mystacial vibrissae. We found a series of potential maps with different spatial configurations that suggested the activation of a large-scale network with generators in several somatosensory and motor areas of both hemispheres. The spatiotemporal activation pattern was stable both across mice and in the same mouse across time. We also performed 16-channel intracortical recordings of the local field potential across cortical layers in different brain areas and found tight spatiotemporal concordance with the generators estimated from the epicranial maps. Epicranial EEG mapping thus allows assessing sensory processing by large-scale neuronal networks in living mice with minimal invasiveness, complementing existing approaches to study the neurophysiological mechanisms of interaction within the network in detail and to characterize their developmental, experience-dependent and lesion-induced plasticity in normal and transgenic animals.

Hemispatial PCA dissociates temporal from parietal ERP generator patterns: CSD components in healthy adults and depressed patients during a dichotic oddball task

Event-related potentials (31-channel ERPs) were recorded from 38 depressed, unmedicated outpatients and 26 healthy adults (all right-handed) in tonal and phonetic oddball tasks developed to exploit the perceptual challenge of a dichotic stimulation. Tonal nontargets were pairs of complex tones (corresponding to musical notes G and B above middle C) presented simultaneously to each ear (L/R) in an alternating series (G/B or B/G; 2-s fixed SOA). A target tone (note A) replaced one of the pair on 20% of the trials (A/B, G/A, B/A, A/G). Phonetic nontargets were L/R pairs of syllables (/ba/, /da/) with a short voice onset time (VOT), and targets contained a syllable (/ta/) with a long VOT. Subjects responded with a left or right button press to targets (counterbalanced across blocks). Target detection was poorer in patients than controls and for tones than syllables. Reference-free current source densities (CSDs; spherical spline Laplacian) derived from ERP waveforms were simplified and measured using temporal, covariance-based PCA followed by unrestricted Varimax rotation. Target-related N2 sinks and mid-parietal P3 sources were represented by CSD factors peaking at 245 and 440 ms. The P3 source topography included a secondary, left-lateralized temporal lobe maximum for both targets and nontargets. However, a subsequent hemispheric spatiotemporal PCA disentangled temporal lobe N1 and P3 sources as distinct factors. P3 sources were reduced in patients compared with controls, even after using performance as a covariate. Results are consistent with prior reports of P3 reduction in depression and implicate distinct parietal and temporal generators of P3 when using a dichotic oddball paradigm.

Spatio-temporal EEG source localization using a three-dimensional subspace FINE approach in a realistic geometry inhomogeneous head model

The subspace source localization approach, i.e., first principle vectors (FINE), is able to enhance the spatial resolvability and localization accuracy for closely-spaced neural sources from EEG and MEG measurements. Computer simulations were conducted to evaluate the performance of the FINE algorithm in an inhomogeneous realistic geometry head model under a variety of conditions. The source localization abilities of FINE were examined at different cortical regions and at different depths. The present computer simulation results indicate that FINE has enhanced source localization capability, as compared with MUSIC and RAP-MUSIC, when sources are closely spaced, highly noise-contaminated, or inter-correlated. The source localization accuracy of FINE is better, for closely-spaced sources, than MUSIC at various noise levels, i.e., signal-to-noise ratio (SNR) from 6 dB to 16 dB, and RAP-MUSIC at relatively low noise levels, i.e., 6 dB to 12 dB. The FINE approach has been further applied to localize brain sources of motor potentials, obtained during the finger tapping tasks in a human subject. The experimental results suggest that the detailed neural activity distribution could be revealed by FINE. The present study suggests that FINE provides enhanced performance in localizing multiple closely spaced, and inter-correlated sources under low SNR, and may become an important alternative to brain source localization from EEG or MEG.

Differentially Variable Component Analysis: Identifying Multiple Evoked Components Using Trial-to-Trial Variability

Electric potentials and magnetic fields generated by ensembles of synchronously active neurons, either spontaneously or in response to external stimuli, provide information essential to understanding the processes underlying cognitive and sensorimotor activity. Interpreting recordings of these potentials and fields is difficult because detectors record signals simultaneously generated by various regions throughout the brain. We introduce a novel approach to this problem, the differentially variable component analysis (dVCA) algorithm, which relies on trial-to-trial variability in response amplitude and latency to identify multiple components. Using simulations we demonstrate the importance of response variability to component identification, the robustness of dVCA to noise, and its ability to characterize single-trial data. We then compare the source-separation capabilities of dVCA with those of principal component analysis and independent component analysis. Finally, we apply dVCA to neural ensemble activity recorded from an awake, behaving macaque—demonstrating that dVCA is an important tool for identifying and characterizing multiple components in the single trial.

ERP/CSD indices of impaired verbal working memory subprocesses in schizophrenia

To disentangle subprocesses of verbal working memory deficits in schizophrenia, long EEG epochs (>10 s) were recorded from 13 patients and 17 healthy adults during a visual word serial position test. ERP generator patterns were summarized by temporal PCA from reference-free current source density (CSD) waveforms to sharpen 31-channel topographies. Patients showed poorer performance and reduced left inferior parietotemporal P3 source. Build-up of mid-frontal negative slow wave (SW) in controls during item encoding, integration, and active maintenance was absent in patients, whereas a sustained mid-frontal SW sink during the retention interval was comparable across groups. Mid-frontal SW sinks (encoding and retention periods) and posterior SW sinks and sources (encoding only) were related to performance in controls only. Data suggest disturbed processes in a frontal-parietotemporal network in schizophrenia, affecting encoding and early item storage.

Processing stages underlying word recognition in the anteroventral temporal lobe

The anteroventral temporal lobe integrates visual, lexical, semantic and mnestic aspects of word processing, through its reciprocal connections with the ventral visual stream, language areas, and the hippocampal formation. We used linear microelectrode arrays to probe population synaptic currents and neuronal firing in different cortical layers of the anteroventral temporal lobe, during semantic judgments with implicit priming and overt word recognition. Since different extrinsic and associative inputs preferentially target different cortical layers, this method can help reveal the sequence and nature of local processing stages at a higher resolution than was previously possible. The initial response in inferotemporal and perirhinal cortices is a brief current sink beginning at approximately 120 ms and peaking at approximately 170 ms. Localization of this initial sink to middle layers suggests that it represents feedforward input from lower visual areas, and simultaneously increased firing implies that it represents excitatory synaptic currents. Until approximately 800 ms, the main focus of transmembrane current sinks alternates between middle and superficial layers, with the superficial focus becoming increasingly dominant after approximately 550 ms. Since superficial layers are the target of local and feedback associative inputs, this suggests an alternation in predominant synaptic input between feedforward and feedback modes. Word repetition does not affect the initial perirhinal and inferotemporal middle layer sink but does decrease later activity. Entorhinal activity begins later (approximately 200 ms), with greater apparent excitatory post-synaptic currents and multiunit activity in neocortically projecting than hippocampal-projecting layers. In contrast to perirhinal and entorhinal responses, entorhinal responses are larger to repeated words during memory retrieval. These results identify a sequence of physiological activation, beginning with a sharp activation from lower level visual areas carrying specific information to middle layers. This is followed by feedback and associative interactions involving upper cortical layers, which are abbreviated to repeated words. Following bottom-up and associative stages, top-down recollective processes may be driven by entorhinal cortex. Word processing involves a systematic sequence of fast feedforward information transfer from visual areas to anteroventral temporal cortex followed by prolonged interactions of this feedforward information with local associations and feedback mnestic information from the medial temporal lobe.

Principal components analysis of Laplacian waveforms as a generic method for identifying ERP generator patterns: II. Adequacy of low-density estimates

OBJECTIVE

To evaluate the comparability of high- and low-density surface Laplacian estimates for determining ERP generator patterns of group data derived from a typical ERP sample size and paradigm.

METHODS

High-density ERP data (129 sites) recorded from 17 adults during tonal and phonetic oddball tasks were converted to a 10-20-system EEG montage (31 sites) using spherical spline interpolations. Current source density (CSD) waveforms were computed from the high- and low-density, but otherwise identical, ERPs, and correlated at corresponding locations. CSD data were submitted to separate covariance-based, unrestricted temporal PCAs (Varimax of covariance loadings) to identify and effectively summarize temporally and spatially overlapping CSD components. Solutions were compared by correlating factor loadings and scores, and by plotting ANOVA F statistics derived from corresponding high- and low-resolution factor scores using representative sites.

RESULTS

High- and low-density CSD waveforms, PCA solutions, and F statistics were remarkably similar, yielding correlations of .9 < or = r < or = .999 between waveforms, loadings, and scores for almost all comparisons at low-density locations except for low-signal CSD waveforms at occipital sites. Each of the first 10 high-density factors corresponded precisely to one factor of the first 10 low-density factors, with each 10-factor set accounting for the meaningful CSD variance (> 91.6%).

CONCLUSIONS

Low-density surface Laplacian estimates were shown to be accurate approximations of high-density CSDs at these locations, which adequately and quite sufficiently summarized group data. Moreover, reasonable approximations of many high-density scalp locations were obtained for group data from interpolations of low-density data. If group findings are the primary objective, as typical for cognitive ERP research, low-resolution CSD topographies may be as efficient, given the effective spatial smoothing when averaging across subjects and/or conditions.

SIGNIFICANCE

Conservative recommendations for restricting surface Laplacians to high-density recordings may not be appropriate for all ERP research applications, and should be re-evaluated considering objective, costs and benefits.

Principal components analysis of Laplacian waveforms as a generic method for identifying ERP generator patterns: I. Evaluation with auditory oddball tasks

OBJECTIVE

To evaluate the effectiveness and comparability of PCA-based simplifications of ERP waveforms versus their reference-free Laplacian transformations for separating task- and response-related ERP generator patterns during auditory oddball tasks.

METHODS

Nose-referenced ERPs (31 sites total) were recorded from 66 right-handed adults during oddball tasks using syllables or tones. Response mode (left press, right press, silent count) and task was varied within subjects. Spherical spline current source density (CSD) waveforms were computed to sharpen ERP scalp topographies and eliminate volume-conducted contributions. ERP and CSD data were submitted to separate covariance-based, unrestricted temporal PCAs (Varimax) to disentangle temporally and spatially overlapping ERP and CSD components.

RESULTS

Corresponding ERP and CSD factors were unambiguously related to known ERP components. For example, the dipolar organization of a central N1 was evident from factorized anterior sinks and posterior sources encompassing the Sylvian fissure. Factors associated with N2 were characterized by asymmetric frontolateral (tonal: frontotemporal R > L) and parietotemporal (phonetic: parietotemporal L > R) sinks for targets. A single ERP factor summarized parietal P3 activity, along with an anterior negativity. In contrast, two CSD factors peaking at 360 and 560 ms distinguished a parietal P3 source with an anterior sink from a centroparietal P3 source with a sharply localized Fz sink. A smaller parietal but larger left temporal P3 source was found for silent count compared to button press. Left or right press produced opposite, region-specific asymmetries originating from central sites, modulating the N2/P3 complex.

CONCLUSIONS

CSD transformation is shown to be a valuable preprocessing step for PCA of ERP data, providing a unique, physiologically meaningful solution to the ubiquitous reference problem. By reducing ERP redundancy and producing sharper, simpler topographies, and without losing or distorting any effects of interest, the CSD-PCA solution replicated and extended previous task- and response-related findings.

SIGNIFICANCE

Eliminating ambiguities of the recording reference, the combined CSD-PCA approach systematically bridges between montage-dependent scalp potentials and distinct, anatomically-relevant current generators, and shows promise as a comprehensive, generic strategy for ERP analysis.

Optimizing PCA methodology for ERP component identification and measurement: theoretical rationale and empirical evaluation

OBJECTIVE

To determine how specific methodological choices affect "data-driven" simplifications of event-related potentials (ERPs) using principal components analysis (PCA). The usefulness of the extracted component measures can be evaluated by knowledge about the variance distribution of ERPs, which are characterized by the removal of baseline activity. The variance should be small before and at stimulus onset (across and within cases), but large near the end of the recording epoch and at ERP component peaks. These characteristics are preserved with a covariance matrix, but lost with a correlation matrix, which assigns equal weights to each sample point, yielding the possibility that small but systematic variations may form a factor.

METHODS

Varimax-rotated PCAs were performed on simulated and real ERPs, systematically varying extraction criteria (number of factors) and method (correlation/covariance matrix, using unstandardized/standardized loadings before rotation).

RESULTS

Conservative extraction criteria changed the morphology of some components considerably, which had severe implications for inferential statistics. Solutions converged and stabilized with more liberal criteria. Interpretability (more distinctive component waveforms with narrow and unambiguous loading peaks) and statistical conclusions (greater effect stability across extraction criteria) were best for unstandardized covariance-based solutions. In contrast, all standardized covariance- and correlation-based solutions included "high-variance" factors during the baseline, confirming findings for simulated data.

CONCLUSIONS

Unrestricted, unstandardized covariance-based PCA solutions optimize ERP component identification and measurement.