No description available.

No description available.

No description available.

We develop techniques and models for analysing and understanding idiopathic generalised epilepsy, and absence epilepsy in particular. We present techniques for analysing the electroencephalogram of patients with idiopathic generalised epilepsy, having spike and wave discharges. The techniques may have clinical applications and can help to further our understanding of the physiological mechanisms involved in spike and wave discharges and absence seizures. We introduce a new model based on the hypothesis that spike and wave discharges initiate through a transition in a multistable system. The emphasis of the model is in understanding how network structure could relate to the pathology of epilepsy, which is explored with application to human electroencephalogram data. We present evidence to support the hypothesis that changes in network connectivity are a factor in the pathology of epilepsy.

Aims: The aim of this thesis was firstly, to prospectively examine cognitive function in relation to epilepsy, with event-related potential (ERP) and neuropsychological assessments, over a one year period. Secondly, to investigate whether electrophysiological measures have some predictive value for behavioural outcomes. Methods: A methodology development study investigated optimal stimulus and recording parameters for the mismatch negativity (MMN). A prospective epilepsy study included investigation of a healthy control group, groups of patients with: primary generalised seizures (PGS), focal with secondary generalised seizures (FSG) or focal seizures (FS) and a chronic pain group undergoing anti-epileptic drug (AED) therapy. Assessments were conducted over the course of 12 months: at baseline (session 1), 6-months follow-up (session 2) and 12- months follow-up (session 3). ERP measures included the MMN, P3b and P3a. Neuropsychology measures were obtained using visual reaction time, spatial working memory, planning, word list learning, perseverative function, story recall memory and word list learning tests. Cross-sectional analyses of data obtained at sessions 1 and 2 were carried out to investigate group differences at the different time points. Longitudinal analyses were carried out to investigate the effect of time on behavioural and ERP changes in relation to the different groups. Finally, the predictive value of baseline ERPs for behavioural outcomes at one year follow-up was explored. Results: The methodology development study demonstrated that a reading task with duration increment tones yielded the most replicable MMNs in healthy controls. The clinical study demonstrated significant changes in some ERP components obtained from standard oddball, dual target oddball and novel sounds tasks, between controls and seizure groups -some of which were only apparent at session 3. Although the seizure groups showed evidence of significant impairments in aspects of behavioural performance, there was no significant progressive deterioration over time. Moreover, although the pain group showed similar behavioural impairments, in contrast to the seizure groups, they showed significant improvements in some aspects of performance over time. Conclusion: An MMN recording protocol was established for use with the subsequent prospective clinical study. The prospective epilepsy study revealed functional cognitive changes that may be linked to the nature of pathophysiology underlying different seizure types.