Point process filters have been applied successfully to decode neural signals and track neural dy... more Point process filters have been applied successfully to decode neural signals and track neural dynamics. Traditionally these methods assume that multiunit spiking activity has already been correctly spike-sorted. As a result, these methods are not appropriate for situations where sorting cannot be performed with high precision, such as real-time decoding for brain-computer interfaces. Because the unsupervised spike-sorting problem remains unsolved, we took an alternative approach that takes advantage of recent insights into clusterless decoding. Here we present a new point process decoding algorithm that does not require multiunit signals to be sorted into individual units. We use the theory of marked point processes to construct a function that characterizes the relationship between a covariate of interest (in this case, the location of a rat on a track) and features of the spike waveforms. In our example, we use tetrode recordings, and the marks represent a four-dimensional vector of the maximum amplitudes of the spike waveform on each of the four electrodes. In general, the marks may represent any features of the spike waveform. We then use Bayes's rule to estimate spatial location from hippocampal neural activity. We validate our approach with a simulation study and experimental data recorded in the hippocampus of a rat moving through a linear environment. Our decoding algorithm accurately reconstructs the rat's position from unsorted multiunit spiking activity. We then compare the quality of our decoding algorithm to that of a traditional spike-sorting and decoding algorithm. Our analyses show that the proposed decoding algorithm performs equivalent to or better than algorithms based on sorted single-unit activity. These results provide a path toward accurate real-time decoding of spiking patterns that could be used to carry out content-specific manipulations of population activity in hippocampus or elsewhere in the brain.
The 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2004
Developing optimal strategies for constructing and testing decoding algorithms is an important qu... more Developing optimal strategies for constructing and testing decoding algorithms is an important question in computational neuroscience, In this field, decoding algorithms are mathematical methods that model ensemble neural spiking activity as they dynamically represent a biological signal. We present a recursive decoding algorithm based on a Bayesian point process model of individual neuron spiking activity and a linear stochastic state-space
The Journal of neuroscience : the official journal of the Society for Neuroscience, Jan 15, 1998
The problem of predicting the position of a freely foraging rat based on the ensemble firing patt... more The problem of predicting the position of a freely foraging rat based on the ensemble firing patterns of place cells recorded from the CA1 region of its hippocampus is used to develop a two-stage statistical paradigm for neural spike train decoding. In the first, or encoding stage, place cell spiking activity is modeled as an inhomogeneous Poisson process whose instantaneous rate is a function of the animal's position in space and phase of its theta rhythm. The animal's path is modeled as a Gaussian random walk. In the second, or decoding stage, a Bayesian statistical paradigm is used to derive a nonlinear recursive causal filter algorithm for predicting the position of the animal from the place cell ensemble firing patterns. The algebra of the decoding algorithm defines an explicit map of the discrete spike trains into the position prediction. The confidence regions for the position predictions quantify spike train information in terms of the most probable locations of the ...
The hippocampus is required for the formation of episodic memories, but the neural mechanisms und... more The hippocampus is required for the formation of episodic memories, but the neural mechanisms underlying memory storage in the hippocampus remain unclear. Currently, the dominant hypothesis is that memory patterns are stored as attractor states in the recurrent CA3 network1. To ...
The hippocampus is required for the encoding, consolidation and retrieval of event memories. Alth... more The hippocampus is required for the encoding, consolidation and retrieval of event memories. Although the neural mechanisms that underlie these processes are only partially understood, a series of recent papers point to awake memory replay as a potential contributor to both consolidation and retrieval. Replay is the sequential reactivation of hippocampal place cells that represent previously experienced behavioral trajectories and occurs frequently in the awake state, particularly during periods of relative immobility. Awake replay may reflect trajectories through either the current environment or previously visited environments that are spatially remote. The repetition of learned sequences on a compressed time scale is well suited to promote memory consolidation in distributed circuits beyond the hippocampus, suggesting that consolidation occurs in both the awake and sleeping animal. Moreover, sensory information can influence the content of awake replay, suggesting a role for awake replay in memory retrieval.
Neural spike train decoding algorithms are important tools for characterizing how ensembles of ne... more Neural spike train decoding algorithms are important tools for characterizing how ensembles of neurons represent biological signals. We present a Bayesian neural spike train decoding algorithm based on a point process model of individual neurons, a linear stochastic state-space model of the biological signal, and a temporal latency parameter. The latency parameter represents the temporal lead or lag between the biological signal and the ensemble spiking activity. We use the algorithm to study whether the representation of position by the ensemble spiking activity of pyramidal neurons in the CA1 region of the rat hippocampus is more consistent with prospective coding, i.e., future position, or retrospective coding, past position. Using 44 simultaneously recorded neurons and an ensemble delay latency of 400 ms, the median decoding error was 5.1 cm during 10 min of foraging in an open circular environment. The true coverage probability for the algorithm's 0.95 confidence regions was 0.71. These results illustrate how the Bayesian neural spike train decoding paradigm may be used to investigate spatio-temporal representations of position by an ensemble of hippocampal neurons.
We analyze the stochastic structure of CA1 hippocampal place cell spiking activity with stimulus–... more We analyze the stochastic structure of CA1 hippocampal place cell spiking activity with stimulus–response models based on inhomogeneous Poisson (IP), inhomogeneous gamma (IG) and inhomogeneous inverse Gaussian (IIG) interspike interval probability densities that have Markov dependence. We present a technique based on quantile–quantile (Q–Q) plots derived from the intensity-rescaling transformation, and use it along with Akaike's (AIC) and Bayesian (BIC)
The formation of spatial memories is believed to depend on long-term potentiation (LTP) of synaps... more The formation of spatial memories is believed to depend on long-term potentiation (LTP) of synapses within the hippocampal network. In this issue of Neuron, Dragoi et al. demonstrate that LTP can cause changes in the hippocampal representation of space without disrupting network dynamics. These results help define the elusive relationship between cellular-level synaptic plasticity and systems-level neural coding.
A paradigm for constructing and analyzing non-Poisson stimulus-response models of neural spike tr... more A paradigm for constructing and analyzing non-Poisson stimulus-response models of neural spike train activity is presented. Inhomogeneous gamma (IG) and inverse Gaussian (IIG) probability models are constructed by generalizing the derivation of the inhomogeneous Poisson (IP) model from the exponential probability density. The resultant spike train models have Markov dependence. Quantile-quantile (Q -Q) plots and Kolmogorov-Smirnov (K -S) plots are
The medial temporal lobe is crucial for the ability to learn and retain new declarative memories.... more The medial temporal lobe is crucial for the ability to learn and retain new declarative memories. This form of memory includes the ability to quickly establish novel associations between unrelated items. To better understand the patterns of neural activity during associative memory formation, we recorded the activity of hippocampal neurons of macaque monkeys as they learned new associations. Hippocampal neurons
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2014
The brain is a massively interconnected network of specialized circuits. Even primary sensory are... more The brain is a massively interconnected network of specialized circuits. Even primary sensory areas, once thought to support relatively simple, feed-forward processing, are now known to be parts of complex feedback circuits. All brain functions depend on millisecond timescale interactions across these brain networks. Current approaches cannot measure or manipulate such large-scale interactions. Here we demonstrate that polymer-based, penetrating, micro-electrode arrays can provide high quality neural recordings from awake, behaving animals over periods of months. Our results indicate that polymer electrodes are a viable substrate for the development of systems that can record from thousands of channels across months to years. This is our first step towards developing a 1000+ electrode system capable of providing high-quality, long-term neural recordings.
Microelectrode arrays for neural interface devices that are made of biocompatible thin-film polym... more Microelectrode arrays for neural interface devices that are made of biocompatible thin-film polymer are expected to have extended functional lifetime because the flexible material may minimize adverse tissue response caused by micromotion. However, their flexibility prevents them from being accurately inserted into neural tissue. This article demonstrates a method to temporarily attach a flexible microelectrode probe to a rigid stiffener using biodissolvable polyethylene glycol (PEG) to facilitate precise, surgical insertion of the probe. A unique stiffener design allows for uniform distribution of the PEG adhesive along the length of the probe. Flip-chip bonding, a common tool used in microelectronics packaging, enables accurate and repeatable alignment and attachment of the probe to the stiffener. The probe and stiffener are surgically implanted together, then the PEG is allowed to dissolve so that the stiffener can be extracted leaving the probe in place. Finally, an in vitro test method is used to evaluate stiffener extraction in an agarose gel model of brain tissue. This approach to implantation has proven particularly advantageous for longer flexible probes (>3 mm). It also provides a feasible method to implant dual-sided flexible probes. To date, the technique has been used to obtain various in vivo recording data from the rat cortex.
The entorhinal cortex (EC) is an essential component of the medial temporal lobe long-term-memory... more The entorhinal cortex (EC) is an essential component of the medial temporal lobe long-term-memory system. Recently, Egorov et al. demonstrated that neurons in layer 5 of the EC show graded persistent activity: each neuron could maintain a constant rate firing over several minutes, even in the absence of input from other neurons. These findings suggest that layer-5 EC neurons could play a central role in the formation of associations among stimuli that occur at different times.
Point process filters have been applied successfully to decode neural signals and track neural dy... more Point process filters have been applied successfully to decode neural signals and track neural dynamics. Traditionally these methods assume that multiunit spiking activity has already been correctly spike-sorted. As a result, these methods are not appropriate for situations where sorting cannot be performed with high precision, such as real-time decoding for brain-computer interfaces. Because the unsupervised spike-sorting problem remains unsolved, we took an alternative approach that takes advantage of recent insights into clusterless decoding. Here we present a new point process decoding algorithm that does not require multiunit signals to be sorted into individual units. We use the theory of marked point processes to construct a function that characterizes the relationship between a covariate of interest (in this case, the location of a rat on a track) and features of the spike waveforms. In our example, we use tetrode recordings, and the marks represent a four-dimensional vector of the maximum amplitudes of the spike waveform on each of the four electrodes. In general, the marks may represent any features of the spike waveform. We then use Bayes's rule to estimate spatial location from hippocampal neural activity. We validate our approach with a simulation study and experimental data recorded in the hippocampus of a rat moving through a linear environment. Our decoding algorithm accurately reconstructs the rat's position from unsorted multiunit spiking activity. We then compare the quality of our decoding algorithm to that of a traditional spike-sorting and decoding algorithm. Our analyses show that the proposed decoding algorithm performs equivalent to or better than algorithms based on sorted single-unit activity. These results provide a path toward accurate real-time decoding of spiking patterns that could be used to carry out content-specific manipulations of population activity in hippocampus or elsewhere in the brain.
The 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2004
Developing optimal strategies for constructing and testing decoding algorithms is an important qu... more Developing optimal strategies for constructing and testing decoding algorithms is an important question in computational neuroscience, In this field, decoding algorithms are mathematical methods that model ensemble neural spiking activity as they dynamically represent a biological signal. We present a recursive decoding algorithm based on a Bayesian point process model of individual neuron spiking activity and a linear stochastic state-space
The Journal of neuroscience : the official journal of the Society for Neuroscience, Jan 15, 1998
The problem of predicting the position of a freely foraging rat based on the ensemble firing patt... more The problem of predicting the position of a freely foraging rat based on the ensemble firing patterns of place cells recorded from the CA1 region of its hippocampus is used to develop a two-stage statistical paradigm for neural spike train decoding. In the first, or encoding stage, place cell spiking activity is modeled as an inhomogeneous Poisson process whose instantaneous rate is a function of the animal's position in space and phase of its theta rhythm. The animal's path is modeled as a Gaussian random walk. In the second, or decoding stage, a Bayesian statistical paradigm is used to derive a nonlinear recursive causal filter algorithm for predicting the position of the animal from the place cell ensemble firing patterns. The algebra of the decoding algorithm defines an explicit map of the discrete spike trains into the position prediction. The confidence regions for the position predictions quantify spike train information in terms of the most probable locations of the ...
The hippocampus is required for the formation of episodic memories, but the neural mechanisms und... more The hippocampus is required for the formation of episodic memories, but the neural mechanisms underlying memory storage in the hippocampus remain unclear. Currently, the dominant hypothesis is that memory patterns are stored as attractor states in the recurrent CA3 network1. To ...
The hippocampus is required for the encoding, consolidation and retrieval of event memories. Alth... more The hippocampus is required for the encoding, consolidation and retrieval of event memories. Although the neural mechanisms that underlie these processes are only partially understood, a series of recent papers point to awake memory replay as a potential contributor to both consolidation and retrieval. Replay is the sequential reactivation of hippocampal place cells that represent previously experienced behavioral trajectories and occurs frequently in the awake state, particularly during periods of relative immobility. Awake replay may reflect trajectories through either the current environment or previously visited environments that are spatially remote. The repetition of learned sequences on a compressed time scale is well suited to promote memory consolidation in distributed circuits beyond the hippocampus, suggesting that consolidation occurs in both the awake and sleeping animal. Moreover, sensory information can influence the content of awake replay, suggesting a role for awake replay in memory retrieval.
Neural spike train decoding algorithms are important tools for characterizing how ensembles of ne... more Neural spike train decoding algorithms are important tools for characterizing how ensembles of neurons represent biological signals. We present a Bayesian neural spike train decoding algorithm based on a point process model of individual neurons, a linear stochastic state-space model of the biological signal, and a temporal latency parameter. The latency parameter represents the temporal lead or lag between the biological signal and the ensemble spiking activity. We use the algorithm to study whether the representation of position by the ensemble spiking activity of pyramidal neurons in the CA1 region of the rat hippocampus is more consistent with prospective coding, i.e., future position, or retrospective coding, past position. Using 44 simultaneously recorded neurons and an ensemble delay latency of 400 ms, the median decoding error was 5.1 cm during 10 min of foraging in an open circular environment. The true coverage probability for the algorithm's 0.95 confidence regions was 0.71. These results illustrate how the Bayesian neural spike train decoding paradigm may be used to investigate spatio-temporal representations of position by an ensemble of hippocampal neurons.
We analyze the stochastic structure of CA1 hippocampal place cell spiking activity with stimulus–... more We analyze the stochastic structure of CA1 hippocampal place cell spiking activity with stimulus–response models based on inhomogeneous Poisson (IP), inhomogeneous gamma (IG) and inhomogeneous inverse Gaussian (IIG) interspike interval probability densities that have Markov dependence. We present a technique based on quantile–quantile (Q–Q) plots derived from the intensity-rescaling transformation, and use it along with Akaike's (AIC) and Bayesian (BIC)
The formation of spatial memories is believed to depend on long-term potentiation (LTP) of synaps... more The formation of spatial memories is believed to depend on long-term potentiation (LTP) of synapses within the hippocampal network. In this issue of Neuron, Dragoi et al. demonstrate that LTP can cause changes in the hippocampal representation of space without disrupting network dynamics. These results help define the elusive relationship between cellular-level synaptic plasticity and systems-level neural coding.
A paradigm for constructing and analyzing non-Poisson stimulus-response models of neural spike tr... more A paradigm for constructing and analyzing non-Poisson stimulus-response models of neural spike train activity is presented. Inhomogeneous gamma (IG) and inverse Gaussian (IIG) probability models are constructed by generalizing the derivation of the inhomogeneous Poisson (IP) model from the exponential probability density. The resultant spike train models have Markov dependence. Quantile-quantile (Q -Q) plots and Kolmogorov-Smirnov (K -S) plots are
The medial temporal lobe is crucial for the ability to learn and retain new declarative memories.... more The medial temporal lobe is crucial for the ability to learn and retain new declarative memories. This form of memory includes the ability to quickly establish novel associations between unrelated items. To better understand the patterns of neural activity during associative memory formation, we recorded the activity of hippocampal neurons of macaque monkeys as they learned new associations. Hippocampal neurons
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2014
The brain is a massively interconnected network of specialized circuits. Even primary sensory are... more The brain is a massively interconnected network of specialized circuits. Even primary sensory areas, once thought to support relatively simple, feed-forward processing, are now known to be parts of complex feedback circuits. All brain functions depend on millisecond timescale interactions across these brain networks. Current approaches cannot measure or manipulate such large-scale interactions. Here we demonstrate that polymer-based, penetrating, micro-electrode arrays can provide high quality neural recordings from awake, behaving animals over periods of months. Our results indicate that polymer electrodes are a viable substrate for the development of systems that can record from thousands of channels across months to years. This is our first step towards developing a 1000+ electrode system capable of providing high-quality, long-term neural recordings.
Microelectrode arrays for neural interface devices that are made of biocompatible thin-film polym... more Microelectrode arrays for neural interface devices that are made of biocompatible thin-film polymer are expected to have extended functional lifetime because the flexible material may minimize adverse tissue response caused by micromotion. However, their flexibility prevents them from being accurately inserted into neural tissue. This article demonstrates a method to temporarily attach a flexible microelectrode probe to a rigid stiffener using biodissolvable polyethylene glycol (PEG) to facilitate precise, surgical insertion of the probe. A unique stiffener design allows for uniform distribution of the PEG adhesive along the length of the probe. Flip-chip bonding, a common tool used in microelectronics packaging, enables accurate and repeatable alignment and attachment of the probe to the stiffener. The probe and stiffener are surgically implanted together, then the PEG is allowed to dissolve so that the stiffener can be extracted leaving the probe in place. Finally, an in vitro test method is used to evaluate stiffener extraction in an agarose gel model of brain tissue. This approach to implantation has proven particularly advantageous for longer flexible probes (>3 mm). It also provides a feasible method to implant dual-sided flexible probes. To date, the technique has been used to obtain various in vivo recording data from the rat cortex.
The entorhinal cortex (EC) is an essential component of the medial temporal lobe long-term-memory... more The entorhinal cortex (EC) is an essential component of the medial temporal lobe long-term-memory system. Recently, Egorov et al. demonstrated that neurons in layer 5 of the EC show graded persistent activity: each neuron could maintain a constant rate firing over several minutes, even in the absence of input from other neurons. These findings suggest that layer-5 EC neurons could play a central role in the formation of associations among stimuli that occur at different times.
Uploads
Papers by Loren Frank