<p> Hearing provides a vital means for infants to discover their environment and communicate with their caregivers. Identifying and discriminating the pitch of sounds is critical for infants in order to acquire information from speech and music. Therefore, how infants process pitch is a fundamental question in research on auditory development. The focus of this dissertation is the use of auditory event related potentials (ERPs) derived from electroencephalogram (EEG) recordings to examine the maturation of pitch perception in early infancy. </p> <p> Pitch perception in adults has been extensively studied, but little is known about the development of pitch perception during early infancy. Infant mismatch responses (MMRs) are ERP components that are elicited by infrequent changes in auditory stimuli. MMR is a promising tool to study infant pitch perception because it can be elicited without attention or a behavioural response. However, previous studies on MMRs in infants have reported inconsistent results, some reporting frontally positive responses while others report frontally negative mismatch responses. In Chapter 2, we examined MMRs to simple pitch changes in infants between 2 and 4 months of age and found both types of infant MMRs are present, but the morphological distributions and developmental trajectories are different. In Chapter 3, we reported that both types of infant MMRs are affected similarly by the amplitude of pitch change but only the positive MMR becomes stronger when stimulus presentation rate increases, which suggests different neural mechanisms for the two types of infant MMRs. The studies reported in Chapter 4 found that only the negative MMR can be elicited readily by changes in pitch patterns, suggesting that it may be functionally similar to mismatch negativity (MMN) in adults. </p> <p> The experiments in Chapter 5 used MMR as the indication of whether infants automatically integrate the frequency components of a complex tone into a single pitch percept, even when the fundamental frequency component (corresponding to the pitch) is removed. Previous studies show that adult MMN is elicited by a pitch change in such tones missing the fundamental. Previous behavioural studies using a conditioned head tum method show that 7-month-olds also perceive pitch with tones missing the fundamental. The results of the present study indicate that infants as young as 4 months of age integrate components into a single pitch percept, but evidence for this in younger infants could not be found. </p> <p> In conclusion, the current dissertation established a promising procedure utilizing infant MMR to study infant pitch perception and contributed to the knowledge of early development of pitch perception by demonstrating dramatic changes in brain response to pitch in harmonic tones in infants between 2 and 4 months old, and to pitch in tones in infants missing the fundamental between 3 and 4 months old . </p> / Thesis / Doctor of Philosophy (PhD)

Human-robot collaboration (HRC) occurs when people and robots work together in a shared environment. Current robots often use rigid programs unsuitable for HRC. Multimodal robot programming offers an easier way to control robots using inputs like voice and gestures. In this scenario, human commands from different sensors trigger the robot’s actions. However, this data-driven approach has challenges: accurately understanding power dynamics, integrating inputs, and precisely controlling the robot. To address this, we introduce EEG signals to improve robot control, requiring reliable signal processing, feature extraction, and accurate classification using machine learning and deep learning. Existing deep learning models struggle to balance accuracy and efficiency. This thesis focuses on whether dilated convolutional neural networks can improve accuracy and reduce training and reaction times compared to the baseline. After using the Morlet wavelet for EEG feature extraction, in the thesis, an existing convolutional neural network as a benchmark is employed and uses the dilated convolution algorithm for comparison. Accuracy, precision, recall, and time are used to assess the comparison algorithm’s performance. The conclusion is that the dilated convolutional neural network performs better than the baseline in accuracy and time parameters. / Samarbete mellan människa och robot (HRC) inträffar när människor och robotar arbetar tillsammans i en delad miljö. Nuvarande robotar använder ofta rigida program som inte är lämpliga för HRC. Multimodal robotprogrammering erbjuder ett enklare sätt att styra robotar med hjälp av röst och gester. I detta scenario utlöser mänskliga kommandon från olika sensorer robotens handlingar. Dock har denna datadrivna ansats utmaningar: att noggrant förstå kraftdynamik, integrera inmatning och exakt styra roboten. För att hantera detta introducerar vi EEG-signaler för att förbättra robotstyrningen, vilket kräver pålitlig signalbehandling, funktionsextraktion och noggrann klassificering med maskininlärning och djupinlärning. Nuvarande djupinlärningsmodeller har svårt att balansera noggrannhet och effektivitet. Den här artikeln fokuserar på om dilaterade konvolutionella neurala nätverk kan förbättra noggrannheten och minska träningstider och reaktionstider jämfört med baslinjen. Efter att ha använt Morlet-våg för EEG-funktionsutvinning använder artikeln en befintlig konvolutionell neural modell som referens och jämför med dilaterad konvolution för att bedöma prestandan. Noggrannhet, precision, recall och tidsparametrar bedömer jämförelsealgoritmens prestanda. Slutsatsen är att det dilaterade konvolutionella neurala nätverket presterar bättre än baslinjen vad gäller noggrannhet och tidsparametrar.

No description available.

The aim of this thesis was to help elucidate the mechanisms that underlie age-related decline in visual selective attention and working memory (WM). Older and younger adults completed a behavioural WM task, after which electroencephalogram (EEG) was recorded as participants perform a localized attentional interference (LAI) task – competition/attentional interference was manipulated by systematically altering the distance between targets and distractors. Older adults showed impaired accuracy and reaction time on the WM and LAI tasks. Two event-related-potentials, indexing spatial attention (N2pc) and target processing (Ptc), displayed attenuated amplitude and increased latency in older adults. Thus, spatial selection, target enhancement and processing speed deficits may contribute to age-related attentional impairments. Furthermore, an unexpected component was found between the N2pc and Ptc in the older adult waveforms. Preliminary analyses suggest this may be the PD, implicated in distractor suppression, which may be differentially contributing to older and younger adults’ electrophysiology and attentional processing.

The aim of this thesis was to help elucidate the mechanisms that underlie age-related decline in visual selective attention and working memory (WM). Older and younger adults completed a behavioural WM task, after which electroencephalogram (EEG) was recorded as participants perform a localized attentional interference (LAI) task – competition/attentional interference was manipulated by systematically altering the distance between targets and distractors. Older adults showed impaired accuracy and reaction time on the WM and LAI tasks. Two event-related-potentials, indexing spatial attention (N2pc) and target processing (Ptc), displayed attenuated amplitude and increased latency in older adults. Thus, spatial selection, target enhancement and processing speed deficits may contribute to age-related attentional impairments. Furthermore, an unexpected component was found between the N2pc and Ptc in the older adult waveforms. Preliminary analyses suggest this may be the PD, implicated in distractor suppression, which may be differentially contributing to older and younger adults’ electrophysiology and attentional processing.

This thesis deals with the problem of newborn seizre detection from the Electroencephalogram (EEG) signals. The ultimate goal is to design an automated seizure detection system to assist the medical personnel in timely seizure detection. Seizure detection is vital as neurological diseases or dysfunctions in newborn infants are often first manifested by seizure and prolonged seizures can result in impaired neuro-development or even fatality. The EEG has proved superior to clinical examination of newborns in early detection and prognostication of brain dysfunctions. However, long-term newborn EEG signals acquisition is considerably more difficult than that of adults and children. This is because, the number of the electrodes attached to the skin is limited by the size of the head, the newborns EEGs vary from day to day, and the newborns are reluctant of being in the recording situation. Also, the movement of the newborn can create artifact in the recording and as a result strongly affect the electrical seizure recognition. Most of the existing methods for neonates are either time or frequency based, and, therefore, do not consider the non-stationarity nature of the EEG signal. Thus, notwithstanding the plethora of existing methods, this thesis applies the discrete wavelet transform (DWT) to account for the non-stationarity of the EEG signals. First, two methods for seizure detection in neonates are proposed. The detection schemes are based on observing the changing behaviour of a number of statistical quantities of the wavelet coefficients (WC) of the EEG signal at different scales. In the first method, the variance and mean of the WC are considered as a feature set to dassify the EEG data into seizure and non-seizure. The test results give an average seizure detection rate (SDR) of 97.4%. In the second method, the number of zero-crossings, and the average distance between adjacent extrema of the WC of certain scales are extracted to form a feature set. The test obtains an average SDR of 95.2%. The proposed feature sets are both simple to implement, have high detection rate and low false alarm rate. Then, in order to reduce the complexity of the proposed schemes, two optimising methods are used to reduce the number of selected features. First, the mutual information feature selection (MIFS) algorithm is applied to select the optimum feature subset. The results show that an optimal subset of 9 features, provides SDR of 94%. Compared to that of the full feature set, it is clear that the optimal feature set can significantly reduce the system complexity. The drawback of the MIFS algorithm is that it ignores the interaction between features. To overcome this drawback, an alternative algorithm, the mutual information evaluation function (MIEF) is then used. The MIEF evaluates a set of candidate features extracted from the WC to select an informative feature subset. This function is based on the measurement of the information gain and takes into consideration the interaction between features. The performance of the proposed features is evaluated and compared to that of the features obtained using the MIFS algorithm. The MIEF algorithm selected the optimal 10 features resulting an average SDR of 96.3%. It is also shown, an average SDR of 93.5% can be obtained with only 4 features when the MIEF algorithm is used. In comparison with results of the first two methods, it is shown that the optimal feature subsets improve the system performance and significantly reduce the system complexity for implementation purpose.

The neural net computer simulations which will be presented here are based on the acceptance of a set of assumptions that for the last twenty years have been expressed in the fields of information processing, neurophysiology and cognitive sciences. First of all, neural networks and their dynamical behaviors in terms of attractors is the natural way adopted by the brain to encode information. Any information item to be stored in the neural net should be coded in some way or another in one of the dynamical attractors of the brain and retrieved by stimulating the net so as to trap its dynamics in the desired item's basin of attraction. The second view shared by neural net researchers is to base the learning of the synaptic matrix on a local Hebbian mechanism. The last assumption is the presence of chaos and the benefit gained by its presence. Chaos, although very simply produced, inherently possesses an infinite amount of cyclic regimes that can be exploited for coding information. Moreover, the network randomly wanders around these unstable regimes in a spontaneous way, thus rapidly proposing alternative responses to external stimuli and being able to easily switch from one of these potential attractors to another in response to any coming stimulus.<p><p>In this thesis, it is shown experimentally that the more information is to be stored in robust cyclic attractors, the more chaos appears as a regime in the back, erratically itinerating among brief appearances of these attractors. Chaos does not appear to be the cause but the consequence of the learning. However, it appears as an helpful consequence that widens the net's encoding capacity. To learn the information to be stored, an unsupervised Hebbian learning algorithm is introduced. By leaving the semantics of the attractors to be associated with the feeding data unprescribed, promising results have been obtained in term of storing capacity. / Doctorat en sciences appliquées / info:eu-repo/semantics/nonPublished

Cette thèse vise à détecter et améliorer l’état cognitif de l’apprenant. Cet état est défini par la capacité d’acquérir de nouvelles connaissances et de les stocker dans la mémoire. Nous nous sommes essentiellement intéressés à améliorer le raisonnement des apprenants, et ceci dans trois environnements : environnement purement cognitif Logique, jeu sérieux LewiSpace et jeu sérieux intelligent Inertia. La détection de cet état se fait essentiellement par des mesures physiologiques (en particulier les électroencéphalogrammes) afin d’avoir une idée sur les interactions des apprenants et l’évolution de leurs états mentaux. L’amélioration des performances des apprenants et de leur raisonnement est une clé pour la réussite de l’apprentissage. Dans une première partie, nous présentons l’implémentation de l’environnement cognitif logique. Nous décrivons des statistiques faites sur cet environnement. Nous avons collecté durant une étude expérimentale les données sur l’engagement, la charge cognitive et la distraction. Ces trois mesures se sont montrées efficaces pour la classification et la prédiction des performances des apprenants. Dans une deuxième partie, nous décrivons le jeu Lewispace pour l’apprentissage des diagrammes de Lewis. Nous avons mené une étude expérimentale et collecté les données des électroencéphalogrammes, des émotions et des traceurs de regard. Nous avons montré qu’il est possible de prédire le besoin d’aide dans cet environnement grâce à ces mesures physiologiques et des algorithmes d’apprentissage machine. Dans une troisième partie, nous clôturons la thèse en présentant des stratégies d’aide intégrées dans un jeu virtuel Inertia (jeu de physique). Cette dernière s’adapte selon deux mesures extraites des électroencéphalogrammes (l’engagement et la frustration). Nous avons montré que ce jeu permet d’augmenter le taux de réussite dans ses missions, la performance globale et par conséquent améliorer l’état cognitif de l’apprenant. / This thesis aims at detecting and enhancing the cognitive state of a learner. This state is measured by the ability to acquire new knowledge and store it in memory. Focusing on three types of environments to enhance reasoning: environment Logic, serious game LewiSpace and intelligent serious game Inertia. Physiological measures (in particular the electroencephalograms) have been taken in order to measure learners’ engagement and mental states. Improving learners’ reasoning is key for successful learning process. In a first part, we present the implementation of logic environment. We present statistics on this environment, with data collected during an experimental study. Three types of data: engagement, workload and distraction, these measures were effective and can predict and classify learner’s performance. In a second part, we describe the LewiSpace game, aimed at teaching Lewis diagrams. We conducted an experimental study and collected data from electroencephalograms, emotions and eye-tracking software. Combined with machine learning algorithms, it is possible to anticipate a learner’s need for help using these data. In a third part, we finish by presenting some assistance strategies in a virtual reality game called Inertia (to teach Physics). The latter adapts according to two measures extracted from electroencephalograms (frustration and engagement). Based on our study, we were able to enhance the learner’s success rate on game missions, by improving its cognitive state.

Electroencephalography and functional magnetic resonance are two different methods for measuring of neural activity. EEG signals have excellent time resolution, fMRI scans capture records of brain activity in excellent spatial resolution. It is assumed that the joint analysis can take advantage of both methods simultaneously. Statistical Parametric Mapping (SPM8) is freely available software which serves to automatic analysis of fMRI data estimated with general linear model. It is not possible to estimate automatic EEG–fMRI analysis with it. Therefore software EEG Regressor Builder was created during master thesis. It preprocesses EEG signals into EEG regressors which are loaded with program SPM8 where joint EEG–fMRI analysis is estimated in general linear model. EEG regressors consist of vectors of temporal changes in absolute or relative power values of EEG signal in the specified frequency bands from selected electrodes due to periods of fMRI acquisition of individual images. The software is tested on the simultaneous EEG-fMRI data of a visual oddball experiment. EEG regressors are calculated for temporal changes in absolute and relative EEG power values in three frequency bands of interest ( 8-12Hz, 12-20Hz a 20-30Hz) from the occipital electrodes (O1, O2 and Oz). Three types of test analyzes is performed. Data from three individuals is examined in the first. Accuracy of results is evaluated due to the possibilities of setting of calculation method of regressor. Group analysis of data from twenty-two healthy patients is performed in the second. Group EEG regressors analysis is realized in the third through the correlation matrix due to the specified type of power and frequency band outside of the general linear model.

The master thesis deals with the joint EEG-fMRI analysis based on a heuristic model that describes the relationship between changes in blood flow in active brain areas and in the electrical activity of neurons. This work also discusses various methods of extracting of useful information from the EEG and their influence on the final result of joined analysis. There were tested averaging methods of electrodes interest, decomposition by principal components analysis and decomposition by independent component analysis. Methods of averaging and decomposition by PCA give similar results, but information about a stimulus vector can not be extracted. Using ICA decomposition, we are able to obtain information relating to the certain stimulation, but there is the problem in the final interpretation and selection of the right components in a blind search for variability coupled with the experiment. It was found out that although components calculated from the time sequence EEG are independent for each to other, their spectrum shifts are correlated. This spectral dependence was eliminated by PCA / ICA decomposition from vectors of spectrum shifts. For this method, each component brings new information about brain activity. The results of the heuristic approach were compared with the results of the joined analysis based on the relative and absolute power approach from frequency bands of interest. And the similarity between activation maps was founded, especially for the heuristic model and the relative power from the gamma band (20-40Hz).