Gregory L. Holmes - US grants

One of the most frequent and important questions asked by parents of children with epilepsy is whether seizures can lead to brain damage. This is a difficult clinical problem to study since cognitive impairment and behavioral abnormalities may be related to the etiological agent responsible for the seizures, age at time of onset of seizures, the type, frequency, or duration of the seizures, or the antiepileptic drugs used to treat the seizures. Many of these variables can be eliminated by using animal models of epilepsy. Work in our laboratory using the kainic acid (KA) model has demonstrated that status epilepticus in prepubescent and mature rats leads to significant deficits in memory, learning and behavior as adults when compared to control littermates without seizures. These rats also had a high incidence of spontaneous recurrent seizures (SRS) and an increased susceptibility to seizures using kindling and flurothyl. However, younger animals (<20 day old) with KA-induced seizures of similar severity were not associated with later neurological deficits. The immature animals also had a low rate of SRS and did not differ from controls in susceptibility to kindling or flurothyl. In preliminary results, we have also found that continuous hippocampal stimulation (CHS) in the mature rat brain, but not immature rats, leads to deficits in memory and activity level. Pathophysiological mechanisms that "protect" the young brain from long-term detrimental effects of prolonged seizures are unknown. Since the observation that prolonged seizures in the immature brain have no-long term consequences may have substantial clinical consequences, we wish to expand these preliminary observations. To verify that age at the time of KA administration is a critical factor in the later development of neurological deficits we will subject KA-treated rats to a variety of behavioral tests at a fixed interval, as opposed to a fixed age, following the KA. The role of spontaneous recurrent seizures on subsequent learning, memory, behavior, and seizure susceptibility will be studied by correlating seizure frequency, using continuous video monitoring, with behavioral testing. To determine if the age-dependent changes are due to a greater recovery or plasticity in the immature brain, we will perform serial behavioral and histological examinations following administration of KA. To determine if the immature brain is "more resistant" to excitatory amino acids presumably released during seizures we will assess the behavioral and histological sequelae following intracerebroventricular administration of glutamate and aspartate. In parallel studies, both competitive and noncompetitive intracerebroventricular NMDA antagonists will be given following KA to determine whether damage which occurs following KA can be altered. Finally, to determine whether our findings that long-term sequelae following prolonged seizures is an age-dependent phenomenon is applicable to other models, we will expand our preliminary studies with CHS. Inferences reached in this study should increase our understanding of the importance of age of onset and the mechanisms for such age-specific differences in neurological sequelae following seizures.

DESCRIPTION (Investigator's Abstract): Epilepsy is one of the most common, yet frightening, neurological disorders encountered in childhood. Of major concern is the cognitive and behavior deterioration seen in some children, particularly after prolonged seizures. There is now a considerable amount of data implicating excessive release of excitatory amino acids (EAA's), principally glutamate, as a cause for neuronal toxicity and a major effort is now being made to develop pharmacological agents that block EAA receptors. However, EAAs arc in learning, memory, and brain plasticity, raising concern about using these agents in the developing animal. Work from our laboratory has demonstrated that the long-term effects of prolonged seizures are age-dependent. Using kainic acid, pilocarpine, and continuous hippocampal stimulation we have demonstrated that prolonged seizures in rats below 20 days of age have substantially fewer behavioral abnormalities and no alteration in seizure susceptibility when compared to older animals. Furthermore, histological lesions can not be demonstrated in young animals despite seizures that are more severe than in older animals. If the immature brain is resistant to the long- term effects of prolonged seizures the question of whether EAA blockers should be administered is even more pertinent. To expand earlier observations we will determine if prolonged seizures in the immature brain are associated with physiological or morphological changes. Following induction of seizures using the pilocarpine or perforant pathway stimulation models in immature and mature rats we will examine hippocampal inhibition using evoked potentials and possible cell loss using cell counting techniques. To examine the physiological basis for this remarkable age-related effect, we will use microdialysis to measure glutamate and GABA release following during perforant pathway stimulation or pilocarpine administration. To examine the long-term effects of EAA receptor antagonists on the developing brain, we will administer NMDA and non- NMDA receptor antagonists to rats with prolonged seizures and age- matched controls. Outcome measures will include learning and memory as measured by the Morris water maze, seizure threshold using flurothyl inhalation, long term potentiation (LTP), a physiological model of learning in vivo, and histological analysis. Results of this study should provide additional insights into our understanding of the long- term consequences of seizures on the developing brain.

The incidence of seizures is highest in the first decade of life and status epilepticus is more common in children than adults. Animal studies have also demonstrated that the immature brain is more prone to seizures than the adult brain, presumably secondary to a developmental imbalance between maturation of excitatory and inhibitory circuits. In addition to the increased risk for epilepsy in children, there are some suggestions that seizures during early development may be more detrimental than those occurring in the mature brain. Onset of non-febrile seizures during the first years of life are significant predictors of poor neurological outcome. The poor prognosis following seizures in young children may be related to the seizures themselves, the underlying etiology of the seizures, antiepileptic drug treatment, or all three. Despite the seriousness of the problem, there have been few studies assessing the benefit and safety of antiepileptic drugs (AEDs) in this age group. This deficiency was recognized by a consensus conference in a recent NIH workshop on AED drugs in children. In this planing grant application we are seeking support for one year to design a randomized clinical trial of AED therapy in children with epilepsy at risk for adverse developmental outcomes and intractable epilepsy. Children between the ages of 1 month to 3 years with frequent seizures and abnormal EEGs will be randomized to one of four AEDs. The treatment goal will be seizure freedom or a greater than 80 percent reduction in seizure frequency from baseline. Children not meeting this treatment goal will be changed to a second study drug. In addition to comparing efficacy and safety of AEDs in young children, we wish to examine those factors that predict both poor developmental outcomes and intractable epilepsy. We are hopeful that this study will provide a template for future studies evaluating therapy in severe childhood epilepsy.

[unreadable] DESCRIPTION (provided by applicant): Newborns are at high risk for seizures and seizure-associated brain injury. Between 20-40 percent of term infants who have seizures are subsequently handicapped and this increases to almost 90 percent in pre-term babies. Despite the clinical importance of this problem, little is known about the pathophysiological basis of seizure-induced brain damage in newborns. The goals of this proposal are to delineate the neurobiological consequences of neonatal seizures. [unreadable] [unreadable] We have demonstrated that neonatal seizures, while not causing cell loss, result in changes in synaptic organization with sprouting of mossy fibers in the CA3 and supragranular regions, reductions in neurogenesis, and impairment in spatial memory. We hypothesize that neonatal seizures perturb the development of the hippocampal network by disrupting normal excitatory and inhibitory synapse development resulting in permanent alterations in hippocampal neuronal circuits and hippocampal network patterned activities. Furthermore, these alterations in hippocampal network properties result in functional impairment. To assess this hypothesis, we will employ ex vivo and in vivo electrophysiological techniques and behavioral studies in three specific aims. In the first aim we hypothesize that seizures in the neonatal period result in alterations in neuronal circuitry with subsequent decreases of place firing field precision and stability and abnormal hippocampal rhythms and patterns. Further, we hypothesize that these abnormalities will be associated with abnormalities in visual-spatial memory. We will assess place cell firing patterns and hippocampal network activity in animals with and without a history of neonatal seizures and compare these findings to performance in the water maze. We also hypothesize that neonatal seizures result in alterations in hippocampal patterned activity. Our second aim complements the first aim by determining the persistence of changes in network activity after neonatal seizures by monitoring hippocampal rhythms and patterns throughout development. Our third specific aim is to determine the effects of neonatal seizures on developmental changes in synaptic connections that can underlie the alterations in ontogenesis of hippocampal network activity described in aim 2 and the long-term effects of neonatal seizures on cognition and place cell physiology described in aim 1. These integrated aims will provide insight into the mechanisms of seizure-induced injury in newborns and form the basis for subsequent therapeutic interventions.

DESCRIPTION (provided by applicant): A serious consequence of status epilepticus (SE) is memory impairment. Yet, remarkably little is known about the pathophysiological mechanisms responsible for this cognitive deficit. Exposing rats to status epilepticus during early adolescence causes impairment of spatial memory in the water maze and, in parallel, marked abnormalities in both the positional firing patterns of place cells and the stability of such patterns. These findings corroborate recent work demonstrating strong relationships between the ongoing activity of place cells and the ability of rats to perform adequately in spatial tasks. Thus, place cells may serve as a cell-level indicator of spatial memory, a very high level cognitive function of the rat hippocampus. A major goal of the studies proposed in this grant application is to rigorously test the idea that place cell abnormalities are markers of spatial memory impairment seen after induction of experimental epilepsy in rats. Our first aim is to determine the timing of place cell abnormalities following SE in adult rats and to compare this time course with the timing of changes in spatial memory and cerebral excitability. In this aim we will compare the time course of seizure-induced changes in hippocampal place cell function (firing rates, precision, and stability) with changes in cognitive function and hippocampal excitability. A critical question asked following SE is whether recurrent seizures or interictal epileptiform discharges (spikes and sharp waves) further harm cognitive function. Place cell studies provide us with the opportunity to examine the effects of seizures and spikes on cognitive function at the single cell level. Using mechanistically-oriented hypotheses we will determine whether spontaneous recurrent seizures and interictal epileptiform activity following SE alters place cell function. A key step in understanding the mechanisms of seizure-induced cognitive deficits is to determine whether place cell and visual-spatial memory impairment extend to other seizure insults. Our third aim is to assess the effects of serial seizures, which result in impaired water maze performance but have a distinctly different pathological pattern than SE, on place cell function. Taken together these studies will provide insight into the mechanisms of seizure-induced cognitive dysfunction which will be critical as we attempt to develop novel therapeutic interventions.

DESCRIPTION (provided by applicant): This revised application requests support for a training program in translational neuroscience research. Translational research is the process of applying ideas, insights, and discoveries generated through basic scientific inquiry to the treatment or prevention of human disease. The major goal of the program is to attract and train talented and motivated individuals interested in learning basic neuroscience techniques that will allow them to conduct research that will have a major impact on neurological disease prevention, modification, or cure. The program will support training post-doctoral fellows (M.D., M.D./Ph.D. or D.O) in a broadly based interdisciplinary research training program. The ideal candidate will be one who has received clinical training in neurology, neurosurgery, or psychiatry and wishes to devote two years to intense training in translational neuroscience research. The training program will occur within the Neuroscience Center at Dartmouth (NCD), a new and unique interdisciplinary group whose mission is to foster collaborative and interactive research in education in the neurosciences. It is the vision of the NCD faculty to produce and disseminate new knowledge and, in doing so, train the next generation of neuroscientists. Post-doctoral fellows will be engaged in research under the tutelage of a faculty sponsor and will have their research experience broadened by regular interactions with other faculty and fellows through conferences, seminars, and courses in neuroscience organized by the NCD and Dartmouth Medical School. In addition, the program faculty and trainees interact closely with faculty and students in other NIH- sponsored training programs at Dartmouth. While this is a resubmission of a new application for a newly created neuroscience center, individual members of the NCD participating in this program all have strong records in research, teaching, and mentoring of pre-doctoral and post-doctoral students. The close relationship between the clinical and basic science faculty, the vigor of the faculty, and the academic environment at Dartmouth makes it highly likely this will be a successful program.

DESCRIPTION (provided by applicant): There is increasing evidence that interictal EEG abnormalities can result in transitory cognitive impairment. In severe cases, where interictal spikes (IS) dominate the EEG recording, cognitive regression can occur. The mechanisms by which IS result in cognitive impairment are not known. In preliminary studies we have shown that following status epilepticus (SE) or recurrent brief flurothyl-induced seizures IS are common. Following SE or recurrent seizures, animals with IS show more impairment than rats without IS. In preliminary work we found that IS result in a substantially reduced likelihood of action potentials (AP) firing in the hippocampal cells that fire in a particular location of the environment, the so-called place cells. Such reduction in AP firing lasts for up to two seconds following the IS. We also have evidence that IS interfere with the replay of place cell firing, normally observed during the awake state and during slow wave sleep consolidation of memory. Our overall hypothesis is that IS, by interfering with firing patterns of critical cells at critical times, result in cognitive impairment. In Specific Aim One we will address the hypothesis that IS interfere with memory, and particularly working memory, by preventing accurate encoding and retention of information. We will first determine location of IS by performing current source density analyses using multi-contact depth electrodes in the hippocampus of epileptic rats. To rigorously address the hypothesis that IS interfere with encoding and retention of memory we will take advantage of the place preference test and the delayed non-matching to sample test, tasks which allow us to record place cells while the animal is engaged in a test of spatial memory. By comparing spatial performance with place cell firing during periods with and without IS we will directly determine the effect of IS on spatial learning and memory. The observation that IS are highly coupled to arousal and EEG state will allow us to control IS frequency by manipulating non-theta activity. In Specific Aim Two we will address the hypothesis that IS during sleep impair cognitive function through disruption of consolidation of memory during replay of place cell firing patterns. In this aim, we will study the effect of nocturnal spikes on recapitulation of place cells firing patterns during sleep using animals previously subjected to status epilepticus or recurrent flurothyl seizures and IS during the awake and sleep states. Project Narrative: If our hypotheses are confirmed the findings would have major clinical implication in regards to the treatment of patients with epileptic encephalopathies. In addition to learning much about the cognitive impairment in epilepsy, our studies will provide enormous insight into processes involved in normal encoding and consolidation of memory.

[unreadable] DESCRIPTION (provided by applicant): The neonatal period eclipses all other epochs of the human life span for the highest incidence of seizures. Neonatal seizures are associated with a high likelihood of adverse neurological outcomes including mental retardation, behavioral disorders, and epilepsy. However, it remains unclear how much of the seizure- related damage is secondary to the etiology of the seizures or the seizures per se. In the prior funding period (2002-2006) we defined the critical role of GABA in seizure genesis in neonates, demonstrating that in the developing hippocampus the GABA effect is complex, showing both excitatory and inhibitory properties. Neonatal seizures were also found to result in long-standing cognitive impairment. While this work showed the hippocampus has an important role in seizure genesis and seizure-related impairment, clinical studies have shown that most children with severe epilepsy have neocortical epilepsy. Remarkably little is known about whether GABA also has a dual role in the developing neocortex, and if so, how this affects normal brain development and seizure susceptibility. The long-term effect of neonatal seizures on neocortical function is also not known. The overall goal of the study is to better understand the role of GABA signaling in normal and abnormal brain development, starting at the cellular level and progressing into behavioral studies. This continuation proposal will revolve around the role of GABA signaling in the neocortex in normal early development as well as during seizures The first specific aim is to study developmental changes in GABA signaling in the rat neocortex. In this aim we will determine the actual developmental changes in GABA signaling in the neocortex during development including the changes in GABA(A) reversal potential (EGABA), GABA(A) driving force (DFGABA, that is the difference between EGABA and membrane potential (Em), action potential threshold, and the excitatory-to-inhibitory switch in the action of GABA. Specific aim two will study the role of depolarizing GABA in the generation of physiological and paroxysmal activities in the immature neocortex. We will take advantage of novel recording techniques developed during the prior funding period to examine GABAergic induced plasticity in the developing neocortex. The third specific aim will evaluate the consequences of neonatal seizures on neocortical function using a combination of powerful behavioral and in vivo electrophysiological techniques. Specifically, we will assess the long-term effect of recurrent neonatal seizures on neocortical synaptic transmission and prefrontal cortex function. PUBLIC HEALTH RELEVANCE This study should provide valuable insights into the mechanisms by which seizures occur in the developing brain and the consequences of such seizures on brain development. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Childhood epilepsy is associated with a significant risk for cognitive impairment. Understanding the mechanisms of these cognitive impairments may lead to novel therapies that minimize the adverse outcomes. We have previously shown that in rats, early-life seizures (ELS) result in substantial cognitive impairment. Preliminary data from our laboratory suggest that ELS are followed by abnormalities in neural coding and EEG oscillatory activity, both of which are believed to play a critical role in cognitive function. This project further investigates the mechanisms leading to cognitive impairments by determining the extent and relevance of brain oscillation dysfunction. We hypothesize that: i) ELS cause deficits in brain oscillatory activity; and ii) These deficits are responsible, at least in part for cognitive impairment. To reach our goal of determining the impact of ELS on oscillatory activity and its relationship to cognitive function in the developing and adult brain we will first characterize how ELS affect the developmental trajectory of oscillatory activity in hippocampus through use of multi-contact laminar electrodes in the hippocampus in non-anesthetized rat pups starting at postnatal day (P) 5. We will then analyze the relationships between oscillations and behavior in rats with and without ELS, both at the single unit and network levels. Place cell development and theta and gamma modulation of place cell firing (phase modulation, phase precession) will be studied in freely moving rats at P20 and P60 in a delayed spatial alternation task, and place cell properties will be compared between correct and incorrect trials in the same animals and between controls and ELS rats. To evaluate the effects of ELS on network properties, we will study spectral coherence on hippocampal (CA1-CA3) and hippocampal-prefrontal network function while rats are involved in a memory task. Having established the role of hippocampal oscillations in ELS-induced cognitive dysfunction in Specific Aim 1, we will then explore ways to improve cognition following ELS in Specific Aim 2. Specifically, we will evaluate the effect of artificially-induced theta and gamma oscillation on hippocampal function following ELS and whether behavioral training animals can modify hippocampal oscillations and subsequent learning. Taken together, these studies are likely to provide insights into the mechanisms and consequences behind network abnormalities of ELS and provide a strong framework for intervention.

DESCRIPTION (provided by applicant): Temporal lobe epilepsy (TLE) is the most common epilepsy in humans undergoing epilepsy surgery. Many of these patients have associated cognitive impairments, particularly in the domain of memory which is an essential, higher cognitive function required for continuity in time, personal history and awareness. The cognitive impairments are no doubt a result of a combination of the brain abnormalities responsible for the disorder, although the mechanisms of this relation remain a mystery. Our laboratory has recently pursued the underlying mechanisms responsible for dynamic cognitive dysfunction in the lithium-pilocarpine model of TLE finding that interictal spikes (IIS), dysfunctions in network oscillation activity (EEG rhythms) and uncoordinated single-neuron firing impact the performance of the animal in a hippocampal- dependent task. We now propose to investigate whether similar mechanisms are at play in humans. These data will ultimately be important for informing therapeutic strategies designed to minimize the cognitive effects of TLE, as well as a better understanding of the pathophysiology of epilepsy on a dynamic level. Three integrated specific aims are proposed. In the first specific aim we will determine the transient effects of IIS on memory, reaction time and related hippocampal oscillations during a hippocampal-dependent memory task in humans. We hypothesize that IIS contribute to memory impairment in patients with TLE through direct impairment of neural processes underlying memory and by disruption of hippocampal oscillations. In the second aim we will investigate the relationships between single unit neural activity and hippocampal oscillations and performance in patients with TLE. Owing to extensive animal work, we hypothesize that deficits in spatial cognition and memory in patients with TLE are due to impaired single unit firing and temporal coding of action potentials. Based on animal data showing impaired coherence in rats with a prior history of seizures, in the third aim we will determine the impact of TLE and related IIS on neuronal network oscillation and cognition, hypothesizing that deficits in working memory in patients with TLE are due to impaired coherence in hippocampal-prefrontal cortex pathways and additional transient coherence disturbances due to IIS. Taken together these three aims will provide considerable insight into the mechanisms of cognitive impairment in TLE. Armed with this information, therapeutic strategies with a scientific basis will be used to treat this devastating comorbidity.

DESCRIPTION (provided by applicant): Febrile seizures are the most common type of seizure seen in young children. Unfortunately, some children with prolonged febrile seizures appear to be at risk for long-term cognitive disturbances. Identifying those individuals at risk for cognitive impairment and discovering the responsible mechanisms would provide opportunities for therapeutic intervention. In preliminary studies through an R21 funding mechanism, we used an immature rat model of long experimental febrile seizures (EFS) and established that a subgroup of rats experiencing these seizures developed spatial memory impairment and aberrant place cell function during adulthood. These deficits in hippocampal-dependent spatial cognition were accompanied by elevated MRI T2 signals in the hippocampus one month after the seizures. These findings demonstrate for the first time a direct causal effect of EFS on function of specific neurons that govern memory performance suggesting that MRI might be a potentially predictive biomarker for individuals at risk for cognitive disturbances. Furthermore, we have found that MRI T2 signals in hippocampus one month after the seizures are associated with inflammatory activation but no overt cell death indicating that inflammatory mediators might contribute to both MRI abnormalities and neuronal dysfunction, providing a target for selective intervention. However, prior to the application of these findings to the management of children with FSE additional critical information is needed. It is not known if hippocampal MRI changes take place early enough to be predictive of cognitive defects, and thus useful for potential intervention or whether the observed cellular and cognitive impairments are result of the FSE or of ensuing epileptogenesis. Determining if hippocampal levels of the inflammatory cytokine interleukin (IL)-12 distinguish FSE rats that became epileptic from those who did not and if this cytokine is involved in the cognitive defects provoked by EFS is necessary. In this proposal we will assess whether the cognitive defects in a subset of rats experiencing FSE are predicted by selective MRI changes that are visible early after the seizures and whether the cognitive and place-cell defects after FSE emerge prior to, and independent from, the epileptogenic process and spontaneous seizures, and define the underlying mechanisms of such deficits. To ascertain whether inflammatory cytokine expression distinguishes rats with cognitive defects after FSE from those with intact hippocampus-dependent memory we will compare rats with and without cytokine changes after EFS. The results of this study will set the stage for therapeutic intervention in children at risk for cognitive problems following febrile seizures. PUBLIC HEALTH RELEVANCE: By far, the most common type of seizure seen in young children occurs as a result of fever. While the outcome in most children with febrile seizures is favorable, a significant number have cognitive deficits following prolonged febrile seizures. Identifying children at risk for cognitive deficits following febrile seizures is important since this will provide opportunities for therapeutic intervention. In this study we will determine if MRI findings after febrile seizures are predictive of cognitive impairment and also study whether inflammation following febrile seizures is responsible for both the MRI changes and cognitive deficits.

DESCRIPTION (provided by applicant): Febrile seizures are the most common type of seizure seen in young children. Unfortunately, some children with prolonged febrile seizures appear to be at risk for long-term cognitive disturbances. Identifying those individuals at risk for cognitive impairment and discovering the responsible mechanisms would provide opportunities for therapeutic intervention. In preliminary studies through an R21 funding mechanism, we used an immature rat model of long experimental febrile seizures (EFS) and established that a subgroup of rats experiencing these seizures developed spatial memory impairment and aberrant place cell function during adulthood. These deficits in hippocampal-dependent spatial cognition were accompanied by elevated MRI T2 signals in the hippocampus one month after the seizures. These findings demonstrate for the first time a direct causal effect of EFS on function of specific neurons that govern memory performance suggesting that MRI might be a potentially predictive biomarker for individuals at risk for cognitive disturbances. Furthermore, we have found that MRI T2 signals in hippocampus one month after the seizures are associated with inflammatory activation but no overt cell death indicating that inflammatory mediators might contribute to both MRI abnormalities and neuronal dysfunction, providing a target for selective intervention. However, prior to the application of these findings to the management of children with FSE additional critical information is needed. It is not known if hippocampal MRI changes take place early enough to be predictive of cognitive defects, and thus useful for potential intervention or whether the observed cellular and cognitive impairments are result of the FSE or of ensuing epileptogenesis. Determining if hippocampal levels of the inflammatory cytokine interleukin (IL)-12 distinguish FSE rats that became epileptic from those who did not and if this cytokine is involved in the cognitive defects provoked by EFS is necessary. In this proposal we will assess whether the cognitive defects in a subset of rats experiencing FSE are predicted by selective MRI changes that are visible early after the seizures and whether the cognitive and place-cell defects after FSE emerge prior to, and independent from, the epileptogenic process and spontaneous seizures, and define the underlying mechanisms of such deficits. To ascertain whether inflammatory cytokine expression distinguishes rats with cognitive defects after FSE from those with intact hippocampus-dependent memory we will compare rats with and without cytokine changes after EFS. The results of this study will set the stage for therapeutic intervention in children at risk for cognitive problems following febrile seizures.

PROJECT SUMMARY Autism, a neurodevelopmental disorder characterized by impairment in communication, social impairment and repetitive and stereotyped behaviors, is a life-long condition with an undetermined etiology and currently is not curable. While autism has a heterogeneous and complex genetic underpinning there is evidence that environmental factors also play a role. Despite the variety of factors that can lead to autism, the phenotype is remarkable uniform, raising the possibility of a ?final common pathway? in this disorder. Functional MRI and electrophysiology studies suggest this ?final common pathway? may be through aberrant neural connectivity during development. In this proposal we wish to evaluate previously recorded EEGs obtained from a large cohort of children with autism, developmental delay and neurotypically developing children extensively evaluated at the NIMH, to determine if there are neural networks characteristics that distinguish children with classic autism from typically developing children and children with developmental disabilities without autism. Ascertaining such changes may provide insight into the pathophysiological mechanisms responsible for the symptoms in autism. In specific aim 1 we will determine whether children with classic autism have neural networks which distinguish them from typically developing children and children with neurodevelopmental disorders but without autism. We will examine awake and sleep recordings for coherence and develop functional connectivity maps using Pearson correlations and partial correlations in all three groups. In specific aim 2 we will determine whether coherence and connectivity maps change over time in children with autism and whether such changes correlate with outcome and in specific aim 3 we will determine whether epileptiform activity is more common in the EEGs of the children with autism than in typically developing children and children with neurodevelopmental disorders but without autism and whether such activity is related to outcome. By taking advantage of this rich data set we wish to better characterize neural connectivity in autism using powerful electrophysiological techniques. Understanding how the brain of a child with autism is ?wired? will play an important role in developing therapeutic interventions.