Epileptic seizures, characterised by paroxysmal disturbance of brain electrical activity, are recognisable by temporary change in clinical state (for example motor signs, behavioural modification or altered conscious level), temporally associated with the cerebral discharge. While analysis of such clinical seizure signs (“semiology”) formed the main basis of epilepsy study from the late 19th century onwards, understanding of the neural basis of semiological expression has advanced relatively little, in comparison to other aspects of epilepsy research. Analysis of ictal clinical signs is today considered essential for diagnosis of epilepsy, offering clues to underlying anatomical localisation and pathophysiology; however, paradoxically, the cerebral substrate of semiological signs remains incompletely understood in many cases and its localising value is therefore debated. Characterising the anatomo-pathophysiological basis of seizure semiology is especially important in the context of epilepsy pre-surgical evaluation, even more so when no radiologically visible lesion is present, since semiological analysis, if validated for a given seizure type, offers crucial localising information. For pharmacoresistant focal epilepsies in which surgical treatment might be possible, a number of cases require intracranial EEG recording. The method of stereoelectroencephalography (SEEG) is particularly useful as this allows simultaneous exploration of multiple, distant brain structures using stereotaxically placed multi-lead electrodes with concurrent video recording. The data thus acquired help form a three dimensional view of spatio-temporal seizure dynamics. Using SEEG it is therefore possible to undertake detailed analysis of semiological patterns and to study their temporal relation to the abnormal electrical cerebral activity occurring in brain networks during seizures. Epileptic seizures characterised clinically by transient cognitive dysfunction, behavioral change and complex motor signs are particularly challenging to analyse and categorise semiologically; indeed any paroxysmal behavioral disturbance must also be analysed with regards to whether it is actually caused by an epileptic discharge or not, since other forms of pathology, particularly psychogenic nonepileptic seizures (PNES), may be difficult to distinguish from epileptic seizures on a purely clinical basis, and require video-EEG recording for confirmation. This issue is particularly pertinent for prefrontal and parietal lobe seizures, which pose specific challenges for electroclinical analysis. PNES have a different and as yet poorly defined neurobiological basis compared to epileptic seizures. However growing understanding of the brain networks underlying emotional dysfunction, complex motor behaviour and altered consciousness, in particular data derived from intracranial studies of epileptic seizures, can help to further knowledge of how altered activity within these neural networks might interact with psychological and other factors in the pathophysiology of PNES. Through detailed observation of multiple epileptic seizures across a large population of patients, it can be appreciated that similarities exist in both clinical pattern and anatomical organisation of seizures. The existence of semiological patterns is in favour of the hypothesis that specific neural circuits underlie some forms of behavioural expression, and thus reinforces the validity of pursuing this line of investigation in epileptic seizures.

Objective: Cognitive Estimation Tests (CETs) are used to assess decision-making. Previous versions include culturally- biased questions likely to disadvantage certain sections of the population. This study aimed to develop a new culture fair questionnaire and assess its reliability and validity. Method: A 30-item questionnaire was developed and assessed for culture fairness. A normative range of answers was gathered, and a scale developed to define level of deviation from typical responses. Performance in a group of people with brain injury was compared to a matched group of healthy controls. Those with brain injury deemed able to make significant life decisions were compared with a group considered to lack this capacity, to determine whether this test may be useful when assessing decision-making capacity. Correlational analyses were conducted to determine whether there was a relationship between the test and performance on the Dysexecutive Questionnaire (DEX), a measure of everyday executive functioning. Test-retest reliability was examined with 30 of the normative sample. Results: Results confirm previous literature showing that those with brain injury perform significantly worse than healthy controls. The test did not discriminate between patients with and without capacity to make important decisions, did not significantly correlate with the total score on the DEX and demonstrated relatively poor consistency. Conclusions: Based on these results, CETs do not appear to be reliable or valid enough for use in clinical assessments. A sub-set of the most sensitive items may prove useful, but further work is required to examine the reliability and validity of this item subset in new samples.

Cysteine Dioxygenase is an enzyme that catalyses the reaction of cysteine to cysteinesulfinic acid and is thought to regulate the intracellular concentrations of its substrate. The enzyme may also be involved in the oxidation of exposed sulfidrils of protein and by implication, involved in cell signalling or regulation. Substrate and products of cysteine dioxygenase are known to be disrupted in a number of neurodegenerative diseases making the protein an important target for medical research. To date, the structure of the protein has not been solved. An understanding of its structure and reaction mechanism will further the understanding of the aetiology of a number of different pathologies. Its structure may also further the development of novel pharmacological drugs.

Mental retardation is the most common brain disease. One of the first genes identified in X-linked mental retardation (XLMR) was the OPHN-1 gene. Mutation of this gene has been described in patients with moderate to severe cognitive impairments. MR is characterized by reduced cognitive function with or without other clinical features, thus providing a direct approach to study the neurobiology of cognition and pathogenesis of MR. I propose in this thesis to clarify the underlying mechanisms responsible for the learning impairments. My first approach was to investigate the functioning of a neuronal population using extracellular recording of fast oscillations which are thought to underlie higher cognitive performance. I showed that \(Ophn-1\) null mice displayed weaker gamma oscillations. Thereafter, Investigation of the synaptic properties of CA3 pyramidal neurons using the patch-clamp technique has been undertaken. I have shown reduced inputs of excitatory and inhibitory neurotransmission to CA3 pyramidal neurons accompanied with reduced frequency dependent facilitation of the inhibitory neurotransmission at 33Hz. Finally, a reduction in readily releasable pool size in inhibitory synapses of CA3 area was unravelled. This defect explained the reduction of frequency of sIPSCs and consequently the reduction in gamma oscillations power in Ophn-1\(^{-/y}\) slices.

This thesis examines perspectives on gender in eating disorders. Chapter one critically reviews research into psychosocial influences on adolescent boys' concerns about eating and body shape. [...] Chapter two is an empirical study of therapists' experiences of working with men with eating disorders. [...] Chapter three provides a reflective account of the author's experiences of being a Trainee Clinical Psychologist working with female patients in an eating disorder service.

In this thesis, we develop some novel models and methods for the analysis of both structural and functional brain images, and their joint analysis with genetic data. In the first project, we present a suite of methods to increase the power of whole-brain genome-wide association studies. We introduce a kernel-based multilocus model to capture the interactions between single nucleotide polymorphisms (SNPs) and model their joint effect on the imaging traits. We provide a fast implementation of voxel- and cluster-wise inferences based on random field theory to take full use of the 3D spatial information in images and account for multiple comparison problems. We also propose a fast permutation procedure to increase the efficiency of standard permutation methods and provide accurate small p-value estimates based on parametric tail approximation. We explore the relationship between 448,294 SNPs and 18,043 genes in 31,662 voxels of the entire brain across 740 elderly subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI), and show boosted power of our approach by making head-to-head comparisons with previous voxel-wise genome-wide association studies. The various advantages of our methods over existing approaches indicate a great potential offered by this novel framework to detect genetic in uences on human brains. In the second project, we present a Bayesian spatial model of multiple sclerosis binary lesion maps based on a spatial generalized linear mixed model with spatially varying coefficients. Our model fully respects the binary nature of the data and the spatial structure of the lesion maps, as opposed to existing massive univariate methods, and produces regularized (smoothed) estimates of lesion incidence without an arbitrary smoothing parameter. Our model also allows for explicit modeling of the spatially varying effects of subject specific covariates such as age, gender, disease duration and disabilities scores, producing spatial maps of these effects and their significance, as well as the (scalar) effect of spatially varying covariates such as the fraction of white matter in each voxel. We apply our model to binary lesion maps derived from T2-weighted MRI images from 250 multiple sclerosis patients classified into five clinical subtypes, and determine the spatial dependence between lesion location and subject specific covariates. We also demonstrate dramatically improved predictive capabilities of our model over existing methods.

Myoclonus Dystonia Syndrome is a childhood onset hyperkinetic movement disorder characterised by alcohol responsive upper body myoclonus and dystonia. A proportion of cases are due to mutations in the maternally imprinted SGCE gene, which encodes the transmembrane epsilon-sarcoglycan protein. Previous studies suggest an increased rate of psychiatric disorders in those with SGCE mutations. This study aimed to establish a cohort of myoclonus dystonia syndrome patients, identify the rate and type of SGCE mutations, determine differences in motor characteristics between mutation positive and negative cases and whether psychiatric disorders form part of the disease phenotype. Eighty-nine probands with clinically suspected MDS were recruited. Information regarding onset and distribution of motor symptoms was collected via systematic questionnaires and video taped examination. SGCE was analysed using direct sequencing and for copy number variants. Psychiatric symptoms were assessed using systematic and standardised questionnaires and compared to a disability-matched, alcohol responsive tremor control group. Nineteen (21%) probands had an SGCE mutation. All had evidence of upper body predominant myoclonus and dystonia during their disease course. Five had contiguous gene deletions ranging from 0.7 to 2.3Mb in size with distinctive clinical features. Recruitment of family members increased the affected SGCE mutation positive group to 27 of whom 21 (77%) had psychiatric symptoms. Obsessive-Compulsive Disorder was eight times more likely (p<0.001) in mutation positive cases, compulsivity being the predominant feature (p<0.001). Generalized Anxiety Disorder (p=0.003) and alcohol dependence (p=0.02) were five times more likely in cases than tremor controls. Overall, SGCE mutations are associated with a narrow clinical and specific psychiatric phenotype. The presence of myoclonus, dystonia, age at onset ≤10 years and a positive family history of the disorder are the strongest predictors of an SGCE mutation. SGCE mutations are likely to have a pleiotropic effect in causing both motor and specific psychiatric symptoms.

The formation of a visual percept in the human brain involves multiple processes, the extent of which may be related to each other within individuals but may show variability between participants. The aim of this thesis was to investigate the relationships between individual variability in various measures of visual processing. The non-invasive neuroimaging methods of magnetic resonance imaging (MRI), magnetic resonance spectroscopy (MRS), magnetoencephalography (MEG) and functional MRI (fMRI) were employed to measure brain structure, neurotransmitter concentration, neuronal oscillations and haemodynamic activity, respectively. Reductions in haemodynamic activity in non-stimulated areas of visual cortex (negative blood oxygen level-dependent responses) were shown to hold useful information about the stimulus, in addition to the responses in stimulated cortex. In general, there were no strong relationships between increased or decreased functional responses in visual cortex and measures of brain structure or of neurotransmitter concentration. Age was the major determinant of individual variability in the frequency of neuronal oscillations. These findings do not replicate results from previous studies that have shown links between individual differences in these measures. This discrepancy was not due to poor repeatability of MRS measures, since methods for optimisation of MRS were identified in this thesis. Simulations were also conducted to determine the sample sizes required in correlational studies involving neuroimaging measures with associated measurement noise. The lack of replication of relationships between neuroimaging measures of individual differences in visual processing is likely to be influenced by low statistical power, due to the small sample sizes and weak relationships tested. Such studies should therefore be conducted and interpreted cautiously, bearing in mind issues of power, demographic mediators of relationships and the likely strength of relationships inferred from the physiological mechanisms linking the variables tested.

Offspring of depressed mothers are at increased risk of developing a variety of psychopathologies. Risk factors and mechanisms for the development of these heterogeneous outcomes are poorly understood. Disruptions in the mother-child relationship may be one mechanism by which maternal depression increases risk for offspring psychopathology. Many adults with depression present with co-occurring psychopathology, but how these co-occurring problems affect offspring risk, or impact upon the mother-child relationship, has rarely been considered. Data were from the Early Prediction of Adolescent Depression study. Mothers with a history of recurrent major depressive disorder and their adolescent offspring were assessed three times between 2007 and 2011. Mothers completed questionnaires assessing their own depression severity and co-occurring psychopathology (anxiety, antisocial behaviour (ASB), and alcohol misuse). Offspring psychopathology (presence of psychiatric disorder, symptoms of depression, anxiety and disruptive behaviour disorder (DBD)) were assessed using the Child and Adolescent Psychiatric Assessment. The mother-child relationship was assessed using parent-rated questionnaires and an interviewer-rated speech sample. Co-occurring problems in mothers predicted new-onset psychiatric disorder in offspring; this remained significant after controlling for maternal depression severity. When investigating the specificity of risk for offspring, maternal co-occurring ASB was specifically associated with offspring DBD, whereas maternal depression severity predicted offspring depression. The mother-child relationship mediated the effect of maternal depression severity on risk for offspring psychopathology. However this was better accounted for by co-occurring maternal ASB, which predicted both maternal warmth and hostility. Maternal hostility was a specific risk factor for offspring DBD. Bidirectional effects were observed between offspring DBD and maternal hostility. Findings highlight the importance of assessing co-occurring psychopathology in mothers with recurrent depression when considering risk for offspring. Parenting interventions that reduce hostility may be beneficial for preventing or reducing adolescent DBD, particularly when depressed mothers report additional ASB. Furthermore, interventions that reduce offspring DBD may also show benefits for the mother-child relationship.

Schizophrenia is a debilitating psychiatric disorder with high heritability. Genome wide association studies (GWASs) have identified association between schizophrenia and an intronic single nucleotide polymorphism (SNP) in zinc finger domain containing 804A (ZNF804A). The biological functions of ZNF804A are largely unknown. Thus, the aim of this thesis was to determine the function of ZNF804A. Yeast two-hybrid (Y2H) screens were used to determine the protein binding partners of ZNF804A in the brain. This identified ZNF804A interacted with transcription factors (zinc finger protein 40 (ZNF40); trans-acting transcription factor 1 (Sp1) and basic helix-loop-helix family, member e40 (Bhlhe40)) and regulators of pre-mRNA splicing, including RNA binding protein, fox-1 and -2 (Rbfox1 and RBFOX2) and neuro-oncological ventral antigen 2 (NOVA2). Therefore, we hypothesised that, via interactions with its protein binding partners, ZNF804A may have a role in regulating transcription and pre-mRNA splicing. Exon arrays were used to determine the effects of ZNF804A knockdown on gene expression in SH-SY5Y neuroblastoma cells. Enrichment analysis on the differentially expressed or spliced genes indicated a significant effect of ZNF804A knockdown on genes involved in nervous system development, particularly synaptic contact and axon guidance. Among the most significantly differentially expressed genes were the known schizophrenia susceptibility genes reelin (RELN) and neuropeptide Y (NPY). Several alternative splicing events were confirmed empirically, including increased exclusion of exon 11a of enabled homolog (ENAH). Consistent with our hypothesis, splicing of exon 11a of ENAH is known to be regulated by RBFOX2. In complementary experiments, exon arrays were used to identify differentially expressed genes in a stable cell line expressing myc-ZNF804A. Enrichment analysis on the differentially expressed genes indicated an over-representation of genes involved in the regulation of epithelial to mesenchymal transition and receptor-mediated axon growth repulsion. Among the genes with the largest fold changes in expression was a gene implicated in synapse development: secreted protein acidic and rich in cysteine (SPARC). Enrichment analysis of the alternatively spliced genes indicated a significant effect of myc-ZNF804A over-expression on genes involved in cell-matrix interactions. These data suggest that ZNF804A regulates the expression of genes implicated in processes underlying schizophrenia pathology, and provide the first evidence that ZNF804A may be involved in the regulation of alternative splicing.