Barak A. Pearlmutter
Publications
-
2.18Impact points
A New Hypothesis for Sleep: Tuning for Criticality.
Neural computation. 03/2009;
We propose that the critical function of sleep is to prevent uncontrolled neuronal feedback while allowing rapid responses and prolonged retention of short-term memories. Through learning, the brain is tuned to react optimally to environmental challenges. Optimal behavior often requires rapid respon... [more] We propose that the critical function of sleep is to prevent uncontrolled neuronal feedback while allowing rapid responses and prolonged retention of short-term memories. Through learning, the brain is tuned to react optimally to environmental challenges. Optimal behavior often requires rapid responses and the prolonged retention of short-term memories. At a neuronal level, these correspond to recurrent activity in local networks. Unfortunately, when a network exhibits recurrent activity, small changes in the parameters or conditions can lead to runaway oscillations. Thus, the very changes that improve the processing performance of the network can put it at risk of runaway oscillation. To prevent this, stimulus-dependent network changes should be permitted only when there is a margin of safety around the current network parameters. We propose that the essential role of sleep is to establish this margin by exposing the network to a variety of inputs, monitoring for erratic behavior, and adjusting the parameters. When sleep is not possible, an emergency mechanism must come into play, preventing runaway behavior at the expense of processing efficiency. This is tiredness.
-
4.46Impact points
Dissecting the cellular contributions to early visual sensory processing deficits in schizophrenia using the VESPA evoked response.
Schizophrenia research. 02/2008; 98(1-3):256-64.
Electrophysiological research has shown clear dysfunction of early visual processing mechanisms in patients with schizophrenia. In particular, the P1 component of the visual evoked potential (VEP) is substantially reduced in amplitude in patients. A novel visual evoked response known as the VESPA (V... [more] Electrophysiological research has shown clear dysfunction of early visual processing mechanisms in patients with schizophrenia. In particular, the P1 component of the visual evoked potential (VEP) is substantially reduced in amplitude in patients. A novel visual evoked response known as the VESPA (Visual Evoked Spread Spectrum Analysis) was recently described. This response has a notably different scalp topography from that of the traditional VEP, suggesting preferential activation of a distinct subpopulation of cells. As such, this method constitutes a potentially useful candidate for investigating cellular contributions to early visual processing deficits. In this paper we compare the VEP and VESPA responses between a group of healthy control subjects and a group of schizophrenia patients. We also introduce an extension of the VESPA method to incorporate nonlinear processing in the visual system. A significantly reduced P1 component was found in patients using the VEP (with a large effect size; Cohen's d=1.6), while there was no difference whatsoever in amplitude between groups for either the linear or nonlinear VESPA. This pattern of results points to a highly specific cellular substrate of early visual processing deficits in schizophrenia, suggesting that these deficits are based on dysfunction of magnocellular pathways with parvocellular processing remaining largely intact.
-
Software platform for rapid prototyping of NIRS brain computer interfacing techniques.
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference. 02/2008; 1:4840-4843.
This paper describes the control system of a next-generation optical brain-computer interface (BCI). Using functional near-infrared spectroscopy (fNIRS) as a BCI modality is a relatively new concept, and research has only begun to explore approaches for its implementation. It is necessary to have a ... [more] This paper describes the control system of a next-generation optical brain-computer interface (BCI). Using functional near-infrared spectroscopy (fNIRS) as a BCI modality is a relatively new concept, and research has only begun to explore approaches for its implementation. It is necessary to have a system by which it is possible to investigate the signal processing and classification techniques available in the BCI community. Most importantly, these techniques must be easily testable in real-time applications. The system we describe was built using LABVIEW, a graphical programming language designed for interaction with National Instruments hardware. This platform allows complete configurability from hardware control and regulation, testing and filtering in a graphical interface environment.
-
A 12-Channel, real-time near-infrared spectroscopy instrument for brain-computer interface applications.
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference. 02/2008; 1:5648-5651.
A continuous wave near-infrared spectroscopy (NIRS) instrument for brain-computer interface (BCI) applications is presented. In the literature, experiments have been carried out on subjects with such motor degenerative diseases as amyotrophic lateral sclerosis, which have demonstrated the suitabilit... [more] A continuous wave near-infrared spectroscopy (NIRS) instrument for brain-computer interface (BCI) applications is presented. In the literature, experiments have been carried out on subjects with such motor degenerative diseases as amyotrophic lateral sclerosis, which have demonstrated the suitability of NIRS to access intentional functional activity, which could be used in a BCI as a communication aid. Specifically, a real-time, multiple channel NIRS tool is needed to realise access to even a few different mental states, for reasonable baud rates. The 12-channel instrument described here has a spatial resolution of 30mm, employing a flexible software demodulation scheme. Temporal resolution of approximately 100ms is maintained since typical topographic imaging is not needed, since we are only interested in exploiting the vascular response for BCI control. A simple experiment demonstrates the ability of the system to report on haemodynamics during single trial mental arithmetic tasks. Multiple trial averaging is not required.
-
3.42Impact points
Isolating endogenous visuo-spatial attentional effects using the novel visual-evoked spread spectrum analysis (VESPA) technique.
The European journal of neuroscience. 01/2008; 26(12):3536-42.
In natural visual environments, we use attention to select between relevant and irrelevant stimuli that are presented simultaneously. Our attention to objects in our visual field is largely controlled endogenously, but is also affected exogenously through the influence of novel stimuli and events. T... [more] In natural visual environments, we use attention to select between relevant and irrelevant stimuli that are presented simultaneously. Our attention to objects in our visual field is largely controlled endogenously, but is also affected exogenously through the influence of novel stimuli and events. The study of endogenous and exogenous attention as separate mechanisms has been possible in behavioral and functional imaging studies, where multiple stimuli can be presented continuously and simultaneously. It has also been possible in electroencephalogram studies using the steady-state visual-evoked potential (SSVEP); however, it has not been possible in conventional event-related potential (ERP) studies, which are hampered by the need to present suddenly onsetting stimuli in isolation. This is unfortunate as the ERP technique allows for the analysis of human physiology with much greater temporal resolution than functional magnetic resonance imaging or the SSVEP. While ERP studies of endogenous attention have been widely reported, these experiments have a serious limitation in that the suddenly onsetting stimuli, used to elicit the ERP, inevitably have an exogenous, attention-grabbing effect. Recently we have shown that it is possible to derive separate event-related responses to concurrent, continuously presented stimuli using the VESPA (visual-evoked spread spectrum analysis) technique. In this study we employed an experimental paradigm based on this method, in which two pairs of diagonally opposite, non-contiguous disc-segment stimuli were presented, one pair to be ignored and the other to be attended. VESPA responses derived for each pair showed a strong modulation at 90-100 ms (during the visual P1 component), demonstrating the utility of the method for isolating endogenous visuo-spatial attention effects.
-
Discovering speech phones using convolutive non-negative matrix factorisation with a sparseness constraint.
Neurocomputing. 01/2008; 72:88-101.
-
2.18Impact points
Optimal coding predicts attentional modulation of activity in neural systems.
Neural computation. 06/2007; 19(5):1295-312.
Neuronal activity in response to a fixed stimulus has been shown to change as a function of attentional state, implying that the neural code also changes with attention. We propose an information-theoretic account of such modulation: that the nervous system adapts to optimally encode sensory stimuli... [more] Neuronal activity in response to a fixed stimulus has been shown to change as a function of attentional state, implying that the neural code also changes with attention. We propose an information-theoretic account of such modulation: that the nervous system adapts to optimally encode sensory stimuli while taking into account the changing relevance of different features. We show using computer simulation that such modulation emerges in a coding system informed about the uneven relevance of the input features. We present a simple feedforward model that learns a covert attention mechanism, given input patterns and coding fidelity requirements. After optimization, the system gains the ability to reorganize its computational resources (and coding strategy) depending on the incoming attentional signal, without the need of multiplicative interaction or explicit gating mechanisms between units. The modulation of activity for different attentional states matches that observed in a variety of selective attention experiments. This model predicts that the shape of the attentional modulation function can be strongly stimulus dependent. The general principle presented here accounts for attentional modulation of neural activity without relying on special-purpose architectural mechanisms dedicated to attention. This principle applies to different attentional goals, and its implications are relevant for all modalities in which attentional phenomena are observed.
-
1.52Impact points
Illusory percepts from auditory adaptation.
The Journal of the Acoustical Society of America. 04/2007; 121(3):1632-41.
Phenomena resembling tinnitus and Zwicker phantom tone are seen to result from an auditory gain adaptation mechanism that attempts to make full use of a fixed-capacity channel. In the case of tinnitus, the gain adaptation enhances internal noise of a frequency band otherwise silent due to damage. Th... [more] Phenomena resembling tinnitus and Zwicker phantom tone are seen to result from an auditory gain adaptation mechanism that attempts to make full use of a fixed-capacity channel. In the case of tinnitus, the gain adaptation enhances internal noise of a frequency band otherwise silent due to damage. This generates a percept of a phantom sound as a consequence of hearing loss. In the case of Zwicker tone, a frequency band is temporarily silent during the presentation of a notched broadband sound, resulting in a percept of a tone at the notched frequency. The model suggests a link between tinnitus and the Zwicker tone percept, in that it predicts different results for normal and tinnitus subjects due to a loss of instantaneous nonlinear compression. Listening experiments on 44 subjects show that tinnitus subjects (11 of 44) are significantly more likely to hear the Zwicker tone. This psychoacoustic experiment establishes the first empirical link between the Zwicker tone percept and tinnitus. Together with the modeling results, this supports the hypothesis that the phantom percept is a consequence of a central adaptation mechanism confronted with a degraded sensory apparatus.
-
Discovering Convolutive Speech Phones Using Sparseness and Non-negativity.
Independent Component Analysis and Signal Separation, 7th International Conference, ICA 2007, London, UK, September 9-12, 2007.; 01/2007
-
5.74Impact points
The VESPA: a method for the rapid estimation of a visual evoked potential.
NeuroImage. 10/2006; 32(4):1549-61.
Faster and less obtrusive means for measuring a Visual Evoked Potential would be valuable in clinical testing and basic neuroscience research. This study presents a method for accomplishing this by smoothly modulating the luminance of a visual stimulus using a stochastic process. Despite its visuall... [more] Faster and less obtrusive means for measuring a Visual Evoked Potential would be valuable in clinical testing and basic neuroscience research. This study presents a method for accomplishing this by smoothly modulating the luminance of a visual stimulus using a stochastic process. Despite its visually unobtrusive nature, the rich statistical structure of the stimulus enables rapid estimation of the visual system's impulse response. The profile of these responses, which we call VESPAs, correlates with standard VEPs, with r=0.91, p<10(-28) for the group average. The time taken to obtain a VESPA with a given signal-to-noise ratio compares favorably to that required to obtain a VEP with a similar level of certainty. Additionally, we show that VESPA responses to two independent stimuli can be obtained simultaneously, which could drastically reduce the time required to collect responses to multiple stimuli. The new method appears to provide a useful alternative to standard VEP methods, and to have potential application both in clinical practice and to the study of sensory and perceptual functions.
-
7.18Impact points
Sparse representations for the cocktail party problem.
The Journal of neuroscience : the official journal of the Society for Neuroscience. 08/2006; 26(28):7477-90.
A striking feature of many sensory processing problems is that there appear to be many more neurons engaged in the internal representations of the signal than in its transduction. For example, humans have approximately 30,000 cochlear neurons, but at least 1000 times as many neurons in the auditory ... [more] A striking feature of many sensory processing problems is that there appear to be many more neurons engaged in the internal representations of the signal than in its transduction. For example, humans have approximately 30,000 cochlear neurons, but at least 1000 times as many neurons in the auditory cortex. Such apparently redundant internal representations have sometimes been proposed as necessary to overcome neuronal noise. We instead posit that they directly subserve computations of interest. Here we provide an example of how sparse overcomplete linear representations can directly solve difficult acoustic signal processing problems, using as an example monaural source separation using solely the cues provided by the differential filtering imposed on a source by its path from its origin to the cochlea [the head-related transfer function (HRTF)]. In contrast to much previous work, the HRTF is used here to separate auditory streams rather than to localize them in space. The experimentally testable predictions that arise from this model, including a novel method for estimating the optimal stimulus of a neuron using data from a multineuron recording experiment, are generic and apply to a wide range of sensory computations.
-
A spectrum of colors: investigating the temporal frequency characteristics of the human visual system using a system identification approach.
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference. 02/2006; 1:3720-3.
Noise input signals are commonly used in both linear and nonlinear system identification of physiological systems. This method can be applied to electrophysiological analysis of the human visual system by controlling the modulation of the contrast of a checkerboard stimulus using a pre-computed nois... [more] Noise input signals are commonly used in both linear and nonlinear system identification of physiological systems. This method can be applied to electrophysiological analysis of the human visual system by controlling the modulation of the contrast of a checkerboard stimulus using a pre-computed noise waveform. In this study we describe how one can obtain an estimate of the linear impulse response of the visual system using noise waveforms. Furthermore, we examine the impulse responses obtained using noise signals with different frequency characteristics, in an attempt to investigate the temporal frequency characteristics of the human visual system. We show that noise signals with frequency content greater than 15 Hz are more effective at evoking these responses than those with little or no power at high frequencies.
-
Bounds on Query Convergence
12/2005;
The problem of finding an optimum using noisy evaluations of a smooth cost function arises in many contexts, including economics, business, medicine, experiment design, and foraging theory. We derive an asymptotic bound E[ (x_t - x*)^2 ] >= O(1/sqrt(t)) on the rate of convergence of a sequence (x... [more] The problem of finding an optimum using noisy evaluations of a smooth cost function arises in many contexts, including economics, business, medicine, experiment design, and foraging theory. We derive an asymptotic bound E[ (x_t - x*)^2 ] >= O(1/sqrt(t)) on the rate of convergence of a sequence (x_0, x_1, >...) generated by an unbiased feedback process observing noisy evaluations of an unknown quadratic function maximised at x*. The bound is tight, as the proof leads to a simple algorithm which meets it. We further establish a bound on the total regret, E[ sum_{i=1..t} (x_i - x*)^2 ] >= O(sqrt(t)) These bounds may impose practical limitations on an agent's performance, as O(eps^-4) queries are made before the queries converge to x* with eps accuracy.
-
6.26Impact points
Fast robust subject-independent magnetoencephalographic source localization using an artificial neural network.
Human brain mapping. 02/2005; 24(1):21-34.
We describe a system that localizes a single dipole to reasonable accuracy from noisy magnetoencephalographic (MEG) measurements in real time. At its core is a multilayer perceptron (MLP) trained to map sensor signals and head position to dipole location. Including head position overcomes the previo... [more] We describe a system that localizes a single dipole to reasonable accuracy from noisy magnetoencephalographic (MEG) measurements in real time. At its core is a multilayer perceptron (MLP) trained to map sensor signals and head position to dipole location. Including head position overcomes the previous need to retrain the MLP for each subject and session. The training dataset was generated by mapping randomly chosen dipoles and head positions through an analytic model and adding noise from real MEG recordings. After training, a localization took 0.7 ms with an average error of 0.90 cm. A few iterations of a Levenberg-Marquardt routine using the MLP output as its initial guess took 15 ms and improved accuracy to 0.53 cm, which approaches the natural limit on accuracy imposed by noise. We applied these methods to localize single dipole sources from MEG components isolated by blind source separation and compared the estimated locations to those generated by standard manually assisted commercial software.
-
Soft-LOST: EM on a Mixture of Oriented Lines
10/2004;
Robust clustering of data into overlapping linear subspaces is a common problem. Here we consider one-dimensional subspaces that cross the origin. This problem arises in blind source separation, where the subspaces correspond directly to columns of a mixing matrix. We present an algorithm that ident... [more] Robust clustering of data into overlapping linear subspaces is a common problem. Here we consider one-dimensional subspaces that cross the origin. This problem arises in blind source separation, where the subspaces correspond directly to columns of a mixing matrix. We present an algorithm that identifies these subspaces using an EM procedure, where the E-step calculates posterior probabilities assigning data points to lines and the M-step repositions the lines to match the points assigned to them. This method, combined with a transformation into a sparse domain and an L1-norm optimisation, constitutes a blind source separation algorithm for the under-determined case.
-
Monaural Source Separation using Spectral Cues
10/2004;
The acoustic environment poses at least two important challenges.
-
Subject-Independent Magnetoencephalographic Source Localization by a Multilayer Perceptron
03/2004;
We describe a system that localizes a single dipole to reasonable accuracy from noisy magnetoencephalographic (MEG) measurements in real time. At its core is a multilayer perceptron (MLP) trained to map sensor signals and head position to dipole location. Including head position overcomes the previo... [more] We describe a system that localizes a single dipole to reasonable accuracy from noisy magnetoencephalographic (MEG) measurements in real time. At its core is a multilayer perceptron (MLP) trained to map sensor signals and head position to dipole location. Including head position overcomes the previous need to retrain the MLP for each subject and session.
-
A Normative Model of Attention: Modulation of Neural Response
12/2003;
When a sensory stimulus is encoded in a lossy fashion for efficient transmission, there are necessarily tradeoffs between the represented fidelity of various aspects of the input pattern. In the model of attention presented here, a top-down signal informs the encoder of these tradeoffs. Given an ens... [more] When a sensory stimulus is encoded in a lossy fashion for efficient transmission, there are necessarily tradeoffs between the represented fidelity of various aspects of the input pattern. In the model of attention presented here, a top-down signal informs the encoder of these tradeoffs. Given an ensemble of input patterns and tradeoff requirements, our system can learn to encode its inputs optimally. This general model is instantiated in a simple network: an autoencoder with a bottleneck, innervated by a top-down attentional signal, trained using backpropagation. The only information the encoder receives concerning the semantics of the top-down attentional signal is from the optimization criterion, which penalizes the system more heavily for errors made near a simple attentional spotlight. The modulation of neural activity learned by this model qualitatively matches that measured in animals during covert visual attention tasks.
-
Brightness Illusions as Optimal Percepts
12/2003;
We show that several brightness illusions can be largely accounted for by assuming that the perceptual system performs simple Bayesian inference using a Gaussian image prior with noisy retinal ganglion cells. This theory accounts for phenomena which have been problematic for lateral-interaction mode... [more] We show that several brightness illusions can be largely accounted for by assuming that the perceptual system performs simple Bayesian inference using a Gaussian image prior with noisy retinal ganglion cells. This theory accounts for phenomena which have been problematic for lateral-interaction models while avoiding the complexities of mid-level vision theories that involve the estimation of structure and albedo.
Following (3)
-
Guido Nolte
Fraunhofer -
David Malone
National University of Ireland, Maynooth -
Sung Chan Jun
Gwangju Institute of Science and Technology
