This laboratory is involved in physiological and histochemical studies of olfactory tissue grown in cell culture in an attempt to create an in vitro model of the olfactory system. The present study is an in-depth ultrastructural study of the morphology of cultured olfactory cells to determine the extent of similarities and differences between cultured tissues and the intact olfactory system in vivo.

No description available.

Modifiable health behaviors are a leading cause of mortality and chronic disease in the United States. Engagement in maladaptive health behaviors is linked to poor physical, psychological, and cognitive outcomes including increased risk of cardiovascular disease, Alzheimer's disease, anxiety, and depression. Using a neurobehavioral approach, I examined the hypothesis that neurobehaviors are impaired in clinical populations, and that exercise improves these neurobehaviors as well as the underlying mechanisms. In the first study, I found that a range of neurobehaviors are affected in individuals with obesity, indicating hyperactivity of the reward system and hypoactivity of the executive system. Using these neurobehaviors as predictors, I created a neurobehavioral model predicting obesity with an accuracy of 65%. In the second study, I examined neurobehaviors in a population of individuals in recovery from substance misuse. I found that neurobehaviors are altered in this population suggesting heightened activity of the executive system supports success in recovery. In the third study, I examined the effects of a long-term exercise program on a range of neurobehaviors and neurophysiology as measured through electroencephalography. I found that long-term exercise improved psychological state, memory, and attention. Additionally, I found that decreased cortical activity in response to exercise is associated with improvements in psychological state. Collectively, these findings suggest that there is a bi-directional relationship between the body and brain, with optimal physical health promoting optimal mental functioning. Additionally, these findings suggest that interventions that support improved neurobehaviors and neural circuitry are critical to promote engagement in positive health behaviors. / Doctor of Philosophy / Modifiable health behaviors are a leading cause of mortality and chronic disease in modern industrialized societies. Engagement in poor health behaviors is linked to increased risk of chronic disease affecting the body and brain including cardiovascular disease, Alzheimer's disease, anxiety, and depression. This dissertation explores the psychological and cognitive factors influencing engagement in healthy behaviors, and the ability of an exercise intervention to improve these factors as well as the underlying mechanisms. In the first study, I found that a range of neurobehavioral factors are impaired in individuals with obesity, and that these factors can be used to predict obesity. In the second study, I examined similar outcomes in a population of individuals in recovery from substance misuse, and found that neurobehaviors are altered in this population suggesting heightened activity of the executive system which supports successful recovery. In the third study I found that long-term exercise improved psychological and cognitive outcomes. Additionally, I found that changes in the electrical activity of the brain in response to exercise are associated with improvements in psychological state. Collectively, these findings suggest that there is a bi-directional relationship between the body and brain, with optimal physical health promoting optimal mental functioning. Additionally, these findings suggest that interventions that support improved neurobehaviors and neural circuitry are critical to promote engagement in positive health behaviors.

Evidence from recent studies suggested that visual areas (V1-V3) might be also actively involved in visually remembered features. The work presented in this thesis extended to other visual attributes, including motion coherence and contrast, with more detailed psychophysical and neuroimaging evidences to support the role of early visual areas in human visual working memory. The mnemonic characteristics of those visual traits were explored from temporal and spatial aspects using conventional psychophysical and functional brain imaging. Along the temporal domain, our results support the multi-channel theory that models the process occurred in early visual cortex when different visual features were perceived. In addition, experiments with “memory masking” procedures indicate that these ongoing representations show specificity of stimuli. Besides answering the question of whether early visual cortex was engaged in VWM, we also addressed the issue of how visual information is processed and held in these areas. We revealed that the visual system might exploit the “image-like” representation to discriminate stimuli with different contrasts, rather than only extracting and retaining the luminance differences. With the advent of up-to-date imaging technology, we also used Magnetic Resonance Imaging (MRI) to investigate the neural correlates of visual WM of contrast in human brain. In addition to the conventional fMRI analysis, we executed multivariate pattern analysis (MVPA), which allows more precise understandings of the structure and neural mechanism of WM. By using MVPA, we confirmed our previous psychophysical results. Along the spatial domain, this thesis provides further evidence of behaviour and brain activity when stimuli were displayed in different hemifields. Our findings showed the extra time and information cost when information relating the contrast of a stimulus was transmitted across hemispheres. Moreover, we unravelled the effect of spatial attention on primary visual cortex while visual features at different spatial locations were processed and memorized.

To improve the efficacy of addiction treatments it is important to understand the learning and behavioural processes involved. The experimental work presented here aims to further our understanding of the behaviour and related brain activity of smokers, using a range of experimental paradigms and Electroencephalographic (EEG) techniques. A range of behavioural tasks adapted from the animal literature for use with humans, were utilised to explore the choices made by smokers and the effect of smoking-related cues on drug-seeking behaviour. Tasks included concurrent choice and variations of the Pavlovian to Instrumental Transfer (PIT) task. EEG data was recorded during these experiments, and analysed using Event Related Potential (ERP) and frequency measures, to identify a neural component related to these effects. Resting EEG data was also collected and analysed to investigate the relationship between EEG frequency measures and individual difference measures. Behavioural results broadly replicated those found in both animal and human research; smoking related cues enhanced responding for smoking- related outcomes showing Pavlovian control of Instrumental behaviour. Extinction of the Pavlovian cues did not reduce instrumental responses in the transfer stage of a PIT task. However no ERP or frequency components were found that consistently correlated with these behavioural effects. The resting EEG data showed higher beta levels (less desynchronisation) in those with longer histories of smoking (four years and over) suggestive of either sensitisation or the loss of inhibitory neural control in long term smokers. In summary, the behavioural data adds support to a growing body of literature regarding the effects of cues on the behaviour of addicted individuals. More work and perhaps other techniques need to be utilised in order to explore the neural correlates of these behavioural effects and the resting data suggests a promising route for further research.

The mesoaccumbens dopamine system has been implicated in many basic psychological processes (e.g. "wanting" and "liking") and illnesses (e.g. addiction, depression, schizophrenia). However, the precise computational functions of nucleus accumbens and dopamine neurons within the system remain unknown. In this thesis, we test some of the current hypotheses regarding the function of this system using a behavioural neurophysiology approach in the rat. The first question we wanted to answer was whether nucleus accumbens neurons process reward-predictive stimuli (e.g. conditioned reinforcers) and reward delivery differently, since previous studies report equivocal findings. To do so, we trained thirsty rats to bar-press on a second-order schedule of saccharin reinforcement, within which the temporal pattern of rats' bar-pressing was reinforced by presentations of a conditioned reinforcer and primary reinforcer (reward). We found that nucleus accumbens neurons typically responded to these conditioned and primary reinforcers with opposite sign, which suggests they were processed differently. We were not sure whether responses to conditioned reinforcers encoded reward-prediction or facilitated a behavioural switch in the rat's behaviour. Indeed, since studies using a variety of experimental techniques have implicated the mesoaccumbens dopamine system in both reward prediction and behavioural switching, we sought to test whether neurons in the nucleus accumbens and dopamine-rich areas of the midbrain respond to outcome-associated stimuli to predict reward or switch behaviour. We found both sets of neurons predominantly did the former. Finally, to understand more about reward consummatory responses from both sets of neurons, we developed a rat behavioural task providing measures of reward "wanting" and "liking". In conclusion, on the basis of our data, the most parsimonious explanation for the function of the mesoaccumbens dopamine system is that it acts to modulate goal-seeking behaviour. Further research is required to identify the function of the interactions between nucleus accumbens and dopamine neurons during goal-seeking and goal consumption.

Dendrites are the most visually striking parts of neurons. Even so many neuron models are of point type and have no representation of space. In this thesis we will look at a range of neuronal models with the common property that we always include spatially extended dendrites. First we generalise Abbott’s “sum-over-trips” framework to include resonant currents. We also look at piece-wise linear (PWL) models and extend them to incorporate spatial structure in the form of dendrites. We look at the analytical construction of orbits for PWL models. By using both analytical and numerical Lyapunov exponent methods we explore phase space and in particular we look at mode-locked solutions. We will then construct the phase response curve (PRC) for a PWL system with compartmentally modelled dendrites. This sets us up so we can look at the effect of multiple PWL systems that are weakly coupled through gap junctions. We also attach a continuous dendrite to a PWL soma and investigate how the position of the gap junction influences network properties. After this we will present a short overview of neuronal plasticity with a special focus on the spatial effects. We also discuss attenuation of distal synaptic input and how this can be countered by dendritic democracy as this will become an integral part of our learning mechanisms. We will examine a number of different learning approaches including the tempotron and spike-time dependent plasticity. Here we will consider Poisson’s equation around a neural membrane. The membrane we focus on has Hodgkin-Huxley dynamics so we can study action potential propagation on the membrane. We present the Green’s function for the case of a one-dimensional membrane in a two-dimensional space. This will allow us to examine the action potential initiation and propagation in a multi-dimensional axon.

Radial glial cells play an important role during embryonic development in mammals. They are not only important for neural production but help to organise the architecture of the neocortex. Glial cells proliferate during the development of the brain in the embryo, before differentiating to produce neurons at a rate which increases towards the end of embryonic brain development. Glial cells communicate via Adenosine tri-phosphate (ATP) mediated calcium waves, which may have the effect of locally synchronising cell cycles, so that clusters of cells proliferate together, shedding cells in uniform sheets. Hence radial glial cells are not only responsible for the production of most neocortical neurons but also contribute to the architecture of the brain. It has been argued that human developmental disorders which are associated with cortical malfunctions such as infantile epilepsies and mental retardation may involve defects in neuronal production and/or architecture and mathematical modelling may shed some light upon these disorders. This thesis investigates, among other things, the conditions under which radial glial cells' cell cycles become `phase locked', radial glia proliferation and stochastic effects. There are various models for the cell cycle and for intracellular calcium dynamics. As part of our work, we marry two such models to form a model which incorporates the effect of calcium on the cell cycle of a single radial glial cell. Furthermore, with this achieved we consider populations of cells which communicate with each other via the secretion of ATP. Through bifurcation analysis, direct numerical simulation and the application of the theory of weakly coupled oscillators, we investigate and compare the behaviour of two models which differ from each other in the time during the cell cycle at which ATP is released. Our results from this suggest that cell cycle synchronisation is highly dependent upon the timing of ATP release. This in turn suggests that a malfunction in the timing of ATP release may be responsible for some cortical development disorders. We also show how the increase in radial glia proliferation may mostly be down to radial glial cells' ability to recruit quiescent cells onto the cell cycle. Furthermore, we consider models with an additive noise term and through the application of numerical techniques show that noise acts to advance the onset of oscillatory type solutions in both models. We build upon these results and show as a proof of concept how noise may act to enhance radial glia proliferation.

No description available.

No description available.