Model building is a fundamental scientific tool. Increasingly there is interest in building neurally-implemented models of cognitive processes with the intention of modelling brains. However, simulation of such models can be prohibitively expensive in both the time and energy required. For example, Spaun - "the world's first functional brain model", comprising 2.5 million neurons - required 2.5 hours of computation for every second of simulation on a large compute cluster. SpiNNaker is a massively parallel, low power architecture specifically designed for the simulation of large neural models in biological real time. Ideally, SpiNNaker could be used to facilitate rapid simulation of models such as Spaun. However the Neural Engineering Framework (NEF), with which Spaun is built, maps poorly to the architecture - to the extent that models such as Spaun would consume vast portions of SpiNNaker machines and still not run as fast as biology. This thesis investigates whether real time simulation of Spaun on SpiNNaker is at all possible. Three techniques which facilitate such a simulation are presented. The first reduces the memory, compute and network loads consumed by the NEF. Consequently, it is demonstrated that only a twentieth of the cores are required to simulate a core component of the Spaun network than would otherwise have been needed. The second technique uses a small number of additional cores to significantly reduce the network traffic required to simulated this core component. As a result simulation in real time is shown to be feasible. The final technique is a novel logic minimisation algorithm which reduces the size of the routing tables which are used to direct information around the SpiNNaker machine. This last technique is necessary to allow the routing of models of the scale and complexity of Spaun. Together these provide the ability to simulate the Spaun model in biological real time - representing a speed-up of 9000 times over previously reported results - with room for much larger models on full-scale SpiNNaker machines.

No description available.

<p dir="ltr">Sinusoidal low frequency alternating current (LFAC) stimulation is a novel mode of electrical modulation observed in the Bioelectroics Lab in 2017. LFAC is capable of blocking the single fiber action potentials (APs) of the earthworm with only a few 100's of µA. The goal of this dissertation was to further explore and characterize the LFAC waveform to determine it's feasibility as a method for block in the mammalian peripheral nervous system (PNS). To better understand the mechanisms of LFAC block (LFACb), a blend of \textit{in-silico} modeling work was explored and the predictions were validated with <i>ex-vivo</i> and <i>in-vivo</i> experiments. </p><p dir="ltr"> This dissertation is divided into five chapters. The first chapter will explore the history of bioelectricity, the current state of <i>in-silico</i> modeling and methods of nerve block used in the PNS. The second chapter explores a major modeling assumption, the conductivity and permittivity of the nerve laminae of a mammalian nerve bundle. Four point electrochemical impedance spectroscopy (EIS) was performed on excised canine vagus nerve to evaluate the electrical properties of the perineurium and epineurium. This study's result, found that the corner frequency of the perineurium (2.6kHz) and epineurium (370Hz) were much lower than previously assumed. This explain a major difference between LFACb and the more established kilohertz frequency alternating current (kHFAC) block. The third chapter revisits the initial earthworm experiments during the discovery of LFACb. The effect of conduction slowing was observed in these earthworm experiments and were also seen in a mammalian canine vagus nerve and in the Horn-Yoshida-Schild (HYS) autonomic unmyelinated axon mode. These experiments showed that LFACb occurs as a cathodic block in which the sodium channels are held inactive. Chapter 4 explored the window between LFACb and LFAC activation (LFACa). The window between the two states was describes by LFAC amplitude and LFAC frequency in an <i>in-vivo</i> rat sciatic nerve and an <i>in-silico</i> model of a myelinated motor neuron, the McIntyre-Richardson-Grill (MRG) axon model. Geometrical effects were also observed by varying the bipolar pair of contacts used to deliver the LFACb waveform from an asymmetrical tripolar cuff electrode. Plantar flexor force measurements and electromyography (EMG) of the lateral gastrocnemius (LG) and soleus (Sol) were used to quantify the effects of the LFAC waveform. Convergence between <i>in-silico</i> modeling and <i>in-vivo</i> results showed promise that modeling efforts could be used with confidence to explore the LFAC block-activation more completly. LFACa was found to be highly dependent on frequency with increasing frequency lowering the threshold of activation. LFACb was shown to be mostly invariant to frequency. The final chapter takes the information found in this dissertation and summarizes it. Future work on LFAC is also proposed and the hypothesized results presented with the findings from this dissertation and available literature. </p>

Upper-limb myoelectric prostheses are controlled by muscle activity information recorded on the skin surface using electromyography (EMG). Intuitive prosthetic control can be achieved by deploying statistical and machine learning (ML) tools to decipher the user's movement intent from EMG signals. This thesis proposes various means of advancing the capabilities of non-invasive, ML-based control of myoelectric hand prostheses. Two main directions are explored, namely classification-based hand grip selection and proportional finger position control using regression methods. Several practical aspects are considered with the aim of maximising the clinical impact of the proposed methodologies, which are evaluated with offline analyses as well as real-time experiments involving both able-bodied and transradial amputee participants. It has been generally accepted that the EMG signal may not always be a reliable source of control information for prostheses, mainly due to its stochastic and non-stationary properties. One particular issue associated with the use of surface EMG signals for upper-extremity myoelectric control is the limb position effect, which is related to the lack of decoding generalisation under novel arm postures. To address this challenge, it is proposed to make concurrent use of EMG sensors and inertial measurement units (IMUs). It is demonstrated this can lead to a significant improvement in both classification accuracy (CA) and real-time prosthetic control performance. Additionally, the relationship between surface EMG and inertial measurements is investigated and it is found that these modalities are partially related due to reflecting different manifestations of the same underlying phenomenon, that is, the muscular activity. In the field of upper-limb myoelectric control, the linear discriminant analysis (LDA) classifier has arguably been the most popular choice for movement intent decoding. This is mainly attributable to its ease of implementation, low computational requirements, and acceptable decoding performance. Nevertheless, this particular method makes a strong fundamental assumption, that is, data observations from different classes share a common covariance structure. Although this assumption may often be violated in practice, it has been found that the performance of the method is comparable to that of more sophisticated algorithms. In this thesis, it is proposed to remove this assumption by making use of general class-conditional Gaussian models and appropriate regularisation to avoid overfitting issues. By performing an exhaustive analysis on benchmark datasets, it is demonstrated that the proposed approach based on regularised discriminant analysis (RDA) can offer an impressive increase in decoding accuracy. By combining the use of RDA classification with a novel confidence-based rejection policy that intends to minimise the rate of unintended hand motions, it is shown that it is feasible to attain robust myoelectric grip control of a prosthetic hand by making use of a single pair of surface EMG-IMU sensors. Most present-day commercial prosthetic hands offer the mechanical abilities to support individual digit control; however, classification-based methods can only produce pre-defined grip patterns, a feature which results in prosthesis under-actuation. Although classification-based grip control can provide a great advantage over conventional strategies, it is far from being intuitive and natural to the user. A potential way of approaching the level of dexterity enjoyed by the human hand is via continuous and individual control of multiple joints. To this end, an exhaustive analysis is performed on the feasibility of reconstructing multidimensional hand joint angles from surface EMG signals. A supervised method based on the eigenvalue formulation of multiple linear regression (MLR) is then proposed to simultaneously reduce the dimensionality of input and output variables and its performance is compared to that of typically used unsupervised methods, which may produce suboptimal results in this context. An experimental paradigm is finally designed to evaluate the efficacy of the proposed finger position control scheme during real-time prosthesis use. This thesis provides insight into the capacity of deploying a range of computational methods for non-invasive myoelectric control. It contributes towards developing intuitive interfaces for dexterous control of multi-articulated prosthetic hands by transradial amputees.

In a world dominated by temporal order, memory capable of processing, encoding and subsequently recalling ordered information is very important. Over the decades this memory, known as serial memory, has been extensively studied, and its effects are well known. Many models have also been developed, and while these models are able to reproduce the behavioural effects observed in human recall studies, they are not always implementable in a biologically plausible manner. This thesis presents the Ordinal Serial Encoding model, a model inspired by biology and designed with a broader view of general cognitive architectures in mind. This model has the advantage of simplicity, and we show how neuro-plausibility can be achieved by employing the principles of the Neural Engineering Framework in the model’s design. Additionally, we demonstrate that not only is the model able to closely mirror human performance in various recall tasks, but the behaviour of the model is itself a consequence of the underlying neural architecture.

Several decades of physiology, imaging and psychophysics research on attention has generated an enormous amount of data describing myriad forms of attentional effects. A similar breadth of theoretical models have been proposed that attempt to explain these effects in varying amounts of detail. However, there remains a need for neurally detailed mechanistic models of attention that connect more directly with various kinds of experimental data -- behavioural, psychophysical, neurophysiological, and neuroanatomical -- and that provide experimentally testable predictions. Research has been conducted that aims to identify neurally consistent principles that underlie selective attentional processing in cortex. The research specifically focuses on describing the functional mechanisms of attentional routing in a large-scale hierarchical model, and demonstrating the biological plausibility of the model by presenting a spiking neuron implementation that can account for a variety of attentional effects. The thesis begins by discussing several significant physiological effects of attention, and prominent brain areas involved in selective attention, which provide strong constraints for developing a model of attentional processing in cortex. Several prominent models of attention are then discussed, from which a set of common limitations in existing models is assembled that need to be addressed by the proposed model. One central limitation is that, for many existing models, it remains to be demonstrated that their computations can be plausibly performed in spiking neurons. Further, few models address attentional effects for more than a single neuron or single cortical area. And finally, few are able to account for different forms of attentional modulation in a single detailed model. These and other limitations are addressed by the Attentional Routing Circuit (ARC) proposed in this thesis. The presentation of the ARC begins with the proposal of a high-level mathematical model for selective routing in the visual hierarchy. The mathematical model is used to demonstrate that the suggested mechanisms allow for scale- and position-invariant representations of attended stimuli to be formed, and provides a functional context for interpreting detailed physiological effects. To evaluate the model's biological plausibility, the Neural Engineering Framework (NEF) is used to implement the ARC as a detailed spiking neuron model. Simulation results are then presented which demonstrate that selective routing can be performed efficiently in spiking neurons in a way that is consistent with the mathematical model. The neural circuitry for computing and applying attentional control signals in the ARC is then mapped on to neural populations in specific cortical laminae using known anatomical interlaminar and interareal connections to support the plausibility of its cortical implementation. The model is then tested for its ability to account for several forms of attentional modulation that have been reported in neurophysiological experiments. Three experiments of attention in macaque are simulated using the ARC, and for each of these experiments, the model is shown to be quantitatively consistent with measured data. Specifically, a study by Womelsdorf et al. (2008) demonstrates that spatial shifts of attention result in a shifting and shrinking of receptive fields depending on the target's position. An experiment by Treue and Martinez-Trujllo (1999) reports that attentional shifts between receptive field stimuli produce a multiplicative scaling of responses, but do not affect the neural tuning sensitivity. Finally, a study by Lee and Maunsell (2010) demonstrates that attentional shifts result in a multiplicative scaling of neural contrast-response functions that is consistent with a response-gain effect. The model accounts for each of these experimentally observed attentional effects using a single mechanism for selectively processing attended stimuli. In conclusion, it is suggested that the ARC is distinguished from previous models by providing a unifying interpretation of attentional effects at the level of single cells, neural populations, cortical areas, and over the bulk of the visual hierarchy. As well, there are several advantages of the ARC over previous models, including: (1) scalability to larger implementations without affecting the model's principles; (2) a significant increase in biological plausibility; (3) the ability to account for experimental results at multiple levels of analysis; (4) a detailed description of the model's anatomical substrate; (5) the ability to perform selective routing while preserving biological detail; and (6) generating a variety of experimentally testable predictions.

In a world dominated by temporal order, memory capable of processing, encoding and subsequently recalling ordered information is very important. Over the decades this memory, known as serial memory, has been extensively studied, and its effects are well known. Many models have also been developed, and while these models are able to reproduce the behavioural effects observed in human recall studies, they are not always implementable in a biologically plausible manner. This thesis presents the Ordinal Serial Encoding model, a model inspired by biology and designed with a broader view of general cognitive architectures in mind. This model has the advantage of simplicity, and we show how neuro-plausibility can be achieved by employing the principles of the Neural Engineering Framework in the model’s design. Additionally, we demonstrate that not only is the model able to closely mirror human performance in various recall tasks, but the behaviour of the model is itself a consequence of the underlying neural architecture.

The goal of the study was to use electrical signals from primary motor cortex to generate accurate predictions of the movement onset time of performed movements, for potential use in asynchronous brain-machine interface (BMI) systems. Four subjects, two with electroencephalogram and two with electrocorticogram electrodes, performed various movements while activity from their primary motor cortices was recorded. An analysis program used several criteria (change point, fractal dimension, spectral entropy, sum of differences, bandpower, bandpower integral, phase, and variance), derived from the neural recordings, to generate predictions of movement onset time, which it compared to electromyogram activity onset time, determining prediction accuracy by receiver operating characteristic curve areas. All criteria, excepting phase and change-point analysis, generated accurate predictions in some cases.

The goal of the study was to use electrical signals from primary motor cortex to generate accurate predictions of the movement onset time of performed movements, for potential use in asynchronous brain-machine interface (BMI) systems. Four subjects, two with electroencephalogram and two with electrocorticogram electrodes, performed various movements while activity from their primary motor cortices was recorded. An analysis program used several criteria (change point, fractal dimension, spectral entropy, sum of differences, bandpower, bandpower integral, phase, and variance), derived from the neural recordings, to generate predictions of movement onset time, which it compared to electromyogram activity onset time, determining prediction accuracy by receiver operating characteristic curve areas. All criteria, excepting phase and change-point analysis, generated accurate predictions in some cases.

Several decades of physiology, imaging and psychophysics research on attention has generated an enormous amount of data describing myriad forms of attentional effects. A similar breadth of theoretical models have been proposed that attempt to explain these effects in varying amounts of detail. However, there remains a need for neurally detailed mechanistic models of attention that connect more directly with various kinds of experimental data -- behavioural, psychophysical, neurophysiological, and neuroanatomical -- and that provide experimentally testable predictions. Research has been conducted that aims to identify neurally consistent principles that underlie selective attentional processing in cortex. The research specifically focuses on describing the functional mechanisms of attentional routing in a large-scale hierarchical model, and demonstrating the biological plausibility of the model by presenting a spiking neuron implementation that can account for a variety of attentional effects. The thesis begins by discussing several significant physiological effects of attention, and prominent brain areas involved in selective attention, which provide strong constraints for developing a model of attentional processing in cortex. Several prominent models of attention are then discussed, from which a set of common limitations in existing models is assembled that need to be addressed by the proposed model. One central limitation is that, for many existing models, it remains to be demonstrated that their computations can be plausibly performed in spiking neurons. Further, few models address attentional effects for more than a single neuron or single cortical area. And finally, few are able to account for different forms of attentional modulation in a single detailed model. These and other limitations are addressed by the Attentional Routing Circuit (ARC) proposed in this thesis. The presentation of the ARC begins with the proposal of a high-level mathematical model for selective routing in the visual hierarchy. The mathematical model is used to demonstrate that the suggested mechanisms allow for scale- and position-invariant representations of attended stimuli to be formed, and provides a functional context for interpreting detailed physiological effects. To evaluate the model's biological plausibility, the Neural Engineering Framework (NEF) is used to implement the ARC as a detailed spiking neuron model. Simulation results are then presented which demonstrate that selective routing can be performed efficiently in spiking neurons in a way that is consistent with the mathematical model. The neural circuitry for computing and applying attentional control signals in the ARC is then mapped on to neural populations in specific cortical laminae using known anatomical interlaminar and interareal connections to support the plausibility of its cortical implementation. The model is then tested for its ability to account for several forms of attentional modulation that have been reported in neurophysiological experiments. Three experiments of attention in macaque are simulated using the ARC, and for each of these experiments, the model is shown to be quantitatively consistent with measured data. Specifically, a study by Womelsdorf et al. (2008) demonstrates that spatial shifts of attention result in a shifting and shrinking of receptive fields depending on the target's position. An experiment by Treue and Martinez-Trujllo (1999) reports that attentional shifts between receptive field stimuli produce a multiplicative scaling of responses, but do not affect the neural tuning sensitivity. Finally, a study by Lee and Maunsell (2010) demonstrates that attentional shifts result in a multiplicative scaling of neural contrast-response functions that is consistent with a response-gain effect. The model accounts for each of these experimentally observed attentional effects using a single mechanism for selectively processing attended stimuli. In conclusion, it is suggested that the ARC is distinguished from previous models by providing a unifying interpretation of attentional effects at the level of single cells, neural populations, cortical areas, and over the bulk of the visual hierarchy. As well, there are several advantages of the ARC over previous models, including: (1) scalability to larger implementations without affecting the model's principles; (2) a significant increase in biological plausibility; (3) the ability to account for experimental results at multiple levels of analysis; (4) a detailed description of the model's anatomical substrate; (5) the ability to perform selective routing while preserving biological detail; and (6) generating a variety of experimentally testable predictions.