Differential polarization decoding for robust coherent optical packet switched networks

This paper proposes the use of the 2-D differential decoding to improve the robustness of dual-polarization optical packet receivers and is demonstrated in a wavelength switching scenario for the first time.

Phonics instruction using pseudowords for success in phonetic decoding

This study examined a Pseudoword Phonics Curriculum to determine if this form of instruction would increase students’ decoding skills compared to typical real-word phonics instruction. In typical phonics instruction, children learn to decode familiar words which allow them to draw on their prior knowledge of how to pronounce the word and may detract from learning decoding skills. By using pseudowords during phonics instruction, students may learn more decoding skills because they are unfamiliar with the “words” and therefore cannot draw on memory for how to pronounce the word. It was hypothesized that students who learn phonics with pseudowords will learn more decoding skills and perform higher on a real-word assessment compared to students who learn phonics with real words. ^ Students from two kindergarten classes participated in this study. An author-created word decoding assessment was used to determine the students’ ability to decode words. The study was broken into three phases, each lasting one month. During Phase 1, both groups received phonics instruction using real words, which allowed for the exploration of baseline student growth trajectories and potential teacher effects. During Phase 2, the experimental group received pseudoword phonics instruction while the control group continued real-word phonics instruction. During Phase 3, both groups were taught with real-word phonics instruction. Students were assessed on their decoding skills before and after each phase. ^ Results from multiple regression and multi-level model analyses revealed a greater increase in decoding skills during the second and third phases of the study for students who received the pseudoword phonics instruction compared to students who received the real-word phonics instruction. This suggests that pseudoword phonics instruction improves decoding skills quicker than real-word phonics instruction. This also suggests that teaching decoding with pseudowords for one month can continue to improve decoding skills when children return to real-word phonics instruction. Teacher feedback suggests that confidence with reading increased for students who learned with pseudowords because they were less intimidated by the approach and viewed pseudoword phonics as a game that involved reading “silly” words. Implications of these results, limitations of this study, and areas for future research are discussed. ^

Phonics Instruction using Pseudowords for Success in Phonetic Decoding

This study examined a Pseudoword Phonics Curriculum to determine if this form of instruction would increase students’ decoding skills compared to typical real-word phonics instruction. In typical phonics instruction, children learn to decode familiar words which allow them to draw on their prior knowledge of how to pronounce the word and may detract from learning decoding skills. By using pseudowords during phonics instruction, students may learn more decoding skills because they are unfamiliar with the “words” and therefore cannot draw on memory for how to pronounce the word. It was hypothesized that students who learn phonics with pseudowords will learn more decoding skills and perform higher on a real-word assessment compared to students who learn phonics with real words. Students from two kindergarten classes participated in this study. An author-created word decoding assessment was used to determine the students’ ability to decode words. The study was broken into three phases, each lasting one month. During Phase 1, both groups received phonics instruction using real words, which allowed for the exploration of baseline student growth trajectories and potential teacher effects. During Phase 2, the experimental group received pseudoword phonics instruction while the control group continued real-word phonics instruction. During Phase 3, both groups were taught with real-word phonics instruction. Students were assessed on their decoding skills before and after each phase. Results from multiple regression and multi-level model analyses revealed a greater increase in decoding skills during the second and third phases of the study for students who received the pseudoword phonics instruction compared to students who received the real-word phonics instruction. This suggests that pseudoword phonics instruction improves decoding skills quicker than real-word phonics instruction. This also suggests that teaching decoding with pseudowords for one month can continue to improve decoding skills when children return to real-word phonics instruction. Teacher feedback suggests that confidence with reading increased for students who learned with pseudowords because they were less intimidated by the approach and viewed pseudoword phonics as a game that involved reading “silly” words. Implications of these results, limitations of this study, and areas for future research are discussed.

Decoding Spontaneous Emotional States in the Human Brain.

Pattern classification of human brain activity provides unique insight into the neural underpinnings of diverse mental states. These multivariate tools have recently been used within the field of affective neuroscience to classify distributed patterns of brain activation evoked during emotion induction procedures. Here we assess whether neural models developed to discriminate among distinct emotion categories exhibit predictive validity in the absence of exteroceptive emotional stimulation. In two experiments, we show that spontaneous fluctuations in human resting-state brain activity can be decoded into categories of experience delineating unique emotional states that exhibit spatiotemporal coherence, covary with individual differences in mood and personality traits, and predict on-line, self-reported feelings. These findings validate objective, brain-based models of emotion and show how emotional states dynamically emerge from the activity of separable neural systems.

Recoding: reprogrammed genetic decoding with an emphasis on antizyme regulatory frameshifting

Recoding embraces mechanisms that augment the rules of standard genetic decoding. The deviations from standard decoding are often purposeful and their realisation provides diverse and flexible regulatory mechanisms. Recoding events such as programed ribosomal frameshifting are especially plentiful in viruses. In most organisms only a few cellular genes are known to employ programed ribosomal frameshifting in their expression. By far the most prominent and therefore well-studied case of cellular +1 frameshifting is in expression of antizyme mRNAs. The protein antizyme is a key regulator of polyamine levels in most eukaryotes with some exceptions such as plants. A +1 frameshifting event is required for the full length protein to be synthesized and this requirement is a conserved feature of antizyme mRNAs from yeast to mammals. The efficiency of the frameshifting event is dependent on the free polyamine levels in the cell. cis-acting elements in antizyme mRNAs such as specific RNA structures are required to stimulate the frameshifting efficiency. Here I describe a novel stimulator of antizyme +1 frameshifting in the Agaricomycotina class of Basidiomycete fungi. It is a nascent peptide that acts from within the ribosome exit tunnel to stimulate frameshifting efficiency in response to polyamines. The interactions of the nascent peptide with components of the peptidyl transferase centre and the protein exit tunnel emerge in our understanding as powerful means which the cell employs for monitoring and tuning the translational process. These interactions can modulate the rate of translation, protein cotranslational folding and localization. Some nascent peptides act in concert with small molecules such as polyamines or antibiotics to stall the ribosome. To these known nascent peptide effects we have added that of a stimulatory effect on the +1 frameshifting in antizyme mRNAs. It is becoming evident that nascent peptide involvement in regulation of translation is a much more general phenomenon than previously anticipated.

ImageLearn - Decoding Britain's Landscape

Abstract Ordnance Survey, our national mapping organisation, collects vast amounts of high-resolution aerial imagery covering the entirety of the country. Currently, photogrammetrists and surveyors use this to manually capture real-world objects and characteristics for a relatively small number of features. Arguably, the vast archive of imagery that we have obtained portraying the whole of Great Britain is highly underutilised and could be ‘mined’ for much more information. Over the last year the ImageLearn project has investigated the potential of "representation learning" to automatically extract relevant features from aerial imagery. Representation learning is a form of data-mining in which the feature-extractors are learned using machine-learning techniques, rather than being manually defined. At the beginning of the project we conjectured that representations learned could help with processes such as object detection and identification, change detection and social landscape regionalisation of Britain. This seminar will give an overview of the project and highlight some of our research results.

Étiologie génétique et cognitive de l’association entre les habiletés en lecture et les dimensions du trouble déficitaire de l’attention avec ou sans hyperactivité (TDAH)

La présente thèse examine les associations entre les dimensions du TDAH et les habiletés en lecture sur les plans phénotypique, génétique et cognitif. En premier lieu, les associations entre les dimensions du TDAH (inattention et hyperactivité/impulsivité) et les habiletés en lecture (décodage et compréhension en lecture) chez des enfants au début du primaire (6-8 ans) ont été examinées. Les résultats révèlent des associations similaires. Toutefois, seules celles entre l’inattention et les habiletés en lecture demeurent après que l’hyperactivité/impulsivité, les symptômes de trouble du comportement et les habiletés non verbales aient été contrôlés. De plus, les associations entre l’inattention et les habiletés en lecture s’expliquent en grande partie par des facteurs génétiques. En second lieu, les associations entre les dimensions du TDAH et les habiletés en lecture (lecture de mots et exactitude/vitesse lors de la lecture d’un texte) ont été étudiées à 14-15 ans. Seule l’inattention demeure associée aux habiletés en lecture après que l’hyperactivité/impulsivité, les habiletés verbales et les habiletés non verbales aient été contrôlées. L’inattention et les habiletés en lecture sont aussi corrélées sur le plan génétique, mais ces corrélations deviennent non significatives lorsque les habiletés verbales sont contrôlées. En dernier lieu, des habiletés cognitives ont été étudiées comme mécanismes sous-jacents potentiels de l’association entre l’inattention et les habiletés en lecture (décodage et compréhension en lecture) à l’enfance. Il apparait que la conscience phonologique, la vitesse de dénomination de chiffres, le traitement temporel bimodal et le vocabulaire sont des médiateurs de l’association entre l’inattention et le décodage alors que la conscience phonologique, la vitesse de dénomination de chiffres et de couleurs et le vocabulaire sont des médiateurs de l’association entre l’inattention et la compréhension en lecture. De plus, des facteurs génétiques communs ont été observés entre certains médiateurs (conscience phonologique, vitesse de dénomination des chiffres et traitement temporel bimodal), l’inattention et le décodage. Somme toute, la présente thèse montre que des facteurs génétiques expliquent en partie ces associations à l’enfance et l’adolescence. Des médiateurs cognitifs sous-tendent ces associations, possiblement par des processus génétiques et environnementaux qui devront être précisés dans le futur.

Interpretable Convolutional Neural Networks for Decoding and Analyzing Neural Time Series Data

Machine learning is widely adopted to decode multi-variate neural time series, including electroencephalographic (EEG) and single-cell recordings. Recent solutions based on deep learning (DL) outperformed traditional decoders by automatically extracting relevant discriminative features from raw or minimally pre-processed signals. Convolutional Neural Networks (CNNs) have been successfully applied to EEG and are the most common DL-based EEG decoders in the state-of-the-art (SOA). However, the current research is affected by some limitations. SOA CNNs for EEG decoding usually exploit deep and heavy structures with the risk of overfitting small datasets, and architectures are often defined empirically. Furthermore, CNNs are mainly validated by designing within-subject decoders. Crucially, the automatically learned features mainly remain unexplored; conversely, interpreting these features may be of great value to use decoders also as analysis tools, highlighting neural signatures underlying the different decoded brain or behavioral states in a data-driven way. Lastly, SOA DL-based algorithms used to decode single-cell recordings rely on more complex, slower to train and less interpretable networks than CNNs, and the use of CNNs with these signals has not been investigated. This PhD research addresses the previous limitations, with reference to P300 and motor decoding from EEG, and motor decoding from single-neuron activity. CNNs were designed light, compact, and interpretable. Moreover, multiple training strategies were adopted, including transfer learning, which could reduce training times promoting the application of CNNs in practice. Furthermore, CNN-based EEG analyses were proposed to study neural features in the spatial, temporal and frequency domains, and proved to better highlight and enhance relevant neural features related to P300 and motor states than canonical EEG analyses. Remarkably, these analyses could be used, in perspective, to design novel EEG biomarkers for neurological or neurodevelopmental disorders. Lastly, CNNs were developed to decode single-neuron activity, providing a better compromise between performance and model complexity.