Search

This thesis aimed to evaluate whether EEG is a sensitive measure of reward dysfunction in alcohol dependency using a monetary incentive delay task. fMRI monetary incentive delay task studies have shown significant differences exist between alcohol dependent and healthy control participants, with alcohol dependent participants showing blunted ventral striatum activation during the anticipation of monetary rewards (Wrase et al., 2007; Beck et al., 2009) and an elevated response in a distributed reward network during reward outcome (Bjork et al., 2008; Luijten et al., 2017). To date, an EEG monetary incentive delay task has not been used to assess reward disturbances in alcohol dependency or other addictive disorders. This thesis provides new insights into this topic, describing three EEG monetary incentive delay experiments in which data were analysed using both event-related potential and multivariate single-trial machine learning methods. Experiment 1 found disrupted valence (monetary gain versus loss cues) sensitivity during reward anticipation within a group of young adult at-risk drinkers. The machine learning analyses demonstrated that low-risk drinkers have significant valence discrimination within the cue-P3 time window, whereas at-risk drinkers were insensitive to valence. Experiment 2 uncovered reduced cue-P3 sensitivity to gain and loss anticipation in alcohol dependent compared to healthy control participants. The event-related potential analyses showed enhanced cue-P3 amplitudes for both monetary gain and monetary loss compared to neutral cues within healthy controls, but this effect was not observed in alcohol dependents. The machine learning analyses demonstrated healthy controls had enhanced loss anticipation discrimination (monetary loss versus neutral cues) across the cue-P3 time window. Experiment 3 found no evidence of disrupted reward processing in alcohol dependent compared to healthy control participants during the outcome phase of the task. Both groups showed comparable results for both forms of analyses, with modulation by reward predication error salience and valence within the FRN and the feedback-P3 time windows, respectively. This thesis found blunted neural processing within the cue-P3 time window during reward anticipation in a group of at-risk drinkers and alcohol dependent participants, compared to healthy controls. The experiments in this thesis uncovered potential EEG components which may be utilised during future psychopharmacological research for the evaluation of treatments for alcohol dependency.

Prader-Willi syndrome (PWS) is a neurodevelopmental disorder resulting from the absent expression of maternally imprinted, paternally expressed genes located in the chromosomal region 15q11-13. This absence of expression is usually either due to a paternal deletion or maternal uniparental disomy (mUPD). Both genetic subtypes share the same core clinical symptoms, such as developmental delays, intellectual disability, hyperphagia, behavioural and social difficulties, and a heightened risk for anxiety and depression. However, only people with PWS due to mUPD seem to be at particularly high risk for developing psychosis. Research on the causes of this differential risk is very scarce. This PhD study aims to propose potential mechanisms of psychosis in PWS and design and develop a study to test them. I hypothesised that the GABA/glutamate equilibrium is disrupted in PWS due to the absence of paternal expression. In addition, the overexpression of maternal genes on chromosome 15 would cause cholinergic dysfunction in people with mUPD and further disrupt GABAergic, glutamatergic, dopaminergic and cholinergic functions, increasing the probability of psychotic symptoms. In line with the predictive processing model of psychosis, the neurotransmitter disruptions proposed would cause a reduction in the use of priors to formulate predictions, detect and generate prediction errors, and update the prediction system. A case-control study comprising clinical, cognitive, neuroimaging, and remote behavioural testing was designed to investigate the hypotheses. The study was organised into four modules. The EEG module comprised the Global-Local paradigm, a sensory gating paradigm, and resting state. The MRS module measured metabolites in the anterior cingulate cortex, the auditory and visual cortex. The cognitive module contained measures of IQ, processing speed and working memory. The behavioural module was added during the COVID-19 pandemic to enable online data collection and focused on measuring the use of priors in perception. All modules were assessed for use in a PWS population and were feasible, with adjustments. Nineteen participants took part in at least part of the study, demonstrating its feasibility in a PWS population. Because of time constraints, COVID-19, and resources available, not all participants completed all measures. For the same reasons, only the EEG Global-Local data was analysed. The EEG paradigms were collected on 19 participants (11 with deletion, 4 with mUPD), and the global-local paradigm was analysed. Preliminary analysis revealed that people with mUPD have smaller P300 in response to an oddball paradigm than deletion and control participants. Participants with deletion have smaller P300 responses than controls showing impaired attention switching and a flaw in the predicting coding system of people with PWS, particularly in those with mUPD. Contrary to what was expected, the mismatch negativity response to the oddball paradigm was not different between mUPD and controls but was significantly earlier in the deletion group. This could indicate a chronic fight or flight state having the potential to disrupt neurotransmission, interoception and perception further, causing additional disruption of the predictive processing system. A planned analysis of the remaining measures is presented in this thesis and will be applied in the future when sufficient data is analysed.

Background: Cognitive impairment is prevalent in Parkinson's disease (PD), with 50% of patients developing dementia within 10 years (Williams-Gray et al., 2013; Aarsland and Kurz, 2010). The presence of mild cognitive impairment (MCI) has previously been associated with increased dementia risk (Pederson et al. 2013). Electroencephalography (EEG) is an inexpensive and non-invasive tool used to assess cognition in aging clinical groups. EEG has previously been shown to have prognostic value in MCI associated with Alzheimer's disease (AD-MCI), yet there is little information on the electrophysiological correlates of PD-MCI. In this study, the feasibility of using a 'dry-EEG' mobile headset to assess cognitive impairment in Parkinson's disease (PD) was investigated. Dry-EEG headsets offer many benefits as the set-up time is significantly shortened, the equipment is cheaper and the headset is portable, which should improve patient access to neuroimaging technologies as called for by a World Health Organisation report (WHO, 2017). Aim: To determine if dry-EEG can adequately record well-known EEG components and to evaluate if these recordings enhance PD-MCI patient classification. Method: 47 PD patients and 37 age and sex-matched controls were recruited to complete a battery of neuropsychological and EEG tasks. PD-MCI status was assigned based on the Movement Disorders Society PD-MCI level I criteria. A comprehensive battery of neuropsychological tests and EEG tasks were conducted. Resting state EEG was also recorded. A mix of both event-related potential (ERP) component analysis and time-frequency spectral analysis was conducted. Differences between groups were assessed using randomisation tests with FDR-corrected pairwise comparisons. Machine learning models were employed to investigate if the addition of EEG recordings improved PD-MCI classification accuracy. Results: A significant proportion of PD participants were classified as PD-MCI according to the Movement Disorders Society PD-MCI criteria. The majority of PD-MCI patients were impaired in at least three cognitive domains, demonstrating that cognitive impairment is pervasive in this PD cohort. Significant differences were found between PD participants and controls on a range of EEG tasks. Machine learning models provided interesting insights on what features were most discriminative for participant classification. Conclusion: The findings of this study show that cognitive impairment is prevalent in PD. The addition of EEG did not aid differential discrimination between cognitively normal and cognitively impaired PD patients at an early stage. The most discriminative features selected by machine learning models are in agreement with the dual syndrome hypothesis of cognition (Kehagia et al., 2013) and those identified in a 10 year follow-up study (Williams-Grey et al., 2013). This suggests that it may not be feasible to use dry-EEG to profile cognition in PD. Longitudinal study of this cohort would investigate the ability of these EEG recordings to predict change to PD cognitive status over time. These findings could also inform the development of a larger trial which uses EEG as an outcome measure to evaluate the success of an intervention, be it pharmacological, cognitive training or exercise-based to improve cognitive outcomes in PD-MCI.an intervention, be it pharmacological, cognitive training or exercise-based to improve cognitive outcomes in PD-MCI.

Studying brain development poses significant methodological challenges when working with vulnerable newborn infant populations. Non-invasive and wearable optical brain imaging technologies may allow for the cot-side study of newborn infants in clinical settings. This has particularly important implications for preterm infants in intensive care, for whom traditional neuroimaging techniques are not readily feasible. Wearable high-density diffuse optical tomography (HD-DOT) has emerged as a promising technology in this pursuit. In this thesis, I describe the development and application of wearable HD-DOT for newborn infants in clinical settings. Specifically, this technology is applied to study the features of functional brain connectivity (FC) associated with neonatal sleep states, active sleep and quiet sleep, in both term-born and preterm infants. HD-DOT is also electromechanically combined with electroencephalography (EEG) to concurrently study haemodynamic cortical and electrical cortical brain activity in the preterm infant. As a result of the constraints posed by lock-down conditions and a pause on clinical studies during the COVID-19 pandemic, the HD- DOT device is also used to study resting state functional connectivity (RSFC) in the adult brain in the home setting. Findings from this work demonstrate the potential of wearable HD-DOT, and combined EEG-HD-DOT, for the cot-side study of cerebral function during sleep in newborn infants, as well as at-home neuroimaging of resting-state networks in adults. This work was supervised by: Dr. Topun Austin, Department of Paediatrics, University of Cambridge, and Dr. Robert J. Cooper, Department of Medical Physics and Biomedical Engineering, UCL.

Bilingualism has been shown to modulate the neural mechanisms of selective attention, with differences between monolinguals and bilinguals observed even when they display equivalent behavioural performance in a selective attention task. This suggests that the crucial consequence of learning and using multiple languages might be that it triggers neuroplastic adaptation that allows bilinguals to achieve optimal performance under increased processing demands. This functional plasticity yielding equivalent outcomes (also known as degeneracy) is a common feature in biological systems, allowing flexible adaptation to changing environments. Yet the exact mechanism by which bilingualism affects selective attention is still not entirely clear. While the currently dominant view suggests that the need for constant management of competing languages in bilinguals increases attentional capacity; another possibility is that this language control may be drawing on the available attentional resources such that they need to be economised to support optimal task performance. Another question concerns the development of this adaptation over time, where the demands of competition and inhibition between co-activated languages might be reconfiguring the patterns of attentional processes right from the onset, such that the effects can be seen by the time children can respond to selective attention tasks. Alternatively, these modifications might have a protracted maturation dependent on the length and intensity of exposure to the demands of bilingualism, in which case they would manifest differently in adults and in children, as well as in speakers with different levels of exposure to L2. Finally, another aspect is to establish the extent to which these modifications might affect attentional processing beyond the language domain, extending to auditory processing more generally. Here I present a series of behavioural and neuroimaging experiments that address these questions. To investigate whether bilingualism enhances attentional processing or triggers redistribution of the existing capacity, I used EEG to track the neural encoding of attended continuous speech in monolingual and bilingual children aged 7-12, in the context of different types of acoustic and linguistic interference. Participants attended to a narrative in English while four different types of interference were presented to the unattended ear. The neural encoding of attended and unattended streams was assessed by reconstructing their speech envelopes from the EEG data in each condition, using the mTRF toolbox. Results showed more accurate reconstruction of the attended envelopes than ignored ones across all conditions for both bilinguals and monolinguals. Critically however, there was no evidence of enhanced attentional processing in bilinguals; instead data showed a pattern consistent with redistribution of the available capacity, economised to achieve optimal performance on the selective attention task. The follow up behavioural experiments tested the limits of this adaptation by using a dual task (dichotic listening + visual attention) to further increase processing load. The results over three experiments (on children, and adults with different levels of exposure to L2) showed consistently comparable performance on both tasks for monolingual and bilingual adults, suggesting that bilingual adaptation can accommodate high processing loads. However there were also subtle differences in performance on the secondary (visual) task between the monolingual and bilingual children, and across the two groups of bilingual adults, suggesting that maturation and exposure do exert influence on this functional adaptation. The findings of the final EEG study on auditory processing beyond language domain indicate comparable but attenuated modification of attentional processing in bilinguals, compared to the first EEG study using linguistic interference. Findings from all experiments are explored in the context of theories of selective attention and bilingualism.

Rapid drinking of excessive quantities of alcohol disrupts normal memory functioning and can lead to the experience of an alcohol-induced memory blackout (MBO), meaning that events which occurred while intoxicated may not be recalled once sober. Although instances of extreme binge-drinking and MBOs are prevalent in student populations, there is a paucity of knowledge surrounding the topic. Critically, 1) we do not know what the wider influences on these behaviours may be, specifically for students studying in Scotland, and 2) we also do not know whether these events leave a lasting impact on memory formation and, if so, whether this deficit is temporary. This thesis addressed these issues firstly with a questionnaire which investigated student drinking behaviours, and then with a series of laboratory-based memory studies which participants carried out sober, after a scaled dose of alcohol, and within 20-hours of experiencing a blackout. We found that students in Scotland frequently binge-drink, with a high prevalence of MBOs, influenced by home country, year of study, and possibly by Scottish culture. We also found control and experimental participant groups performed recall and recognition memory tasks with similar behavioural accuracy while sober and that performance dropped but did not differ following alcohol. However, ERP evidence suggested a shift in neural strategy in MBO participants compared to controls. Further, after-blackout performance remained impaired in more cognitively demanding tasks. In sum, evidence suggests that our Scottish-based student population drink to excess, with a large proportion experiencing regular MBOs. Alcohol impairs behavioural memory performance for all, but underlying differences in memory strategy, and a lasting deficit following MBO, can be seen in those who frequently blackout. These findings highlight the importance of further investigating the trajectory of blackout experiences, and any damage they leave in their wake.

Electroencephalography (EEG) is a neurophysiological tool that can be used to study the neural correlates of cognitive processes. Using a tool like EEG over traditional pen and paper tests of cognition can offer greater objectivity of assessments and might also be more sensitive to conditions in which such cognitive processes are impaired (e.g., memory impairment in Alzheimer's disease; AD). Fast Periodic Visual Stimulation (FPVS) is a method used with EEG to study neural correlates of cognitive processes by virtue of rapid and periodic presentation of stimuli. In an FPVS paradigm a stream of stimuli is presented at a high frequency (e.g., six images per second: 6 Hz) wherein 'oddball' stimuli that are categorically different to the standard stimuli are presented periodically (e.g., every sixth image belongs to a different category). The high signal-to-noise ratio and quick administration time (~ 2 minutes for an FPVS task) of this method makes it a promising candidate for use in clinical practice. Therefore, my PhD research has focussed on developing and refining the FPVS method to measure cognitive processes that are impaired early on in conditions such as AD and semantic dementia as well as improving its implementation by testing the reliability of a mobile EEG version of a FPVS memory assessment tool. The main findings of my PhD were: 1) repetition of stimuli in an FPVS paradigm plays a significant role in strengthening the signal measured but is not the sole driver of the oddball response, 2) a mobile EEG version of this tool can reliably detect the FPVS memory signal, 3) implicit semantic discrimination can be measured using FPVS EEG, in the absence of any pre-task encoding, instructions or repetition of stimuli and 4) pre-task encoding of stimuli results in significantly greater FPVS responses compared to repetition priming alone. Overall, the findings of this PhD contribute to the FPVS literature, and the single- vs multiple- systems of memory debate. They also provide supporting evidence for the potential use of FPVS-based assessments to detect impairments in memory in conditions such as AD.

Recognition memory decisions can be influenced by short-term masked priming. The overarching aim of the thesis was to investigate the effects of short-term repetition and conceptual priming on familiarity and recollection, respectively. Previous research has found that repetition priming of test stimuli (chair - CHAIR) increases familiarity regardless of the study status. Similarly, increasing conceptual fluency of test stimuli through associative priming (table - CHAIR) has been found to increase familiarity for both studied and unstudied stimuli. The convergent explanation for this phenomenon is fluency (mis)attribution to memory, with more fluently processed stimuli being judged as studied or familiar. One branch of the thesis investigated the on/off switching of the fluency attribution to memory system and the neural dynamics underlying it. In two experiments (Chapters 2 and 3) we found that fluency is attributed to prior exposure when study and test modalities match (at least partially). Memory- and priming- sensitive event-related potential (ERP) components were found, but these did not appear to capture the fluency attribution process (i.e., there were no group differences). However, using dynamic causal modelling (DCM), we found that group modulated the connectivity strength between right perirhinal cortex and right dorsolateral prefrontal cortex. The second branch of the thesis examined the mechanism underlying conceptual priming effects on recognition memory. Recent studies have reported that non-associative conceptual priming (sofa - CHAIR) increases correct recollection only. Rather than fitting the fluency-attribution-to-memory framework, we propose that these findings are due to encoding context reactivation, conceptual primes acting as retrieval cues for the encoded memory trace of the target word. In a series of experiments, we biased the encoded context of target words and used conceptual primes: (1) related to the target and to the encoded context, (2) related to the target and related to a different context (than the encoded one), and (3) unrelated primes. Conceptual priming effects on recollection were not consistently observed; nonetheless, only primes related to the encoded context increased recollection compared to the other priming conditions, supporting the encoding context reactivation account. In contrast, conceptual primes related to the different context, and sometimes both types of related primes increased familiarity compared to unrelated primes, in line with the fluency attribution account of familiarity. Overall, our findings show distinct mechanisms by which repetition priming and non-associative conceptual priming influence recognition memory judgments.

Benzodiazepines (BZDs), non-selective positive allosteric modulators (PAMs) of GABAA receptors, are highly efficacious across a range of therapeutic indications. However, even at low receptor occupancy (RO), BZDs are associated with significant side effects such as somnolence and are liable to abuse and to the development of tachyphylaxis, all of which limit their clinical utility, particularly in chronic indications such as epilepsy. As many of these undesirable properties are mediated by α1 subunit-containing GABAA receptors, there has been a concerted effort to develop α2/3/5 subtype selective PAMs for chronic treatment of a range of neurological and neuropsychiatric disorders. This thesis highlights the biomarker-enabled approach taken in the research and development program of darigabat, a compound identified as demonstrating the desired functional selectivity for GABAA receptors containing α2/3/5 subunits, with negligible activity at α1 subunits. The wide-ranging biomarker toolkit employed in both preclinical and clinical studies included measures of receptor occupancy, pharmacodynamic activity, and efficacy, and was designed to answer some key research questions. These included demonstrating the translation of the in vitro darigabat profile to the clinical profile; identifying attributes that differentiated GABAA receptor subtype-selective PAMs from BZDs (a critical requirement for the continued development of darigabat); understanding the relationship between darigabat dose, occupancy, and pharmacology; and has provided some understanding of the potential utility of darigabat in a range of CNS disorders. This work, which spans more than 10 years, has not only identified the therapeutic potential of darigabat, which is currently in Phase 2 clinical trials in focal epilepsy and Phase 1 clinical trials in panic, but has furthered our understanding of GABAA receptor pharmacology and the promise these compounds hold for the treatment of serious and disabling disorders.

Individual differences have been related to face recognition ability. For example, empathy has been related to face recognition; however, empathy poses limitations for the development of effective interventions. Empathy is proposed as part of compassion, which relates to self-compassion and mindfulness that may be more effectively applied in face recognition than empathy alone. In Chapter 2, four experiments explored the relationship between empathy, compassion, mindfulness and unfamiliar face identification. Mindfulness and common humanity related to better recognition but not matching. Face identification did not relate to empathy or compassion for others. Chapter 3 explored the factor structure of empathy, compassion and mindfulness measures to further investigate if suggested latent variables predict face recognition performance. Attentional mindfulness notions merged to predict face recognition performance. All self-compassion facets and the mindfulness facet of 'non-react' collectively predicted false alarm rates. Finally, positive empathy and compassion notions did not predict recognition performance. Chapter 4 investigated the neural basis of the relationship between mindfulness and face recognition by examining the face recognition ability and corresponding ERP activity of experienced meditators and non-meditators. Experienced meditators' face recognition ability was not as good as that of non-meditators. Experienced meditators, compared to non-meditators, displayed smaller N1 and larger P300 amplitudes in the study phase, but no group differences were observed in the P100, N170 and FN400 components. Chapter 5 explored the effects of a brief mindfulness intervention on face recognition. Participants, who either trained in meditation or listened to an audiobook were tested, post-intervention, with face recognition memory tasks while their ERP component activity was recorded. Findings suggested better face recognition in the meditation group, who also displayed larger N1 amplitudes and shorter P100 latencies. No further significant ERP amplitude differences between the meditation and control groups were observed. Chapter 5 embedded mindfulness instructions in the face-composite construction process to facilitate identification. Participants memorised an unfamiliar face and, 24 hours later, constructed the face using the EvoFIT facial composite system, either with or without mindfulness instructions. More composites constructed with mindfulness, than the normal EvoFIT process, were correctly identified. Overall, the present findings provide sufficient justification for further research into the role of mindfulness in face recognition. Attentional mindfulness notions may inform new directions in face recognition research, aiming to practically assist the face recognition process.

This thesis aims to investigate the cognitive changes that follow the transition from wakefulness to sleep. All studies presented have been driven by the hypothesis that consciousness is selectively required for some cognitive processes and not others. As such, fluctuations in alertness are hypothesised expected to produce a larger effect in those processes that are more reliant on consciousness. In chapter 2 we begin by investigating the effect of alertness on conflict monitoring. Conflict monitoring is the environmental scanning process by which the brain detects conflicting information and identifies the need for top-down behavioural control. Using a combined Go/NoGo and Go-left/Go-right paradigm, we investigated effects of alertness on reaction times and responsiveness. We revealed an interaction between the task requirement to monitor conflict (Go/NoGo) and the level of alertness. Crucially, the difference between both tasks was only observed in the later stage of drowsiness. This suggests that conflict monitoring is more susceptible to reductions in alertness. Our results from Chapter 2 also provide further evidence for the ecological validity of the Hori scale as a way to classify of the sleep onset period into functionally meaningful sub-stages. However, manual scoring of trials is an arduous process subject to interrater variability. In order to overcome these limitations, we developed an automated algorithm to classify trials into 3 alertness categories using Hori stage principles. This microstaging method was published in the journal Neuroimage (Jahannathan, Ezquerro-Nassar, et al., 2018) and is included in its published form. The method in Chapter 3 was then employed in Chapters 4 and 5, where we investigated effects of alertness and sleep deprivation in an auditory Simon task. This research was published in Journal of Neuroscience (Canales-Johnson, et al., 2020) and the manuscript is included in this thesis. Participants attended two sessions in the lab and performed an auditory Simon task, after a night of normal sleep or a night of partial sleep deprivation. We hypothesised that conflict detection processes necessary for an immediate conflict effect would be preserved under drowsiness, in line with studies showing residual processing for local, short stimuli during early sleep. Similarly, we predicted that sleep deprivation would not impair conflict detection. Furthermore, we investigated the capacity to modify behavioural control using information from the previous trial via top-down mechanisms, known as conflict adaptation. The process that mediates the conflict adaptation effect is hypothesised to require consciousness. Therefore, we hypothesised that trial-by-trial conflict adaptation would be impaired during drowsiness and sleep deprivation. In Chapter 4, we confirmed our hypothesis that the conflict detection is preserved under drowsiness, as indicated by slower reaction times for incongruent stimuli. However, we revealed an interaction between alertness, previous trial congruency and current trial congruency, suggesting a deleterious effect of drowsiness on conflict adaptation. At odds with our hypothesis, sleep deprivation was found to have a main effect on reaction times but it did not interact with any of the variables. In Chapter 5, we found an increase in theta power associated to incongruent trials, a classic neural marker of cognitive control. This effect was observed during wakefulness but not during drowsiness, suggestive of potential alternative mechanisms underlying cognitive control under reduced alertness. Overall, the findings of this thesis suggest that alertness has a different bearing depending on the cognitive process required. This has wider implications for the functional role of consciousness and suggests that sleep onset period is a useful framework to evaluate theories of consciousness.

Memory changes are commonplace over the lifespan. With advances in neuroimaging there is accumulating evidence that changes in neural dynamics may underscore changes in memory functioning. Oscillations in the theta range have long been associated with memory processes. However, how theta activity might relate to the ageing brain, and more specifically the losses in memory thought to occur during the ageing process, is less clear. This research aimed to explore these questions by analysing a large dataset of 227 adult participants from across the lifespan. Memory was assessed across different domains (episodic, associative, and working memory), and was compared with participants underlying resting-state electroencephalography (EEG). Older participants displayed significant losses in memory performance across all memory domains, as well as decreased theta power across frontal, medial parietal, and occipital regions. Amongst younger participants working memory correlated with theta power in medial frontal, left temporal, and right temporoparietal regions, however this effect was lost amongst older adults. These findings suggest that resting theta power diminishes over the lifespan in numerous cortical regions, which corresponds with a decline in memory performance. Moreover, these findings suggest a nuanced relationship between working memory and theta power, providing evidence of a unique association between theta power and working memory performance only amongst younger adults. Taken together, resting-state EEG measures appear to carry promise in understanding the nature of neurodynamic changes over the lifespan, with theta functioning dynamically changing as a function of age. Whilst this study is limited by its cross-sectional and correlational design, future longitudinal and causational research, possibly using non-invasive brain stimulation, would help clarify the effects observed.

Fluctuations during sustained attention can cause momentary lapses in performance which can have a significant impact on safety and wellbeing. However, it is less clear how unrelated tasks impact current task processes, and whether potential disturbances can be detected by autonomic and central nervous system measures in naturalistic settings. In a series of five experiments, I sought to investigate how prior attentional load impacts semi-naturalistic tasks of sustained attention, and whether neurophysiological and psychophysiological monitoring of continuous task processes and performance could capture attentional lapses. The first experiment explored various non-invasive electrophysiological and subjective methods during multitasking. The second experiment employed a manipulation of multitasking, task switching, to attempt to unravel the negative lasting impacts of multitasking on neural oscillatory activity, while the third experiment employed a similar paradigm in a semi-naturalistic environment of simulated driving. The fourth experiment explored the feasibility of measuring changes in autonomic processing during a naturalistic sustained monitoring task, autonomous driving, while the fifth experiment investigated the visual demands and acceptability of a biological based monitoring system. The results revealed several findings. While the first experiment demonstrated that only self-report ratings were able to successfully disentangle attentional load during multitasking; the second and third experiment revealed deficits in parieto-occipital alpha activity and continuous performance depending on the attentional load of a previous unrelated task. The fourth experiment demonstrated increased sympathetic activity and a smaller distribution of fixations during an unexpected event in autonomous driving, while the fifth experiment revealed the acceptability of a biological based monitoring system although further research is needed to unpick the effects on attention. Overall, the results of this thesis help to provide insight into how autonomic and central processes manifest during semi-naturalistic sustained attention tasks. It also provides support for a neuro- or biofeedback system to improve safety and wellbeing.

Background: Alzheimer's Research UK reports 850,000 individuals are living with dementia in the UK today and this is estimated to increase to one million by 2025. Alzheimer's Disease (AD), a type of dementia, accounts for approximately 60% of all diagnoses and as yet, there is no cure. Diagnosis rates in high income countries are below half and achieving acceptable levels of coverage and access to care is the challenge currently being faced (World Health Organization, 2015). Mild cognitive impairment (MCI) is a pre-clinical form of AD (Petersen, 2011) and is manifested by mild symptoms affecting cognitive function, without interference in daily activities. Neurodegeneration, defined by synaptic dysfunction, occurs decades prior to any symptoms emerging (Albert et al., 2011) interrupting electrical communication between neurones. Methods: A multi assessment approach using a wireless, dry-EEG system was used to explore the potential of recording brain activity outside the laboratory and in a more informal setting. Participants with MCI and neurotypical controls completed a neuropsychological assessment and computerized tasks designed to probe cognitive function where deficits are expected in MCI. Event related potential technique was applied to the EEG and features of interest were extracted for further analysis using supervised machine learning methods. Analysis and Results: Analysis was carried out on a group level and individual level. The latter involved training algorithms on a range of data models to identify patterns in the data with feature combinations predicting group membership of individuals with accuracy. Randomisation tests confirmed that there no differences found at a group level when considering the groups' performance during episodic memory, language, executive function (working memory) or attention tasks. For analysis at an individual level, improved model accuracies with the addition of ERP features to neuropsychological features were reported for the language and attention task. A combination of features, namely an ERP feature and neuropsychological feature were selected by the algorithms as features contributing to the best model. Discussion: The four EEG tasks in this study were designed to probe cognitive function known to deteriorate in individuals with MCI. Considering the trajectory of cognitive decline is essential when interpreting differences, or indeed no differences between the groups. However, a difference between the groups with respect to episodic memory would have been expected since it is the cognitive function upon which their diagnosis is based: this finding most likely points to the low power of the study. Feature selection was of interest across the tasks and provided insight into potential predictors of progression to AD relative to the cognitive function. However, due to the heterogeneity of MCI, it is with caution that these features are discussed since it is unknown who within the MCI group would progress to AD. Conclusion: The study was a proof of concept exploring cognitive function using a wireless dry-EEG headset outside a laboratory or hospital setting. The benefits of this system included a quick and easy set-up without compromising the quality of data. This approach could provide an efficient and affordable way to extend research into more accessible locations such as health centres and at home.

Despite large strides in our understanding of bilingual language processing at the lexical and syntactic level, including activation in L1/L2 and co-activation of languages (RHM, Kroll and Stewart, 1994; BIA - Dijkstra & Van Heuven & Grainger, 1998; BIA+, Dijkstra & Van Heuven, 2002), relatively little is known about the structure and activation of bilingual semantics. Current models provide at best a sketchy account of how the two languages compare and potentially interact at the level of meaning (The Sense Model, Finkbeiner et al, 2004; The distributed Feature Model, de Groot, 1992), and it is imperative that more empirical work is done to ascertain a more detailed picture of bilingual semantics (Multilink, Dijkstra et al, 2018). The empirical work presented in this thesis therefore aims to examine whether stronger conceptual links in the L1 afford less processing effort to achieve intra-sense conceptual mapping than in L2 (Part 1), and whether the bilingual inter-sense mapping includes co-activations of meaning (Part 2). Crucially, we use sentence processing throughout in a bid to provide an account of bilingual semantics beyond the lexical level.

Human faces convey essential information for social behaviour, such as information about others’ mental states and intentions. Crucially, many studies have claimed that several facial features such as configural facial information, emotional expressions, and gaze direction modulate how faces gain access to perceptual awareness. However, the procedures employed in said studies suffer from multiple methodological issues and limitations. In a series of experiments, I tested whether configural facial features, emotional expressions, and gaze direction modulate how faces gain access to awareness. To achieve this, I used stringent procedures that allow measurement of perceptual sensitivity and decision criterion to the location and identity of faces. I used these measures to assess how long it takes faces to reach awareness as they overcome Continuous Flash Suppression – an interocular suppression technique that can render images invisible for several seconds. Using classical and Bayesian analyses, I found that configural face processing (which occurs for upright, but not inverted faces) promotes faces’ access to awareness. Similarly, faces making eye contact gain access to awareness faster than faces looking away. Contrary to past claims, however, I found that faces expressing negative emotional expressions (anger or fear) do not enter awareness faster than neutral expressions. In another series of experiments, I measured the minimal exposure durations required for configural facial processing, emotion processing, metacognition, and conscious access. To this end, I used a newly developed LCD tachistoscope that can present images with sub-millisecond precision and examined both behavioural (psychophysical) and neural (electroencephalography) markers of processing. I found that configural face processing promotes faces’ access to awareness by showing that upright faces require shorter exposure durations than inverted faces to be seen. Crucially, only around four milliseconds of exposure were required to find this advantage. Fearful expressions, however, do not gain access to awareness faster than neutral expressions. Evidence from neural markers expanded this by showing that the exposure duration required for configural facial processing is the same as that required for faces to reach conscious access. Finally, around six milliseconds of exposure were required for emotion processing. Together, these findings shed light on the factors that affect access of faces to awareness: configural facial information and gaze direction can modulate faces’ access to perceptual awareness; and such modulation is due to perceptual sensitivity rather than decision criterion. Furthermore, the perceptual processing of faces follows a hierarchical pattern: configural information precedes and facilitates access to awareness, and emotion processing follows awareness.

Previous visual search studies have revealed that target identification can be facilitated by foreknowledge of a salient distractor’s feature(s). The term ‘templates-for-rejection’ (coined by Woodman & Luck, 2007) has been used to refer to this effect and mechanisms of active suppression are proposed to be involved (e.g.,Gaspelin & Luck, 2019; Sawaki & Luck, 2010). To date, there has been an absence of research into the possible underlying neural mechanisms of templates-for-rejection for stimuli possessing socio-motivational relevance. The first three experiments of this thesis therefore used facial expression stimuli to investigate this issue. Experiment one (Chapter 2) presented face pairs bilaterally and participants reported the sex of the non-neutral face (neutral template-for-rejection block), or the non-angry face (angry template-for-rejection block). For comparison, participants also reported the sex of the neutral, or angry face (neutral vs. angry template-for-selection blocks). The template-matching face appeared beside either a happy, sad, surprised, or disgusted face. Electroencephalography (EEG) was recorded with event-related-potentials (ERPs) time-locked to the onset of the faces. Accuracy and response times were also measured. An ERP component previously considered to reflect active suppression was observed in response to neutral, but not angry, templates-for-rejection. However, what appeared to be active suppression of neutral templates could have been attentional capture by the more emotionally salient face on the opposite side. Experiments two (neutral templates) and three (angry templates) (Chapters 3 and 4) used the same faces, but with one face appearing on the midline and the other to the left or right of fixation, in order to isolate the ERP response to the lateral stimulus. For the neutral templates-for-rejection only, active suppression was revealed, consistent with previous studies using non-motivationally relevant stimuli. This effect appeared to arise at a relatively late stage of processing, suggesting that distractor representations may have been suppressed within visual working memory (VWM) to prevent ongoing engagement with task-irrelevant stimuli. In line with the findings from Experiment one, there was no evidence of active suppression of angry templates. This may indicate the resistance of angry facial expressions to suppression, but it is also the case that angry faces may have been harder than neutral faces to perceptually discriminate from paired emotional expressions; thus, an interpretation in terms of task difficulty for the absence of active suppression of angry face templatesfor-rejection cannot be ruled out. Experiment four (Chapter 5) was therefore carried out in part to explore the role of task difficulty on ERP indices of active suppression using non-motivationally relevant colour singletons. Attentional control was better under easy (crowded array), compared with hard (sparse array) task conditions. For crowded displays only, active suppression of non-targets occurred during late attention processing which was earlier than in experiment two. It is possible therefore that an apparent lack of active suppression for angry templates-for-rejection in experiments one and three may have been due to task difficulty as well as the aversive motivational significance of the stimuli. In contrast to experiment two (neutral template findings), there was no evidence that continued engagement with the non-motivationally relevant stimuli was actively terminated within the late VWM processing stage. Notably, the ERP profile of response in Experiment four corresponded to a template for rejection impairment with respect to speed and accuracy of target identification. Experiment five (Chapter 6) sought to explore the role of increased perceptual difficulty under equivalent high cognitive demands as experiment four, in order to investigate the conditions necessary for a behavioural templates-for-rejection benefit to emerge. To conclude, this thesis demonstrates that active suppression of template distractors may be influenced by the motivational relevance and threat value of stimuli, and also cognitive and perceptual task demands. The absence of a clear behavioural template-for-rejection benefit across experiments is explained in terms of delayed active suppression due to high cognitive demands associated with the requirements to switch between template features.

Time series signals, such as Electroencephalogram (EEG) and Electrocardiogram (ECG) represent the complex dynamic behaviours of biological systems. The analysis of these signals using variety of nonlinear methods is essential for understanding variability within EEG and ECG, which potentially could help unveiling hidden patterns related to underlying physiological mechanisms. EEG is a time varying signal, and electrodes for recording EEG at different positions on the scalp give different time varying signals. There might be correlation between these signals. It is important to know the correlation between EEG signals because it might tell whether or not brain activities from different areas are related. EEG and ECG might be related to each other because both of them are generated from one co-ordinately working body. Investigating this relationship is of interest because it may reveal information about the correlation between EEG and ECG signals. This thesis is about assessing variability of time series data, EEG and ECG, using variety of nonlinear measures. Although other research has looked into the correlation between EEGs using a limited number of electrodes and a limited number of combinations of electrode pairs, no research has investigated the correlation between EEG signals and distance between electrodes. Furthermore, no one has compared the correlation performance for participants with and without medical conditions. In my research, I have filled up these gaps by using a full range of electrodes and all possible combinations of electrode pairs analysed in Time Domain (TD). Cross-Correlation method is calculated on the processed EEG signals for different number unique electrode pairs from each datasets. In order to obtain the distance in centimetres (cm) between electrodes, a measuring tape was used. For most of our participants the head circumference range was 54-58cm, for which a medium-sized I have discovered that the correlation between EEG signals measured through electrodes is linearly dependent on the physical distance (straight-line) distance between them for datasets without medical condition, but not for datasets with medical conditions. Some research has investigated correlation between EEG and Heart Rate Variability (HRV) within limited brain areas and demonstrated the existence of correlation between EEG and HRV. But no research has indicated whether or not the correlation changes with brain area. Although Wavelet Transformations (WT) have been performed on time series data including EEG and HRV signals to extract certain features respectively by other research, so far correlation between WT signals of EEG and HRV has not been analysed. My research covers these gaps by conducting a thorough investigation of all electrodes on the human scalp in Frequency Domain (FD) as well as TD. For the reason of different sample rates of EEG and HRV, two different approaches (named as Method 1 and Method 2) are utilised to segment EEG signals and to calculate Pearson’s Correlation Coefficient for each of the EEG frequencies with each of the HRV frequencies in FD. I have demonstrated that EEG at the front area of the brain has a stronger correlation with HRV than that at the other area in a frequency domain. These findings are independent of both participants and brain hemispheres. Sample Entropy (SE) is used to predict complexity of time series data. Recent research has proposed new calculation methods for SE, aiming to improve the accuracy. To my knowledge, no one has attempted to reduce the computational time of SE calculation. I have developed a new calculation method for time series complexity which could improve computational time significantly in the context of calculating a correlation between EEG and HRV. The results have a parsimonious outcome of SE calculation by exploiting a new method of SE implementation. In addition, it is found that the electrical activity in the frontal lobe of the brain appears to be correlated with the HRV in a time domain. Time series analysis method has been utilised to study complex systems that appear ubiquitous in nature, but limited to certain dynamic systems (e.g. analysing variables affecting stock values). In this thesis, I have also investigated the nature of the dynamic system of HRV. I have disclosed that Embedding Dimension could unveil two variables that determined HRV.

In this thesis, brain-computer interfaces (BCIs) are proposed as the future of our communication with machines. Once our thoughts can be directly understood and actioned, huge gains in terms of productivity and usability from consumer to professional applications can be envisioned. However, in its current state, BCI systems are limited in use to research and medical applications (in assistive technologies and in diagnoses). This project presents a BCI system for image retrieval where the system is able to identify the specific image that the user is looking at. In contrast to the majority of BCI research, we contribute work under several real world restrictions. Namely, our processing must be fully automated, use a recording technology available to consumers, and should be able to search for images not present during training, i.e. we apply zero-shot learning. To achieve this goal, we first investigate a discriminative feature extraction process for EEG brain data and establish a baseline evaluation of exemplar and category-level decoding. Once competitive decoding rates are achieved, we extend this framework to address our zero-shot restriction by making use of computer vision, linguistic, and traditional machine learning techniques. Finally, we demonstrate improvement over these results by designing a set of metric-learning-based deep neural network architectures, representing a more data-driven and state-of-the-art approach. This results in the first time a multi-modal network based on metric learning has been applied to BCI image retrieval. Our network resulted in a peak decoding rate of 5.85\% which we demonstrate is well situated against existing studies in order to give context in a field where appropriate result comparison is difficult.

Our manual interactions with objects represent the most fundamental activity in our everyday life. Whereas the grasp of an object is driven by the perceptual senses, using an object for its function relies on learnt experience to retrieve. Recent theories explain how the brain takes decisions based on perceptual information, yet the question of how does it retrieve object knowledge to use tools remains unanswered. Discovering the neuronal implementation of the retrieval of object knowledge would help understanding praxic impairments and provide appropriate neurorehabilitation. This thesis reports five investigations on the neuronal oscillatory activity involved in accessing object knowledge. Employing an original paradigm combining EEG recordings with tool use training in virtual reality, I demonstrated that beta oscillations are crucial to the retrieval of object knowledge during object recognition. Multiple evidence points toward an access to object knowledge during the 300 to 400 ms of visual processing. The different topographies of the beta oscillations suggest that tool knowledge is encoded in distinct brain areas but generally located within the left hemisphere. Importantly, learning action information about an object has consequences on its manipulations. Multiplying tool use knowledge about an object increases the beta desynchronization and slows down motor control. Furthermore, the present data report an influence of language on object manipulations and beta oscillations, in a way that learning the name of an object speeds up its use while impedes its grasp. This shred of evidence led to the formulation of three testable hypotheses extending contemporary theories of object manipulation and semantic memory. First, the preparation of object transportation or use could be distinguished by the synchronization/desynchronization patterns of mu and beta rhythms. Second, action competitions originate from both perceptuo-motor and memory systems. Third, accessing to semantic object knowledge during object processing could be indexed by the bursts of desynchronization of high-beta oscillations in the brain.

How do we lose and regain consciousness? The space between healthy wakefulness and unconsciousness encompasses a series of gradual and rapid changes in brain activity. In this thesis, I investigate computational measures applicable to the electroencephalogram to quantify the loss and recovery of consciousness from the perspective of modern theoretical frameworks. I examine three different transitions of consciousness caused by natural, pharmacological and pathological factors: sleep, sedation and coma. First, I investigate the neural dynamics of falling asleep. By combining the established methods of phase-lag brain connectivity and EEG microstates in a group of healthy subjects, a unique microstate is identified, whose increased duration predicts behavioural unresponsiveness to auditory stimuli during drowsiness. This microstate also uniquely captures an increase in frontoparietal theta connectivity, a putative marker of the loss of consciousness prior to sleep onset. I next examine the loss of behavioural responsiveness in healthy subjects undergoing mild and moderate sedation. The Lempel-Ziv compression algorithm is employed to compute signal complexity and symbolic mutual information to assess information integration. An intriguing dissociation between responsiveness and drug level in blood during sedation is revealed: responsiveness is best predicted by the temporal complexity of the signal at single- channel and low-frequency integration, whereas drug level is best predicted by the complexity of spatial patterns and high-frequency integration. Finally, I investigate brain connectivity in the overnight EEG recordings of a group of patients in acute coma. Graph theory is applied on alpha, theta and delta networks to find that increased variability in delta network integration early after injury predicts the eventual coma recovery score. A case study is also described where the re-emergence of frontoparietal connectivity predicted a full recovery long before behavioural improvement. The findings of this thesis inform prospective clinical applications for tracking states of consciousness and advance our understanding of the slow and fast brain dynamics underlying its transitions. Collating these findings under a common theoretical framework, I argue that the diversity of dynamical states, in particular in temporal domain, and information integration across brain networks are fundamental in sustaining consciousness.

The present research project adopted physiological measures to investigate the rela- tionship between anxiety and working memory (WM). The importance of WM for maths processing and the sensitivity of physiological measures to both anxious responses and executive processes make these measures appealing for advancing our understanding of maths anxiety (MA). The overarching research question asked whether anxiety affects the accuracy of the representations in WM, process that is thought to be central in performing maths task and that may explain the relationship between maths performance and MA. In this thesis the construct of anxiety was considered under two aspects. First, as source of noise causing a decrease in performance to tasks relying on WM. Second, as evolutionary determined physiological reaction to stressors. The former conceptualization of anxiety informed the parts of this thesis investigating the effects of anxiety over WM. The latter provided the starting point to investigate how physiological measures may provide insights on the relationship between the effects of MA on WM processes and maths performance. The first two studies in the thesis aimed at addressing the question of whether anxiety affects WM. WM was conceptualized as a limited capacity system where information is temporarily maintained and manipulated in order to perform a task. It is necessary for performing complex cognitive tasks as well as for later encoding of information in long term memory (Baddeley, 1992). The efficiency of such system can be compromised by events external or internal to the individual. One factor that may impact WM efficiency is anxiety. Anxiety is a state of tension in the face of threat and it is characterised by worry and increased physiological arousal. Task-irrelevant self-directed thoughts might reduce WM resources otherwise needed for performing cognitive tasks (Eysenck et al., 2007). Moreover, the effect that anxiety has on maths performance, as in the case of maths anxiety (MA), might be explained by the effect of anxiety on WM processing (Ashcraft and Kirk, 2001). The present thesis investigates the interplay between these three constructs with the aid of physiological measures. Attentional control has been found to be a central function of WM processing (Baddeley, 1992; Eysenck, 2007). In the first experiment I look at the involvement of attentional processes in memory encoding by investigating whether ERP prestimulus-memory effects (pSMEs) are of attentive nature (Otten et al., 2006). If that was the case, investigating how anxiety affected attentional processed prior memory encoding would have provided a starting point to address how MA may exert its influence on the performance of maths tasks. Contrary to my prediction, the pSME seemed to be better explained by semantic processing, rather than recruitment of attentional processes. In the second experiment, I investigated how anxiety affects maintenance in WM. To this aim I used experimentally induced by threat-of-shock might impact WM as conceptualized by resource models. Resource models of WM give a central role to attentional resources (Ma et al, 2014). To do so I implemented a modified NPU-threat test protocol (Shmitz & Grillon, 2012) combined with task in which WM was assessed through precision of recall (Bays et al., 2009) and anxiety was assessed by means of the startle reflex. In the different conditions, the probability of receiving a shock was manipulated and the estimates of the distributions fitted to the data were compared to assess whether anxiety decreased precision of recall. In a third experiment I looked at the specific case of MA. It is thought that MA impairs maths performance by depleting WM resources (Ashcraft and Kirk, 2001). While MA impacts on implicit processes such as WM, very few studies have attempted to investigate implicit measures of MA with varying success. In the experiment I attempt to replicate the behavioural findings of Rubinsten et al. (2010 and 2012) in which they use an affective priming task combined with a verification task. Furthermore, I investigate the startle reflex and heart rate variability as implicit physiological measures, which have never been studied in the context of MA.

Chronic pain in Parkinson's disease (PD) is significantly higher than in the general population, and is often overlooked to be a side effect of the impaired movement and stiffness associated with the disease. There is scientific rational to link the degeneration which occurs in the PD brain to the increased prevalence of chronic pain. Although acute pain thresholds have been shown to be lower in PD, there is a lack of neuroimaging research which investigates the role of central processing in PD. Therefore, this thesis documents one of the first neuroimaging studies to investigate whether central pain processing is abnormal in people with PD. The perception of pain can be modulated by subject driven processing such as attention, salience assignment and cognition, which is regarded as top-down modulation and is often abnormal in chronic pain conditions. A technique to investigate top-down processing is to study the neural activity during the anticipation of acute pain stimulus. Hence, in this thesis, we used EEG to record the anticipatory response to acute pain stimuli delivered by a CO2 laser. Our first experiment compared people with PD whilst off their medication to a healthy age-matched cohort and concluded that the PD patients showed an amplified anticipatory response and a higher sensitivity to the laser. In contrast, following dopamine increasing medication, no difference was shown between the PD and healthy controls. To investigate the role of dopamine in altering pain perception in PD, the second study investigated the effect of manipulating the dopamine D2 receptor in young healthy volunteers. The study reported that the agonist and antagonist induced reductions in neural activity in the parietal and temporal lobes during pain perception, and the antagonist induced a reduction in the activity of the insula during pain perception, and intensified the effect of certainty during anticipation on the rating of pain. The patient and the D2R manipulation studies demonstrated that dopamine may play a role in the top-down processing of pain and hence explain why chronic pain in PD is prevalent. Importantly the pain rating and sensitivity were not changed by Levodopa administration or D2R manipulation and supports the notion that dopamine is central to modulating top-down processing, rather than directly coding the intensity of pain. Therefore, together with further neuroimaging research conducted during the PhD, there is strong evidence of a disruption in the central processing of pain in PD and promotes the use of alternative therapies in the treatment of chronic pain in PD.

Brain-computer interface (BCI) research targets movement-free communication between the user and an electronic device using information encoded in the electrophysiological activity of the brain without involving neuromuscular pathways. In BCIs, voluntary modulation of the sensorimotor rhythms (SMRs) is the most common approach for controlling objects in real and virtual spaces using task-related multi-class classification (MC) of electroencephalography (EEG), where information about direction of movement or imagined movement of the limb is not explicitly utilized, but instead different limbs are used to attempt direction control. The work presented in this thesis is aimed at decoding the 3D trajectory of imagined arm movements from EEG, to establish the utility of non-invasive signals for controlling virtual limb(s) or limb prostheses in a more natural way. There is a growing body of evidence that decoding of (executed) arm movement trajectories from EEG is possible, but one of the major research questions in non-invasive BCI research is: 'Can 3D trajectories associated with imagined 3D limb movements be decoded or predicted from EEG?' To date, only a few studies have attempted to address this research question. This PhD thesis builds on the methodology to address this research question through a series of offline and online experiments and has resulted in novel supporting evidence and new methodology resulting in three main contributions. Contribution 1: Slow cortical potentials (SCPs) in the low delta (0-2Hz) band have, predominantly, been found to encode the trajectory of limb movements when using techniques such as multiple linear regression (mLR). This thesis presents a comparative analysis indicating that band power of mu (8-12Hz) and beta (12-28Hz) EEG oscillation encode more information from the 3D trajectory of arm movements compared to the SCPs, and that mLR and standard band-pass filtered EEG potentials may occlude information in higher frequency component thus resulting in SCP predominance. Contribution 2: The thesis shows for the first time that an assumed 3D trajectory of 'imagined' arm movements may be decoded from the band power of mu (8-12Hz), beta (12-28Hz), and low gamma (28-40Hz) oscillations and SCPs provide less information to enable decoding of 'imagined' 3D arm movement trajectories. Contribution 3: The final contribution of this thesis, to the best of the author's knowledge, is the first attempt at real-time control of two virtual arms using 3D trajectories of imagined arm movement decoded from EEG.

According to embodied accounts of human cognition, the mind is at the interface of the body and the environment. For practical reasons, however, neuroscientific research on human cognition has mostly been confined to the laboratory until now. The emergence of portable brain and body imaging research methods offers an unprecedented opportunity to capture the expression of cognitive processes during active behaviours performed in real-world contexts. In the present thesis, electroencephalography (EEG) was used to investigate embodied aspects of human cognition in motion and in the real-world. This approach, however, presents new challenges in terms of signal processing because of the increased noise related to whole body movements. As the necessary signal processing tools were not well-established, the current work involved the development of new solutions to address the specific requirements of mobile EEG data before real-world brain recordings could be validly interpreted. In a series of Event Related Potential (ERP) experiments, real-world conditions were compared to traditional lab-based conditions. The neural marker of attention (P300 ERP) was recorded when participants performed an attentional task while walking through the university's corridors versus standing in the lab. Differences in the classic P300 ERP effect show that attentional processes in the real-world are not the same as those recorded in the lab. Following up on this finding, the attenuation of the P300 effect under real-world conditions was shown to be driven by cognitive demands related to displacement through space rather than the act of walking itself. This is a demonstration, at a brain level, that when walking in the real-world, cognitive resources are reallocated to the processing of visual flow and vestibular information associated with displacement. The findings reflect the dynamic interplay between mind, body, and environment, providing innovative evidence strengthening the embodied framework of human cognition. The same dynamic interplay between body, environment and cognitive function is uniquely represented in real-world navigation. The literature on spatial navigation in humans, however, mainly involved navigating virtual reality environments often while lying on a scanner bed. Most of the evidence on the neural markers of spatial navigation comes from intracranially recorded brain oscillations in rodents. The innovation in this thesis was to investigate brain oscillations associated with cognitive function underlying real-world navigation in humans using surface electrodes. The present work demonstrates that human brain dynamics related to navigational cognitive processes can be recorded in active exploration of real-world environments. The key finding resulting from this novel approach is that real-world spatial navigation is associated with specific neural signatures underlying distinct cognitive functions. Frontal low-frequency oscillations were found to be associated with wayfinding, while parietal high-frequency oscillations were associated with spatial memory. Furthermore, these neural correlates were found to be dynamically modulated depending on the body's contextual positioning within the environment. Therefore, these findings again provide evidence in support of the embodiment theory of cognition. The final study addressed the concern that findings might reflect walking speed variation. The existing animal literature has shown that low-frequency bands are modulated by walking speed. This study characterised the specific modulations in spectral power as a function of walking speed in humans. Critically the pattern showed no similarity to the spectral patterns found in relation to real-world spatial navigation, confirming the cognitive interpretation of this work. Taken together, these findings provide innovative real-world evidence supporting the theoretical embodiment framework. The neural correlates of attention, memory, and spatial navigation were found to be modulated by the dynamic experience of one's environment. Beyond this work's theoretical implications for cognitive sciences, the present findings offer new perspectives for real-world application.

The dementias are persistent or progressive disorders affecting more than one cognitive domain that interfere with an individual’s ability to function at work or home, and represent a decline from a previous level of function. In this thesis I consider the neurophysiology of dementia at a number of levels. I investigate the ways in which the connectivity and function of the brain predisposes to the specific focal patterns of neurodegeneration seen in the various dementias. I aim to identify the mesoscopic changes that occur in individuals with neurodegeneration and how these relate to their cognitive difficulties. I show how, by assessing patients in whom there is focal disruption of brain networks and observing the outcomes in comparison to controls, I can gain insight into the mechanisms by which the normal brain makes predictions and processes language. In Chapter 1, I set the scene for the focussed experimental investigations of model diseases by beginning with an introductory, clinically-focussed review that sets out the features, aetiology, management, epidemiology and prognosis of the dementias. This places these model diseases in the context of the broader clinical challenge posed by the dementias. In Chapter 2, I turn to ‘prototypical’ model diseases that represent neurodegenerative tauopathies with predominantly cortical (Alzheimer’s disease, AD) and subcortical (Progressive Supranuclear Palsy, PSP) disease burdens. I investigate the neurophysiological causes and consequences of Tau accumulation by combining graph theoretical analyses of resting state functional MR imaging and in vivo ‘Tau’ PET imaging using the ligand AV-1451. By relating Tau distribution to the functional connectome I provide in vivo evidence consistent with ‘prion-like’ trans-neuronal spread of Tau in AD but not PSP. This provides important validation of disease modification strategies that aim to halt or slow down the progression of AD by sequestration of pathological Tau in the synapse. In contrast, I demonstrate associations consistent with regional vulnerability to Tau accumulation due to metabolic demand and a lack of trophic support in PSP but not AD. With a cross-sectional approach, using Tau burden as a surrogate marker of disease severity, I then go on to show how the changes in functional connectivity that occur as disease progresses account for the contrasting cognitive phenotypes in AD and PSP. In advancing AD, functional connectivity across the whole brain becomes increasingly random and disorganised, accounting for symptomatology across multiple cognitive domains. In advancing PSP, by contrast, disrupted cortico-subcortical and cortico-brainstem interactions meant that information transfer passed through a larger number of cortical nodes, reducing closeness centrality and eigenvector centrality, while increasing weighted degree, clustering, betweenness centrality and local efficiency. Together, this resulted in increasingly modular processing with inter-network communication taking less direct paths, accounting for the bradyphrenia characteristic of the ‘subcortical dementias’. From chapter 3 onwards, I turn to the in-depth study of a model disease called non-fluent variant Primary Progressive Aphasia (nfvPPA). This disease has a clear clinical phenotype of speech apraxia and agrammatism, associated with a focal pattern of mild atrophy in frontal lobes. Importantly, general cognition is usually well preserved until late disease. In chapter 3 itself, I relate an experiment in which patients with nfvPPA and matched controls performed a receptive language task while having their brain activity recorded with magnetoencephalography. I manipulated expectations and sensory detail to explore the role of top-down frontal contributions to predictive processes in speech perception. I demonstrate that frontal neurodegeneration led to inflexible and excessively precise predictions, and that fronto-temporal interactions play a causal role in reconciling prior predictions with degraded sensory signals. The discussion here concentrates on the insights provided by neurodegenerative disease into the normal function of the brain in processing language. Overall, I demonstrate that higher level frontal mechanisms for cognitive and behavioural flexibility make a critical functional contribution to the hierarchical generative models underlying speech perception In chapter 4, I precisely define the sequence processing and statistical learning abilities of patients with nfvPPA in comparison to patients with non-fluent aphasia due to stroke and neurological controls. I do this by exposing participants to a novel, mixed-complexity artificial grammar designed to assess processing of increasingly complex sequencing relationships, and then assessing the degree of implicit rule learning. I demonstrate that agrammatic aphasics of two different aetiologies are not disproportionately impaired on complex sequencing relationships, and that the learning of phonological and non-linguistic sequences occurs independently in health and disease. In chapter 5, I summarise the synergies between the experimental chapters, and explain how I have applied a systems identification framework to a diverse set of experimental methods, with the common goal of defining the physiology of dementia. I then return to the results of chapter 3 with a clinical focus to explain how inflexible predictions can account for subjective speech comprehension difficulties, auditory processing abnormalities and (in synthesis with chapter 4) receptive agrammatism in nfvPPA. Overall, this body of work has contributed to knowledge in several ways. It has achieved its tripartite aims by: 1) Providing in vivo evidence consistent with theoretical models of trans-neuronal Tau spread (chapter 2), and a comprehensive clinical account of the previously poorly-understood receptive symptomatology of nfvPPA (chapter 5), thus demonstrating that systems neuroscience can provide a translational bridge between the molecular biology of dementia and clinical trials of therapies and medications. In this way, I begin to disentangle the network-level causes of neurodegeneration from its consequences. 2) Providing evidence for a causal role for fronto-temporal interactions in language processing (chapter 3), and demonstrating domain separation of statistical learning between linguistic and non-linguistic sequences (chapter 4), thus demonstrating that studies of patients with neurodegenerative disease can further our understanding of normative brain function. 3) Successfully integrating neuropsychology, behavioural psychophysics, functional MRI, structural MRI, magnetoencephalography and computational modelling to provide comprehensive research training, as the platform for a future research programme in the physiology of dementia.

This doctoral thesis outlines several methodological advances in network science aimed towards uncovering rapid, complex interdependencies of electromagnetic brain activity recorded from the Electroencephalogram (EEG). This entails both new analyses and modelling of EEG brain network topologies and a novel approach to analyse rapid dynamics of connectivity. Importantly, we implement these advances to provide novel insights into pathological brain function in Alzheimer's disease. We introduce the concept of hierarchical complexity of network topology, providing both an index to measure it and a model to simulate it. We then show that the topology of functional connectivity estimated from EEG recordings is hierarchically complex, existing in a scale between random and star-like topologies, this is a paradigm shift from the established understanding that complexity arises between random and regular topologies. We go on to consider the density appropriate for binarisation of EEG functional connectivity, a methodological step recommended to produce compact and unbiased networks, in light of its new-found hierarchical complexity. Through simulations and real EEG data, we show the benefit of going beyond often recommended sparse representations to account for a broader range of hierarchy level interactions. After this, we turn our attention to assessing dynamic changes in connectivity. By constructing a unified framework for multivariate signals and graphs, inspired by network science and graph signal processing, we introduce graph-variate signal analysis which allows us to capture rapid fluctuations in connectivity robust to spurious short-term correlations. We define this for three pertinent brain connectivity estimates - Pearson's correlation coefficient, coherence and phase-lag index - and show its benefit over standard dynamic connectivity measures in a range of simulations and real data. Applying these novel methods to EEG datasets of the performance of visual short-term memory binding tasks by familial and sporadic Alzheimer's disease patients, we uncover disorganisation of the topological hierarchy of EEG brain function and abnormalities of transient phase-based activity which paves the way for new interpretations of the disease's affect on brain function. Hierarchical complexity and graph-variate dynamic connectivity are entirely new methods for analysing EEG brain networks. The former provides new interpretations of complexity in static connectivity patterns while the latter enables robust analysis of transient temporal connectivity patterns, both at the frontiers of analysis. Although designed with EEG functional connectivity in mind, we hope these techniques will be picked up in the broader field, having consequences for research into complex networks in general.

Alzheimer’s disease (AD) is a neurodegenerative disorder in which the death of brain cells causes memory loss and cognitive decline. As AD progresses, changes in the electrophysiological brain activity take place. Such changes can be recorded by the electroencephalography (EEG) and magnetoencephalography (MEG) techniques. These are the only two neurophysiologic approaches able to directly measure the activity of the brain cortex. Since EEGs and MEGs are considered as the outputs of a nonlinear system (i.e., brain), there has been an interest in nonlinear methods for the analysis of EEGs and MEGs. One of the most powerful nonlinear metrics used to assess the dynamical characteristics of signals is that of entropy. The aim of this thesis is to develop entropy-based approaches for characterization of EEGs and MEGs paying close attention to AD. Recent developments in the field of entropy for the characterization of physiological signals have tried: 1) to improve the stability and reliability of entropy-based results for short and long signals; and 2) to extend the univariate entropy methods to their multivariate cases to be able to reveal the patterns across channels. To enhance the stability of entropy-based values for short univariate signals, refined composite multiscale fuzzy entropy (MFE - RCMFE) is developed. To decrease the running time and increase the stability of the existing multivariate MFE (mvMFE) while keeping its benefits, the refined composite mvMFE (RCmvMFE) with a new fuzzy membership function is developed here as well. In spite of the interesting results obtained by these improvements, fuzzy entropy (FuzEn), RCMFE, and RCmvMFE may still lead to unreliable results for short signals and are not fast enough for real-time applications. To address these shortcomings, dispersion entropy (DispEn) and frequency-based DispEn (FDispEn), which are based on our introduced dispersion patterns and the Shannon’s definition of entropy, are developed. The computational cost of DispEn and FDispEn is O(N) – where N is the signal length –, compared with the O(N2) for popular sample entropy (SampEn) and FuzEn. DispEn and FDispEn also overcome the problem of equal values for embedded vectors and discarding some information with regard to the signal amplitudes encountered in permutation entropy (PerEn). Moreover, unlike PerEn, DispEn and FDispEn are relatively insensitive to noise. As extensions of our developed DispEn, multiscale DispEn (MDE) and multivariate MDE (mvMDE) are introduced to quantify the complexity of univariate and multivariate signals, respectively. MDE and mvMDE have the following advantages over the existing univariate and multivariate multiscale methods: 1) they are noticeably faster; 2) MDE and mvMDE result in smaller coefficient of variations for synthetic and real signals showing more stable profiles; 3) they better distinguish various states of biomedical signals; 4) MDE and mvMDE do not result in undefined values for short time series; and 5) mvMDE, compared with multivariate multiscale SampEn (mvMSE) and mvMFE, needs to store a considerably smaller number of elements. In this Thesis, two restating-state electrophysiological datasets related to AD are analyzed: 1) 148-channel MEGs recorded from 62 subjects (36 AD patients vs. 26 age-matched controls); and 2) 16-channel EEGs recorded from 22 subjects (11 AD patients vs. 11 age-matched controls). The results obtained by MDE and mvMDE suggest that the controls’ signals are more and less complex at respectively short (scales between 1 to 4) and longer (scales between 5 to 12) scale factors than AD patients’ recordings for both the EEG and MEG datasets. The p-values based on Mann-Whitney U-test for AD patients vs. controls show that the MDE and mvMDE, compared with the existing complexity techniques, significantly discriminate the controls from subjects with AD at a larger number of scale factors for both the EEG and MEG datasets. Moreover, the smallest p-values are achieved by MDE (e.g., 0.0010 and 0.0181 for respectively MDE and MFE using EEG dataset) and mvMDE (e.g., 0.0086 and 0.2372 for respectively mvMDE and mvMFE using EEG dataset) for both the EEG and MEG datasets, illustrating the superiority of these developed entropy-based techniques over the state-of-the-art univariate and multivariate entropy approaches. Overall, the introduced FDispEn, DispEn, MDE, and mvMDE methods are expected to be useful for the analysis of physiological signals due to their ability to distinguish different types of time series with a low computation time.

People with schizophrenia are known to have difficulties with episodic memory (EM). The purpose of this investigation was to examine the relationship between theta-power and: i) behavioural measures of EM performance, ii) event- related potential (ERP) indices of recollection and, iii) measures of schizophrenia symptomatology. In doing so, the aim was to gain a better understanding of the basic neural mechanisms that contribute to successful EM performance, and how these may differ for people with schizophrenia. The present investigation adopted an endophenotypic approach and collected measures of schizotypy from student participants to minimise patient factors that can confound interpretations. Fifty- four participants were asked to complete a reality-monitoring exclusion EM paradigm whilst electroencephalogram (EEG) data were collected. Measures of theta-power and ERPs were time-locked to words presented during the retrieval phase. There was a significant positive correlation between theta-power over Fz between 600-1000ms post-stimulus presentation and estimates of recollection in the imagine condition as well as a significant negative correlation between these measures of theta-power for perceive items and ERP indices of recollection for imagine items. There was also a significant positive correlation between measures of frontal theta-power in the imagine condition and negative schizotypy. The epoch employed means it is likely these measures of theta- power reflect processes contributing to the content-specific retrieval of imagined items, and post-retrieval processes acting in service of differentiating imagined items in EM. Results are discussed in terms of suggestions for interventions and directions for future research.

Human flexible behaviour is often seen in everyday life tasks. These tasks (e.g., making coffee) are composed of actions (e.g., pouring sugar) that are typically nested within a large task structures made of superordinate components and nested subcomponents. Task switching has been adopted widely as an effective tool to explore the mechanisms of flexible behaviour and can be easily adapted to real-life situations. The core hypothesis explored in this thesis was that there might be another level of control that is responsible for coordinating and scheduling actions in task switching, which I label "meta-control". My first project aimed to establish robust behavioural evidence of meta-control. To test this hypothesis, switching efficiency was treated as a measure of meta-control influence. In a novel design, participants' beliefs about switch probability were manipulated through explicit instruction, allowing manipulation of meta-level control independent of the specific sequence of tasks required. The first three behavioural experiments demonstrated behaviorally that instructions influenced the efficiency of switching even when low-level (bottom-up) experience was matched, and that this effect was motivation-dependent. In follow-up studies, this meta-control influence was found to bias voluntary task selection. Two EEG studies aimed to characterize the level at which instruction affected processing. Motor and task levels were examined with lateralized motor potentials and oscillatory alpha power, respectively. Effects of instruction only existed on alpha power. Overall, the results suggest that expectancy prompts the adoption of distinct control modes across sequences of trials. The second project explored meta-control in a task switching paradigm with a hierarchical task structure in very short (2-4 trial) sequences that were preceded by instructions about switch frequency or switch position. The experiments focused on the benefits and costs of sequence structure, based on the hypothesis that trial sequences are treated as coherent units. Three behavioural studies were conducted for testing this hypothesis. The first utilized instructions about switch frequency to test for sequence transition effects and their influence on switching performance at the trial level. In two subsequent experiments, instructions made explicit the position of a task switch. The results confirmed that instructions about switch position helped participants to build useful sequence representations, and that alternating between sequences influenced trial-level switch processes. Generally, sequence representations have a persisting influence across trials and a pronounced impact at the first trial position of sequences. The experiments in this thesis provide evidence of meta-control in task switching. The first conclusion is that meta-control can be studied with the novel experimental design introduced in this thesis and was found to be reward dependent. The second conclusion is that meta-control acts as a set of parameters that can modulate the mode of control in a sustained way across trials.

Spontaneous reactivation of memory traces during sleep enables long-term consolidation and integration of memories with prior knowledge. Emotion can enhance the consolida-tion of memories during sleep and during wakefulness. However, at the same time, emo-tion may disrupt consolidation of contextual information. Furthermore, sleep not only consolidates emotional memories, but may also influence emotional salience in terms of valence and arousal. The work presented in this thesis utilises behavioural declarative and procedural memory testing; subjective arousal, valence, and likeability testing; targeted memory reactivation (TMR); polysomnography (PSG); electroencephalography (EEG); and functional magnetic resonance imaging (fMRI), to investigate sleep-related processing and memory consolidation of emotional and neutral stimuli. In Chapter 2, we investigate sleep-dependent object-location memory and processing of valence and arousal. We use negative and neutral pictures paired with matching sounds as the stimuli. During slow-wave-sleep (SWS) or wakefulness, we replay half of the stimuli. TMR during SWS results in memory stabilization of the cued stimuli regardless of stimuli's emotionality. Furthermore, TMR enhances arousal habituation of the negative cued stimuli. TMR during wake has no effects in memory or emotional ratings. In Chapter 3, we use fMRI to further investigate the effects of TMR during SWS on emotional associative memory and arousal processing. At the behavioural level, our results indicate that the effects of TMR on arousal depend on whether the stimuli are negative or neutral and the duration of SWS. The neuroimaging results suggest that TMR has different effects on the neural correlates of arousal for negative and neutral stimuli. Furthermore, there is an interaction of TMR, emotion, and rapid-eye-movement sleep (REM) duration on the neural correlates of location memory. In Chapter 4, we investigate whether the neural traces of negative and neutral memories differ when they are reactivated during SWS. We trained an EEG classifier on wake data while participants imagined performing a motor sequence task they had previously practiced. Subsequently, we classified SWS data while auditory cues of the task were replayed. We find that the classification is above chance for both sequences, having either negative or neutral items. The classification rate of the negative sequence is positively correlated with pre-sleep performance and negatively correlated with overnight improvement. The classification rate of the neutral sequence is positively correlated with REM duration. Lastly, in Chapter 5, we investigate how implicit and explicit associative memory and emotional ratings change across periods of sleep or wakefulness. We use happy, peaceful, scary, and sad musical excerpts as stimuli. Our findings suggest that sleep stabilizes valence and arousal ratings and may stabilize explicit memory too. Conversely, wake makes valence more positive, increases arousal ratings and deteriorates explicit memory. Furthermore, emotions influence how ratings of valence and arousal and explicit memory change. These results shed light on the role of sleep in emotional processing and memory con-solidation. We provide evidence that TMR modulates emotional salience and associated memories differently for negative and neutral stimuli, both at the behavioural and at the neural level. Further work needs to be done to elucidate the different networks associated in the processing and consolidation of emotional and neutral stimuli, and the roles of SWS and REM.

Down’s Syndrome (DS) is a genetic disorder associated with intellectual disability, accelerated aging and a propensity for early-onset Alzheimer’s disease (AD). Beta-amyloid plaques are one of the pathological hallmarks of AD, and also a common characteristic of the older DS brain. AD treatment trials are now moving towards administration of the intervention at preclinical stages, with the goal of preventing cognitive decline in the first place, rather than trying to halt or reverse existing pathology. Consequently, it has become essential to develop biomarkers of AD, which can: 1. Predict clinical changes and 2. Track the effectiveness of putative preventative treatments. The strong association between DS and AD means that this research is particularly important for people with DS and it presents a high-risk group for exploring predictive biomarkers. Electroencephalography (EEG) is a non-invasive and inexpensive measure of cortical activity, which is being evaluated with the typically developing (TD) population as a potential biomarker of AD. This thesis aims to evaluate EEG as a potential predictor of cognitive decline associated with DS-AD. There are several potential EEG measures that could be explored. Following a review of the literature, the predictive potential of the following event-related potentials (ERPs): mismatch negativity (MMN) and P300 (P3a and P3b), were chosen for exploration with cross-sectional and longitudinal investigations. The thesis begins by exploring how the ERPs differ for a cross-section of 36 adults with DS and 39 age- and gender-matched TD controls. As expected, the MMN waveform was smaller for adults with DS than TD controls. However, the P3b waveform was predominantly absent for adults with DS, whilst the P3a response was significantly enlarged. The P3a response was also enlarged for the adults with DS who scored lower on a neuropsychological measure. The neuropsychological measure indexes frontal functions, which are compromised early in DS-AD. This experiment also provided evidence that MMN was related to age in DS, with increasing latencies and decreasing amplitudes for older participants. The differences in MMN amplitude between the groups (DS, TD) were isolated to the older adults. These findings lend support to the premature aging hypothesis of DS. The thesis also included a longitudinal follow-up in which 34 adults with DS underwent a repeated cognitive examination one year after their EEG and initial cognitive assessment. The analyses found that adults with DS who had lower MMN amplitudes at the initial assessment were more likely to decline at the cognitive follow-up. This finding suggests that MMN may be a potentially useful clinical tool for predicting the cognitive decline associated with DS-AD.