Simulation of background neuronal activity and noise in human intracranial microwire recordings.

BACKGROUND: Human intracranial microwire recordings allow measurement of neuronal activity in human subjects at a fine temporal and spatial scale. The recorded extracellular potentials represent a mixture of action potentials from nearby neurons, local field potentials, and other noise sources. Signal processing of these recordings is used to separate the activity of putative single neurons from other background and noise. To better understand the separation of single neuron activity, one approach is to simulate the signals produced by neurons firing action potentials combined with background activity and noise. NEW METHOD: This paper characterizes the background activity and noise in human intracranial microwire recordings and presents an accurate and efficient method of simulation using an infinite impulse response filter to color white noise. RESULTS AND COMPARISON: This method reproduces the power spectral density of the background activity and noise over a frequency range of 1-5000 Hz and is over 200 times faster than previously used methods. It thus facilitates large scale studies of variation of noise sources, field potentials, and processing parameters. It performs equivalently in terms of spike sorting to simulation using white noise. Another advantage is that the simulated signals are known to arise from a pseudorandom number generator and cannot be the result of detecting simulated background spiking activity. CONCLUSIONS: This approach provides a rapid and accurate method of simulating background noise and neural activity in human intracranial microwire recordings. It is suitable for use in large scale simulations to study spike sorting in this type of recording.

Two kinds of memory signals in neurons of the human hippocampus.

Prior studies of the neural representation of episodic memory in the human hippocampus have identified generic memory signals representing the categorical status of test items (novel vs. repeated), whereas other studies have identified item specific memory signals representing individual test items. Here, we report that both kinds of memory signals can be detected in hippocampal neurons in the same experiment. We recorded single-unit activity from four brain regions (hippocampus, amygdala, anterior cingulate, and prefrontal cortex) of epilepsy patients as they completed a continuous recognition task. The generic signal was found in all four brain regions, whereas the item-specific memory signal was detected only in the hippocampus and reflected sparse coding. That is, for the item-specific signal, each hippocampal neuron responded strongly to a small fraction of repeated words, and each repeated word elicited strong responding in a small fraction of neurons. The neural code was sparse, pattern-separated, and limited to the hippocampus, consistent with longstanding computational models. We suggest that the item-specific episodic memory signal in the hippocampus is fundamental, whereas the more widespread generic memory signal is derivative and is likely used by different areas of the brain to perform memory-related functions that do not require item-specific information.

Spiking activity in the human hippocampus prior to encoding predicts subsequent memory.

Encoding activity in the medial temporal lobe, presumably evoked by the presentation of stimuli (postonset activity), is known to predict subsequent memory. However, several independent lines of research suggest that preonset activity also affects subsequent memory. We investigated the role of preonset and postonset single-unit and multiunit activity recorded from epilepsy patients as they completed a continuous recognition task. In this task, words were presented in a continuous series and eventually began to repeat. For each word, the patient's task was to decide whether it was novel or repeated. We found that preonset spiking activity in the hippocampus (when the word was novel) predicted subsequent memory (when the word was later repeated). Postonset activity during encoding also predicted subsequent memory, but was simply a continuation of preonset activity. The predictive effect of preonset spiking activity was much stronger in the hippocampus than in three other brain regions (amygdala, anterior cingulate, and prefrontal cortex). In addition, preonset and postonset activity around the encoding of novel words did not predict memory performance for novel words (i.e., correctly classifying the word as novel), and preonset and postonset activity around the time of retrieval did not predict memory performance for repeated words (i.e., correctly classifying the word as repeated). Thus, the only predictive effect was between preonset activity (along with its postonset continuation) at the time of encoding and subsequent memory. Taken together, these findings indicate that preonset hippocampal activity does not reflect general arousal/attention but instead reflects what we term "attention to encoding."

Estimates of distributed coding of visual objects by single neurons in the human brain depend on which spike sorting technique is used.

OBJECTIVE: To determine whether the estimated fraction and degree of distribution of visual responsive neurons in human intracranial microwire recordings depends upon the spike sorting method used. APPROACH: A large dataset of human intracranial microwire recordings from four brain areas was sorted into single unit activity (SUA) and multiunit activity (MUA) using 4 spike sorting methods previously applied to this type of recording. The responses were examined for visual responses to 33 objects which were presented. MAIN RESULTS: The 4 spike sorting techniques examined here yielded fractions of responsive SUA varying from 8% in the left anterior cingulate cortex to 27% in the right amgdala. The fraction of responsive SUA and MUA depended on the type of spike sorting being used as well as brain area and side being recorded from. Agreement between spike sorting techniques was low (0.04-0.16 on the 0-1 AMIall scale). SIGNIFICANCE: Prior estimates of the fraction of single neurons in the human medial temporal lobe coding semantic memory of visual objects have yielded fractions ranging from 0.04% by very strict response criteria to 47% by other criteria. A variety of explanations of these differences have been posited, including differences in the type of memory being tested, differences in visual stimuli, as well as technical differences such as spike sorting techniques. This study shows the dependence of the reported fraction of neurons encoding visual objects on the spike sorting technique employed and confirms a distributed representation of visual objects by single neurons in the human brain. Both the variation in the responsive fractions with spike sorting technique and low levels of agreement between techniques highlight the need for better understanding of the signals being extracted in human intracranial microwire recordings.

Coding of episodic memory in the human hippocampus.

Neurocomputational models have long posited that episodic memories in the human hippocampus are represented by sparse, stimulus-specific neural codes. A concomitant proposal is that when sparse-distributed neural assemblies become active, they suppress the activity of competing neurons (neural sharpening). We investigated episodic memory coding in the hippocampus and amygdala by measuring single-neuron responses from 20 epilepsy patients (12 female) undergoing intracranial monitoring while they completed a continuous recognition memory task. In the left hippocampus, the distribution of single-neuron activity indicated that only a small fraction of neurons exhibited strong responding to a given repeated word and that each repeated word elicited strong responding in a different small fraction of neurons. This finding reflects sparse distributed coding. The remaining large fraction of neurons exhibited a concurrent reduction in firing rates relative to novel words. The observed pattern accords with longstanding predictions that have previously received scant support from single-cell recordings from human hippocampus.

Comparison of combined spike detection and clustering using mutual information.

BACKGROUND: Spike sorting techniques involve both detection of spike waveform events and classification of those events into clusters of similar waveform shape. The one existing method of evaluating the combined effects of both detection and classification depends on assignment of cluster correspondence. Other methods of evaluation have focused on either clustering or detection, but not both, although these two steps may interact. NEW METHOD: This paper develops an information theoretic measure of agreement between the output of two spike sorting techniques, AMIall, which can be used even when the number of waveform events detected by the two techniques differs. RESULTS: AMIall is shown to be a useful measure for studying variations of parameters of spike sorting techniques in two examples: comparing outputs for simulated noisy spike sorting and spike sorting of human single neuron recordings. Comparison with existing methods Computing AMIall does not require an explicit assignment of cluster correspondence, thereby eliminating a potential source of variation. By providing a single measure of performance, computing AMIall is very useful when comparing large numbers of algorithmic or parametric variations of spike sorting techniques; prior comparison techniques have often required multiple measures of performance which complicates large scale comparisons. CONCLUSIONS: The use of AMIall to measure agreement between spike sorting techniques facilitates the comparison of the outputs of those techniques, including variations in both spike detection and waveform clustering. This measure should be useful for broad based and large scale comparisons between spike sorting techniques.

Single neuron recordings of bilinguals performing in a continuous recognition memory task.

We report the results of a bilingual continuous recognition memory task during which single- and multi-neuron activity was recorded in human subjects with intracranial microwire implants. Subjects (n = 5) were right-handed Spanish-English bilinguals who were undergoing evaluation prior to surgery for severe epilepsy. Subjects were presented with Spanish and English words and the task was to determine whether any given word had been seen earlier in the testing session, irrespective of the language in which it had appeared. Recordings in the left and right hippocampus revealed notable laterality, whereby both Spanish and English items that had been seen previously in the other language (switch trials) triggered increased neural firing in the left hippocampus. Items that had been seen previously in the same language (repeat trials) triggered increased neural firings in the right hippocampus. These results are consistent with theories that propose roles of both the left- and right-hemisphere in real-time linguistic processing. Importantly, this experiment presents the first instance of intracranial recordings in bilinguals performing a task with switching demands.

Gender differences in human single neuron responses to male emotional faces.

Well-documented differences in the psychology and behavior of men and women have spurred extensive exploration of gender's role within the brain, particularly regarding emotional processing. While neuroanatomical studies clearly show differences between the sexes, the functional effects of these differences are less understood. Neuroimaging studies have shown inconsistent locations and magnitudes of gender differences in brain hemodynamic responses to emotion. To better understand the neurophysiology of these gender differences, we analyzed recordings of single neuron activity in the human brain as subjects of both genders viewed emotional expressions. This study included recordings of single-neuron activity of 14 (6 male) epileptic patients in four brain areas: amygdala (236 neurons), hippocampus (n = 270), anterior cingulate cortex (n = 256), and ventromedial prefrontal cortex (n = 174). Neural activity was recorded while participants viewed a series of avatar male faces portraying positive, negative or neutral expressions. Significant gender differences were found in the left amygdala, where 23% (n = 15∕66) of neurons in men were significantly affected by facial emotion, vs. 8% (n = 6∕76) of neurons in women. A Fisher's exact test comparing the two ratios found a highly significant difference between the two (p < 0.01). These results show specific differences between genders at the single-neuron level in the human amygdala. These differences may reflect gender-based distinctions in evolved capacities for emotional processing and also demonstrate the importance of including subject gender as an independent factor in future studies of emotional processing by single neurons in the human amygdala.

Distributed representation of visual objects by single neurons in the human brain.

It remains unclear how single neurons in the human brain represent whole-object visual stimuli. While recordings in both human and nonhuman primates have shown distributed representations of objects (many neurons encoding multiple objects), recordings of single neurons in the human medial temporal lobe, taken as subjects' discriminated objects during multiple presentations, have shown gnostic representations (single neurons encoding one object). Because some studies suggest that repeated viewing may enhance neural selectivity for objects, we had human subjects discriminate objects in a single, more naturalistic viewing session. We found that, across 432 well isolated neurons recorded in the hippocampus and amygdala, the average fraction of objects encoded was 26%. We also found that more neurons encoded several objects versus only one object in the hippocampus (28 vs 18%, p < 0.001) and in the amygdala (30 vs 19%, p < 0.001). Thus, during realistic viewing experiences, typical neurons in the human medial temporal lobe code for a considerable range of objects, across multiple semantic categories.

Sparse and distributed coding of episodic memory in neurons of the human hippocampus.

Neurocomputational models hold that sparse distributed coding is the most efficient way for hippocampal neurons to encode episodic memories rapidly. We investigated the representation of episodic memory in hippocampal neurons of nine epilepsy patients undergoing intracranial monitoring as they discriminated between recently studied words (targets) and new words (foils) on a recognition test. On average, single units and multiunits exhibited higher spike counts in response to targets relative to foils, and the size of this effect correlated with behavioral performance. Further analyses of the spike-count distributions revealed that (i) a small percentage of recorded neurons responded to any one target and (ii) a small percentage of targets elicited a strong response in any one neuron. These findings are consistent with the idea that in the human hippocampus episodic memory is supported by a sparse distributed neural code.

Testing for effects of different stimuli on neuronal firing relative to background activity.

OBJECTIVE: Statistical testing for effects of stimuli on the responses of large populations of recorded neurons is a key technique for analyzing data generated using multi-channel recording systems. Combining statistical tests for differences of the responses to different stimuli and tests for changes from background firing, while appealing as apparently focusing analysis on neurons which react to the stimuli, can lead to significant overestimates of the magnitudes of the effects of stimulation or even erroneous identification of responses as being statistically significant. APPROACH: As this type of combination is common in the neurophysiological literature, we derive a non-parametric hypothesis test for changes of responses to different types of stimuli relative to background firing. MAIN RESULTS: This test can be used to avoid improperly combining two statistical tests. We also model the overestimation of effects of stimulus type that can be observed in typical single-unit recordings when applying two statistical tests sequentially. SIGNIFICANCE: Over-estimation of the fraction of neurons with a response can range up to 1800% in plausible recording scenarios. The method of testing derived here avoids this type of error.

A statistical method for predicting seizure onset zones from human single-neuron recordings.

OBJECTIVE: Clinicians often use depth-electrode recordings to localize human epileptogenic foci. To advance the diagnostic value of these recordings, we applied logistic regression models to single-neuron recordings from depth-electrode microwires to predict seizure onset zones (SOZs). APPROACH: We collected data from 17 epilepsy patients at the Barrow Neurological Institute and developed logistic regression models to calculate the odds of observing SOZs in the hippocampus, amygdala and ventromedial prefrontal cortex, based on statistics such as the burst interspike interval (ISI). MAIN RESULTS: Analysis of these models showed that, for a single-unit increase in burst ISI ratio, the left hippocampus was approximately 12 times more likely to contain a SOZ; and the right amygdala, 14.5 times more likely. Our models were most accurate for the hippocampus bilaterally (at 85% average sensitivity), and performance was comparable with current diagnostics such as electroencephalography. SIGNIFICANCE: Logistic regression models can be combined with single-neuron recording to predict likely SOZs in epilepsy patients being evaluated for resective surgery, providing an automated source of clinically useful information.

Firing behavior and network activity of single neurons in human epileptic hypothalamic hamartoma.

OBJECTIVE: Human hypothalamic hamartomas (HH) are intrinsically epileptogenic and are associated with treatment-resistant gelastic seizures. The basic cellular mechanisms responsible for seizure onset within HH are unknown. We used intra-operative microwire recordings of single neuron activity to measure the spontaneous firing rate of neurons and the degree of functional connection between neurons within the tumor. TECHNIQUE: Fourteen patients underwent transventricular endoscopic resection of HH for treatment-resistant epilepsy. Prior to surgical resection, single neuron recordings from bundled microwires (total of nine contacts) were obtained from HH tissue. Spontaneous activity was recorded for two or three 5-min epochs under steady-state general anesthesia. Off-line analysis included cluster analysis of single unit activity and probability analysis of firing relationships between pairs of neurons. RESULTS: Altogether, 222 neurons were identified (mean 6 neurons per recording epoch). Cluster analysis of single neuron firing utilizing a mixture of Gaussians model identified two distinct populations on the basis of firing rate (median firing frequency 0.6 versus 15.0 spikes per second; p < 10(-5)). Cluster analysis identified three populations determined by levels of burst firing (median burst indices of 0.015, 0.18, and 0.39; p < 10(-15)). Unbiased analysis of spontaneous single unit behavior showed that 51% of all possible neuron pairs within each recording epoch had a significant level of firing synchrony (p < 10(-15)). The subgroup of neurons with higher median firing frequencies was more likely to demonstrate synchronous firing (p < 10(-7)). CONCLUSION: Hypothalamic hamartoma tissue in vivo contains neurons which fire spontaneously. The activity of single neurons is diverse but distributes into at least two electrophysiological phenoytpes. Functional linkage between single neurons suggests that HH neurons exist within local networks that may contribute to ictogenesis.

Neurons in the human hippocampus and amygdala respond to both low- and high-level image properties.

A large number of studies have demonstrated that structures within the medial temporal lobe, such as the hippocampus, are intimately involved in declarative memory for objects and people. Although these items are abstractions of the visual scene, specific visual details can change the speed and accuracy of their recall. By recording from 415 neurons in the hippocampus and amygdala of human epilepsy patients as they viewed images drawn from 10 image categories, we showed that the firing rates of 8% of these neurons encode image illuminance and contrast, low-level properties not directly pertinent to task performance, whereas in 7% of the neurons, firing rates encode the category of the item depicted in the image, a high-level property pertinent to the task. This simultaneous representation of high- and low-level image properties within the same brain areas may serve to bind separate aspects of visual objects into a coherent percept and allow episodic details of objects to influence mnemonic performance.

In situ single-unit recording of hypothalamic hamartomas under endoscopic direct visualization.

OBJECTIVE: Hypothalamic hamartomas (HHs) are associated with refractory epilepsy and are amenable to surgical treatment. The gelastic seizures associated with HHs originate within the HH lesion, but the responsible cellular mechanisms are unknown. Microelectrode patch-clamp recordings from HH neurons in resected slice preparations show that small HH neurons spontaneously fire with intrinsic pacemaker-like activity. We questioned whether spontaneous firing of HH neurons was present in situ, and we hypothesized that single-unit field recordings from HH tissue could be obtained with instrumentation passed through the endoscope before surgical resection. TECHNIQUE: After informed consent was obtained, patients undergoing transventricular, endoscopic resection of an HH for intractable epilepsy were eligible for study. After placement of the endoscope, a bundled microwire (total of 9 contacts) was placed into the HH under direct visualization. Spontaneous activity was recorded for two or three 5-minute epochs, under steady-state general anesthesia. The wire was advanced 0.5 to 1 mm within the lesion between recording epochs. RESULTS: A total of thirteen 5-minute recordings were obtained from 5 patients. Noise levels were comparable to extraoperative microwire recordings for temporal lobe epilepsy. Single-neuron spike activity was isolated from a total of 5 channels obtained during recording of 3 sessions in 3 patients. CONCLUSION: We have shown that single-unit recordings from HH lesions can be successfully obtained in situ under direct endoscopic visualization. We believe that this is the first report using the working channel of a neuroendoscope to make physiological recordings of deep structures in humans.

Interference and noise in human intracranial microwire recordings.

Human intracranial microwire recordings have typically had poor signal-to-noise ratios (SNRs), often below 10 dB. The physiological signal source is a fixed-amplitude one; thus, SNR must be improved by reducing either noise or interference. An understanding of the interference sources, how they are coupled to the recording system, and their relative magnitudes is needed to improve SNR. We measured potentially interfering sources in a controlled laboratory model of microwire recordings. Specifically considered were interference from power lines, fluorescent lights, radio transmitters, and other nearby electrical devices. In the presence of typical mismatches in impedance (100 kohm) and loop area (30 cm2), the greatest sources of interference are capacitive coupling to power lines (11.4 microV(rms)), capacitive coupling to fluorescent lights (9.7 microV(rms)), and nonpower line capacitive interference (8.6 microV(rms)). The model and techniques employed here to study human microwire recordings may also be applied to other neurophysiological recordings.

Alternate task inhibits single-neuron category-selective responses in the human hippocampus while preserving selectivity in the amygdala.

One fifth of neurons in the medial-temporal lobe of human epilepsy patients respond selectively to categories of images, such as faces or cars. Here we show that responses of hippocampal neurons are rapidly modified as subjects alternate (over 60 sec) between two tasks (1) identifying images from a category, or (2) playing a simple video game superimposed on the same images. Category-selective responses, present when a subject identifies categories, are eliminated when the subject shifts to playing the game for 87% of category-selective hippocampal neurons. By contrast, responses in the amygdala are present during both tasks for 72% of category-selective amygdalar neurons. These results suggest that attention to images is required to evoke selective responses from single neurons in the hippocampus, but is not required by neurons in the amygdala.

Assessing the direct effects of deep brain stimulation using embedded axon models.

To better understand the spatial extent of the direct effects of deep brain stimulation (DBS) on neurons, we implemented a geometrically realistic finite element electrical model incorporating anisotropic and inhomogenous conductivities. The model included the subthalamic nucleus (STN), substantia nigra (SN), zona incerta (ZI), fields of Forel H2 (FF), internal capsule (IC) and Medtronic 3387/3389 electrode. To quantify the effects of stimulation, we extended previous studies by using multi-compartment axon models with geometry and orientation consistent with anatomical features of the brain regions of interest. Simulation of axonal firing produced a map of relative changes in axonal activation. Voltage-controlled stimulation, with clinically typical parameters at the dorso-lateral STN, caused axon activation up to 4 mm from the target. This activation occurred within the FF, IC, SN and ZI with current intensities close to the average injected during DBS (3 mA). A sensitivity analysis of model parameters (fiber size, fiber orientation, degree of inhomogeneity, degree of anisotropy, electrode configuration) revealed that the FF and IC were consistently activated. Direct activation of axons outside the STN suggests that other brain regions may be involved in the beneficial effects of DBS when treating Parkinsonian symptoms.

Differences in mnemonic processing by neurons in the human hippocampus and parahippocampal regions.

Different structures within the medial-temporal lobe likely make distinct contributions to declarative memory. In particular, several current psychological and computational models of memory predict that the hippocampus and parahippocampal regions play different roles in the formation and retrieval of declarative memories [e.g., Norman, K. A., & O'Reilly, R. C. Modeling hippocampal and neocortical contributions to recognition memory: A complementary-learning systems approach. Psychological Review, 110, 611-646, 2003]. Here, we examined the neuronal firing patterns in these two regions during recognition memory. Recording directly from neurons in humans, we find that cells in both regions respond to novel stimuli with an increase in firing (excitation). However, already on the second presentation of a stimulus, neurons in these regions show very different firing patterns. In the parahippocampal region there is dramatic decrease in the number of cells responding to the stimuli, whereas in the hippocampus there is recruitment of a large subset of neurons showing inhibitory (decrease from baseline firing) responses. These results suggest that inhibition is a mechanism used by cells in the human hippocampus to support sparse coding in mnemonic processing. The findings also provide further evidence for the division of labor in the medial-temporal lobe with respect to declarative memory processes.

Robust micromechanical neurite elicitation in synapse-competent neurons via magnetic bead force application.

The ability to engineer living networks of interconnected neurons with specified connectivity would facilitate the study of synaptogenesis and information processing in the nervous system. Previously, we found that a neurite can be elicited from embryonic chick forebrain neurons by direct mechanical means using magnetic bead force application (MBFA); however, our previous studies and others focused on young, synapse-incompetent neurons. To address this issue, we tested cultures of embryonic chick forebrain neurons of varying age and found that neurites could be micromechanically elicited via MBFA at all ages tested, which ranged between 7 and 22 embryonic equivalent (EE) days (days in ovo plus days in vitro). The probability of neurite initiation was at least 40% for all ages, with a maximum of approximately 80% after 2-4 days in vitro, and a decrease to approximately 60% by day 10 in vitro. The force required to elicit a neurite was approximately 1500 pN with a minimum of approximately 700 pN at embryonic equivalent day 14. The probability of success was similar for two rates of force application (10 and 500 pN/s). Neurite initiation via micromechanical force is robust with respect to cell age, and micromechanical force can induce neurites in synapse-competent neurons.