Search

Feedback is important to improve writing quality; however, to provide timely and personalized feedback is a time-intensive task. Currently, most literature focuses on providing (human or machine) support on product characteristics, especially after a draft is submitted. However, this does not assist students who struggle during the writing process. Therefore, in this study, we investigate the use of keystroke analysis to predict writing quality throughout the writing process. Keystroke data were analyzed from 126 English as a second language learners performing a timed academic summarization task. Writing quality was measured using participants’ final grade. Based on previous literature, 54 keystroke features were extracted. Correlational analyses were conducted to identify the relationship between keystroke features and writing quality. Next, machine learning models (regression and classification) were used to predict final grade and classify students who might need support at several points during the writing process. The results show that, in contrast to previous work, the relationship between writing quality and keystroke data was rather limited. None of the regression models outperformed the baseline, and the classification models were only slightly better than the majority class baseline (highest AUC = 0.57). In addition, the relationship between keystroke features and writing quality changed throughout the course of the writing process. To conclude, the relationship between keystroke data and writing quality might be less clear than previously posited.

Differential evolution (DE) is a popular evolutionary algorithm inspired by Darwin’s theory of evolution and has been studied extensively to solve different areas of optimisation and engineering applications since its introduction by Storn in 1997. This study aims to review the massive progress of DE in the research community by analysing the 192 articles published on this subject from 1997 to 2021, particularly studies in the past five years. The methodology used to search for relevant DE papers and an overview of the original DE are firstly explained. Recent advances in the modifications proposed to enhance the effectiveness and efficiency of the original DE are reviewed by analysing the strengths and weaknesses of each published work, followed by the potential applications of these DE variants in solving different real-world engineering problems. In contrast to most existing DE review papers, additional analyses are performed in this survey by investigating the impacts of various parameter settings on given DE variants to identify their optimal values required for solving certain problem classes. The qualities of modifications incorporated into selected DE variants are also evaluated by measuring the performance gains achieved in terms of search accuracy and/or efficiency against the original DE. The additional surveys conducted in this study are anticipated to provide more insightful perspectives for both beginners and experts of DE research, enabling their better understanding about current research trends and new motivations to outline appropriate strategic planning for future development works.

When it originated in the late 19th century, psychometrics was a field with both a scientific and a social mission: Psychometrics provided new methods for research into individual differences and at the same time considered these methods a means of creating a new social order. In contrast, contemporary psychometrics—because of its highly technical nature and its limited involvement in substantive psychological research—has created the impression of being a value-free discipline. In this article, we develop a contrasting characterization of contemporary psychometrics as a value-laden discipline. We expose four such values: that individual differences are quantitative (rather than qualitative), that measurement should be objective in a specific sense, that test items should be fair, and that the utility of a model is more important than its truth. Our goal is not to criticize psychometrics for supporting these values but rather to bring them into the open and to show that they are not inevitable and are in need of systematic evaluation.

Purpose The Irritable Bowel Syndrome Quality of Life (IBS-QoL) questionnaire is a commonly used and validated IBS-specific QoL instrument. However, this questionnaire is in contrast to the EQ-5D-5L, not preference-based and as such does not allow calculation of QALYs. The objective of this study was to describe the convergent- and known-group validity of both questionnaires and to develop a mapping algorithm from EQ-5D-5L which enable IBS-QoL scores to be transformed into utility scores for use in economic evaluations. Methods We used data from two multicenter randomized clinical trials, which represented the estimation and external validation dataset. The convergent validity was investigated by examining correlations between the EQ-5D-5L and IBS-QoL and the known-group validity by calculating effect sizes. Ordinary least squares (OLS), censored least absolute deviations (CLAD), and mixture models were used in this mapping approach. Results 283 IBS patients were included (n = 189 vs. n = 84). Mean IBS-QoL score was 71.13 (SD 15.66) and mean EQ-5D-5L utility score was 0.73 (SD 0.19). The overall sensitivity of the IBS-QoL and EQ-5D-5L to discriminate between patient and disease characteristics was similar. CLAD model 4, containing the total IBS-QoL score and squared IBS-SSS (IBS severity scoring system), was chosen as the most appropriate model to transform IBS-QoL scores into EQ-5D-5L utility scores. Conclusion This study reports the development of an algorithm where the condition-specific questionnaire IBS-QoL can be used to calculate utility values for use in economic evaluations. Including a clinical measure, IBS-SSS, in the model improved the performance of the algorithm.

Robust scientific evidence shows that human performance predictions are more valid when information is combined mechanically (with a decision rule) rather than holistically (in the decision-maker's mind). Yet, information is often combined holistically in practice. One reason is that decision makers lack the knowledge of evidence-based decision making. In a performance prediction task, we tested whether watching an educational video on evidence-based decision making increased decision-makers' use of a decision rule and their prediction accuracy immediately after the manipulation and a month later. Furthermore, we manipulated whether participants earned incentives for accurate predictions. Existing research showed that incentives decrease decision-rule use and prediction accuracy. We hypothesized that this is the case for decision makers who did not receive educational information about evidence-based decision making, but that incentives increase decision-rule use and prediction accuracy for participants who received educational information. Our results showed that educational information increased decision-rule use. This resulted in increased prediction accuracy, but only immediately after receiving the educational information. In contrast to the existing literature, incentives slightly increased decision-rule use. We did not find evidence that this effect was larger for educated participants. Providing decision makers with educational information may be effective to increase decision-rule use in practice. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

In this article, we propose a new method to ob-tain feedback controllers of an unknown dynamical systemdirectly from noisy input/state data. The key ingredient ofour design is a new matrix S-lemma that will be proven inthis article. We provide both strict and nonstrict versionsof this S-lemma, which are of interest in their own right.Thereafter, we will apply these results to data-driven con-trol. In particular, we will derive nonconservative designmethods for quadratic stabilization, H2 and H∞ control, allin terms of data-based linear matrix inequalities. In contrastto previous work, the dimensions of our decision variablesare independent of the time horizon of the experiment. Ourapproach, thus, enables control design from large datasets.

Designing visually appealing layouts for multimedia documents containing text, graphs and images requires a form of creative intelligence. Modelling the generation of layouts has recently gained attention due to its importance in aesthetics and communication style. In contrast to standard prediction tasks, there are a range of acceptable layouts which depend on user preferences. For example, a poster designer may prefer logos on the top-left while another prefers logos on the bottom-right. Both are correct choices yet existing machine learning models treat layouts as a single choice prediction problem. In such situations, these models would simply average over all possible choices given the same input forming a degenerate sample. In the above example, this would form an unacceptable layout with a logo in the centre. In this paper, we present an auto-regressive neural network architecture, called LayoutMCL, that uses multi-choice prediction and winner-takes-all loss to effectively stabilise layout generation. LayoutMCL avoids the averaging problem by using multiple predictors to learn a range of possible options for each layout object. This enables LayoutMCL to generate multiple and diverse layouts from a single input which is in contrast with existing approaches which yield similar layouts with minor variations. Through quantitative benchmarks on real data (magazine, document and mobile app layouts), we demonstrate that LayoutMCL reduces Fr\'echet Inception Distance (FID) by 83-98% and generates significantly more diversity in comparison to existing approaches., Comment: 9 pages

Significance In the adult brain, multimodal speech perception that interfaces with a bidirectional interaction of perception and production speech systems is increasingly accepted. Speech perception in infancy is already highly multisensory, suggesting an early emerging representation for speech across sensory modalities. We provide electrophysiological evidence for sensorimotor influences on auditory speech discrimination responses in 3-mo-old infants who are several months away from producing canonical babbling. Auditorily, infants discriminated both contrasts tested. However, a tongue-tip articulatory inhibition diminished the /ɗa/-/ɖa/ discrimination (both phones involve the tongue-tip movement as a place of articulation), whereas accentuating the /ba/-/ɗa/ discrimination (different articulators involved in the articulation of the two phones). These findings suggest that prebabbling infants’ speech perception is more robustly multisensory than previously considered., While there is increasing acceptance that even young infants detect correspondences between heard and seen speech, the common view is that oral-motor movements related to speech production cannot influence speech perception until infants begin to babble or speak. We investigated the extent of multimodal speech influences on auditory speech perception in prebabbling infants who have limited speech-like oral-motor repertoires. We used event-related potentials (ERPs) to examine how sensorimotor influences to the infant’s own articulatory movements impact auditory speech perception in 3-mo-old infants. In experiment 1, there were ERP discriminative responses to phonetic category changes across two phonetic contrasts (bilabial–dental /ba/-/ɗa/; dental–retroflex /ɗa/-/ɖa/) in a mismatch paradigm, indicating that infants auditorily discriminated both contrasts. In experiment 2, inhibiting infants’ own tongue-tip movements had a disruptive influence on the early ERP discriminative response to the /ɗa/-/ɖa/ contrast only. The same articulatory inhibition had contrasting effects on the perception of the /ba/-/ɗa/ contrast, which requires different articulators (the lips vs. the tongue) during production, and the /ɗa/-/ɖa/ contrast, whereby both phones require tongue-tip movement as a place of articulation. This articulatory distinction between the two contrasts plausibly accounts for the distinct influence of tongue-tip suppression on the neural responses to phonetic category change perception in definitively prebabbling, 3-mo-old, infants. The results showing a specificity in the relation between oral-motor inhibition and phonetic speech discrimination suggest a surprisingly early mapping between auditory and motor speech representation already in prebabbling infants.

Children's learning capabilities change while growing up. One framework that describes the cognitive and neural development of children's growing learning abilities is the two‐component model. It distinguishes processes that integrate separate features into a coherent memory representation (associative component) and executive abilities, such as elaboration, evaluation, and monitoring, that support memory processing (strategic component). In an fMRI study using an object‐location association paradigm, we investigated how the two components influence memory performance across development. We tested children (10–12 years, n = 31), late adolescents (18 years, n = 29), and adults (25+ years, n = 30). For studying the associative component, we also probed how the utilisation of prior knowledge (schemas) facilitates memory across age groups. Children had overall lower retrieval performance, while adolescents and adults did not differ from each other. All groups benefitted from schemas, but this effect did not differ between groups. Performance differences between groups were associated with deactivation of the dorsal medial prefrontal cortex (dmPFC), which in turn was linked to executive functioning. These patterns were stronger in adolescents and adults and seemed absent in children. Thus, the children's executive system, the strategic component, is not as mature and thus cannot facilitate memory performance in the same way as in adolescents/adults. In contrast, we did not find age‐related differences in the associative component; with activity in the angular gyrus predicting memory performance systematically across groups. Overall, our results suggest that differences of executive rather than associative abilities explain memory differences between children, adolescents, and adults., In an fMRI study, we investigated whether memory differences between children, adolescents, and adults stem from developmental changes in executive abilities, the strategic component, or rather from differences in mechanisms related to binding different features together into a memory representation, the associative component. Overall, our results suggest that differences of executive rather than associative abilities explain memory differences between children, adolescents and adults.

The “Ising model” refers to both the statistical and the theoretical use of the same equation. In this article, we introduce both uses and contrast their differences. We accompany the conceptual introduction with a survey of Ising-related software packages in R. Since the model’s different uses are best understood through simulations, we make this process easily accessible with fully reproducible examples. Using simulations, we show how the theoretical Ising model captures local-alignment dynamics. Subsequently, we present it statistically as a likelihood function for estimating empirical network models from binary data. In this process, we give recommendations on when to use traditional frequentist estimators as well as novel Bayesian options.

Personality analysis is an important area of research in several fields, including psychology, psychiatry, and neuroscience. With the recent dramatic improvements in machine learning, it has also become a popular research area in computer science. While the current computational methods are able to interpret behavioral cues (e.g., facial expressions, gesture, and voice) to estimate the level of (apparent) personality traits, accessible assessment tools are still substandard for practical use, not to mention the need for fast and accurate methods for such analyses. In this study, we present multimodal deep architectures to estimate the Big Five personality traits from (temporal) audio-visual cues and transcribed speech. Furthermore, for a detailed analysis of personality traits, we have collected a new audio-visual dataset, namely: Self-presentation and Induced Behavior Archive for Personality Analysis (SIAP). In contrast to the available datasets, SIAP introduces recordings of induced behavior in addition to self-presentation (speech) videos. With thorough experiments on SIAP and ChaLearn LAP First Impressions datasets, we systematically assess the reliability of different behavioral modalities and their combined use. Furthermore, we investigate the characteristics and discriminative power of induced behavior for personality analysis, showing that the induced behavior indeed includes signs of personality traits.

In contrast to conventional (single-label) classification, the setting of multilabel classification (MLC) allows an instance to belong to several classes simultaneously. Thus, instead of selecting a single class label, predictions take the form of a subset of all labels. In this paper, we study an extension of the setting of MLC, in which the learner is allowed to partially abstain from a prediction, that is, to deliver predictions on some but not necessarily all class labels. This option is useful in cases of uncertainty, where the learner does not feel confident enough on the entire label set. Adopting a decision-theoretic perspective, we propose a formal framework of MLC with partial abstention, which builds on two main building blocks: First, the extension of underlying MLC loss functions so as to accommodate abstention in a proper way, and second the problem of optimal prediction, that is, finding the Bayes-optimal prediction minimizing this generalized loss in expectation. It is well known that different (generalized) loss functions may have different risk-minimizing predictions, and finding the Bayes predictor typically comes down to solving a computationally complexity optimization problem. In the most general case, given a prediction of the (conditional) joint distribution of possible labelings, the minimizer of the expected loss needs to be found over a number of candidates which is exponential in the number of class labels. We elaborate on properties of risk minimizers for several commonly used (generalized) MLC loss functions, show them to have a specific structure, and leverage this structure to devise efficient methods for computing Bayes predictors. Experimentally, we show MLC with partial abstention to be effective in the sense of reducing loss when being allowed to abstain.

We develop purely nonparametric methods for the analysis of repeated measures designs with missing values. Hypotheses are formulated in terms of purely nonparametric treatment effects. In particular, data can have different shapes even under the null hypothesis and therefore, a solution to the nonparametric Behrens-Fisher problem in repeated measures designs will be presented. Moreover, global testing and multiple contrast test procedures as well as simultaneous confidence intervals for the treatment effects of interest will be developed. All methods can be applied for the analysis of metric, discrete, ordinal, and even binary data in a unified way. Extensive simulation studies indicate a satisfactory control of the nominal type-I error rate, even for small sample sizes and a high amount of missing data (up to 30%). We apply the newly developed methodology to a real data set, demonstrating its application and interpretation.

Information from camera traps is used for inferences on species presence, richness, abundance, demography, and activity. Camera trap placement design is likely to influence these parameter estimates. Herein we simultaneously generate and compare estimates obtained from camera traps (a) placed to optimize large carnivore captures and (b) random placement, to infer accuracy and biases for parameter estimates. Both setups recorded 25 species when same number of trail and random cameras (n = 31) were compared. However, species accumulation rate was faster with trail cameras. Relative abundance indices (RAI) from random cameras surrogated abundance estimated from capture-mark-recapture and distance sampling, while RAI were biased higher for carnivores from trail cameras. Group size of wild-ungulates obtained from both camera setups were comparable. Random cameras detected nocturnal activities of wild ungulates in contrast to mostly diurnal activities observed from trail cameras. Our results show that trail and random camera setup give similar estimates of species richness and group size, but differ for estimates of relative abundance and activity patterns. Therefore, inferences made from each of these camera trap designs on the above parameters need to be viewed within this context.

A reliable forecasting model is required for photovoltaic (PV) power output because solar energy is highly volatile. Another driver for the need of a reliable forecasting model is concept drift, which means that the statistical properties of the data change over time. In this paper, an online forecasting method to handle concept drift is proposed. First, the problem of forecasting in batch learning is transformed into a forecasting in online learning setting. Then, an online learning algorithm is applied, which is good for handling concept drift. Through experiments using the real‐world data, it is shown that the method noticeably improves performance compared to the case where a trained model is used. Under various concept drift scenarios, the method improves performance by up to 87.3%. It is also shown that the re‐training method (a representative existing method) has several limitations. This method requires several issues to be solved, such as selection of a proper window size, and this is evident through results showing different performance under different settings. In contrast, the method shows a reliable and desirable performance under various concept drift scenarios and thus outperforms the re‐training method. The method improves performance by up to 79%.

Students academic performance could be affected by excessive use of the smartphone. This study focuses on analyzing the level of cell phone addiction in engineering university students. It also seeks to determine if there is a difference by gender in this behavior. 306 engineering students from a Technological Institute in Veracruz, Mexico participated in the study. The instrument used to obtain data was SAS-SV (Smartphone addiction scale-short version) designed by [1]. In order to identify the set of indicators with the highest factor loadings, an exploratory factor analysis was carried out with extraction of components and orthogonal rotation with the Varimax method. To identify if there is a difference by gender, the t test is used to contrast the hypothesis about two independent population means. The findings demonstrated the extraction of two components, which differs from the one-dimensional model proposed by [1], and no difference was found between the groups of male and female students.

Road signs recognition plays an important role in improving traffic safety for both drivers and pedestrians. To ensure this recognition, many approaches are proposed by researchers. To overcome the limitations of the existing methods, Deep Learning approaches are used. This type of approaches achieves high recognition performances, and is also less sensitive to real world adverse conditions. However, they are in contrast very computationally expensive due essentially to three main factors, which are more precisely, the size of input images, the type of used layers, and the number of used parameters. From this perspective, the objective of this work is to adopt an approach that aims to reduce this computational complexity, in order to ensure a fast and efficient classification of traffic signs, especially for low and limited resources environments. The adopted approach reaches good classification accuracies, and that by using BTSCD dataset.

Background The Quality of Life Impact Refractive Correction (QIRC) questionnaire is a Rasch-validated instrument to assess the quality of life of ametropes with refractive correction. The original QIRC was validated in the United Kingdom. This study aimed to validate the Malay version of the QIRC among refractive correction wearers in Malaysia using Rasch analysis. Methods The original 20-item QIRC was forward-backward translated into Malay in preparation for the Pilot Malay QIRC. The pilot version was pre-tested on 105 spectacle/contact lens-corrected myopes, and the results were reviewed and cross-culturally adapted to produce the Final Malay QIRC. The final version was self-administered to a new sample of 304 participants. A Rasch analysis was conducted to evaluate the items and response categories of the Pilot and the Final Malay QIRC. Test-retest reliability was also analysed on the Final Malay QIRC. Results Based on the pre-test findings, Rasch analysis revealed a multidimensional scale (functional scale [Items 1 to 13] and emotional scale [Items 14 to 20], which were separated in subsequent analysis), unordered response categories for the functional scale (Category 3 was collapsed into Category 2), one misfit item (Item 3 was removed) and six items required modification (Items 4, 6 to 9, and 12 were reworded and cross-culturally adapted). In the Final Malay QIRC, both the functional and emotional scales had ordered response categories, good person reliability (functional, 0.80; emotional, 0.81) and separation index (functional, 2.01; emotional, 2.06), well-targeted items (targeting precision: functional, 0.28 logits; emotional, 0.08 logits), and satisfactory fit statistics (infit and outfit mean square were less than 1.50 for all items). A noticeable differential item functioning (DIF) between genders was found in Item 18 (DIF contrast, 0.40 logits; p = 0.04). Test-retest reliability analysis demonstrated a high intraclass correlation coefficient (0.94) and Cronbach’s alpha (0.97) with a coefficient of repeatability of ±8.14 units. Conclusions The Malay-translated version of the QIRC has good psychometric characteristics for assessing the quality of life of refractive correction wearers in Malaysia. This translated and cross-culturally adapted Malay QIRC is a valid and reliable instrument that can be used in routine clinical practice.

The Cia-Cia language, which is spoken by around 79.000 people in the Bau-Bau area of Buton Island, does not have its own writing system. In 2009, the Korean alphabet, Hangeul, was approved by the Bau-Bau city government for transcribing Cia-Cia, owing to the similarity of phonemes in Korean to those in Cia-Cia. This research aimed to compare the acoustics of monophthongs in the Korean and Cia-Cia languages with an experimental phonetic approach and to discuss writing system problems in Cia-Cia when adopting the Korean writing system. Based on the classification, the Cia-Cia vowels /i, e, a, u, and o/ are equivalent to the Korean vowels /이 ([i]), 에 ([e]), 아 ([a]), 우 ([u]), 오 ([o])/. However, there are two Korean vowels that have no Cia-Cia equivalents namely, /으/ ([ɨ]) and /어/ ([ə]). In general, the vowel equivalents between the two languages have significant differences in terms of their acoustic characteristics. Nonetheless, unlike other vowel equivalents, the Cia-Cia vowel /u/ and Korean vowel /우/ ([u]) when pronounced show similar phonetic features in terms of position and oral cavity opening level. In contrast, the Cia-Cia vowel /a/ and Korean vowel /아/ ([a]) when pronounced is shown to have the same oral cavity opening level, but different tongue positions. The use of the vowel /으/ ([ɨ]) in Hangeul's writing in transcribing particular Cia-Cia words is a unique feature of Korean grammar. This renders some transcriptions different from the pronunciation of the source Cia-Cia word. This is because of the limitations of Korean syllables, which, unlike Cia-Cia words, are not able to be written as double consonants.

Background Evidence supports an inverse relationship between weight status and motor competence, but most work utilizes body mass index as the proxy for weight status. Body mass index fails to account for essential components of body composition, which may be critical for motor performance. The purpose of this investigation was to examine the relationship between fundamental motor skills competency and body composition (i.e., fat mass, fat percentage, and fatfree mass) as measured by bio-electrical impedance analysis and body mass index in children. Methods Two hundred forty-four children from the Southeastern portion of the United States participated in this project (6.05 ± 2.01 years, 53.3% male). Fundamental motor skills were measured using the Test of Gross Motor Development – 2nd edition and body composition was assessed with the Tanita SC-331S Body Composition Analyzer (bio-electrical impedance analysis). Body mass index was calculated using CDC normative growth charts. Results Bio-electrical impedance analysis measures accounted for 23.1%, F(3, 241) = 24.10, p p = .086 variance in locomotor and object control subscales, respectively; body mass index accounted for 8.4% (locomotor) and 0.1% (object control) variance. For the Test of Gross Motor Development -2nd edition total score, bio-electrical impedance analysis measures accounted for 24.4% F(3, 241) = 25.90, p p p Conclusions Different components of body composition (i.e., fat free mass) were associated with different aspects of fundamental motor skills competency. Excess body fat may be a morphological constraint to proficient locomotor performance when transporting the body through space. In contrast, body composition did not significantly predict object manipulation performance. More work is needed to understand the causality and directionality of this relationship; however, bio-electrical impedance analysis accounts for more variance in fundamental motor skills performance than body mass index in a field-based setting.

The association between two ordinal variables can be expressed with a polychoric correlation coefficient. This coefficient is conventionally based on the assumption that responses to ordinal variables are generated by two underlying continuous latent variables with a bivariate normal distribution. When the underlying bivariate normality assumption is violated, the estimated polychoric correlation coefficient may be biased. In such a case, we may consider other distributions. In this paper, we aimed to provide an illustration of fitting various bivariate distributions to empirical ordinal data and examining how estimates of the polychoric correlation may vary under different distributional assumptions. Results suggested that the bivariate normal and skew-normal distributions rarely hold in the empirical datasets. In contrast, mixtures of bivariate normal distributions were often not rejected.

Color is an important element to consider when shaping urban characteristics. However, previous studies seldom included quantitative analyses of color relationships between urban agglomerations within proximal regions and with similar cultures to distinguish and shape individual urban personalities. This study focused on Xiamen, Zhangzhou, and Quanzhou metropolitan areas, which are influenced by Minnan culture, and collected natural and cultural landscape network images that collectively represent the urban landscape in China. Color extraction, computer vision processing technologies, and clustering algorithms, such as k-means partitioning, hierarchical methods, and co-occurrence frequency, were applied using image recognition. We then established an urban color database and quantified color attributes. Finally, we conducted a comparative analysis of dominant colors and color combination associations in Xiamen, Zhangzhou, and Quanzhou metropolitan areas to explore their similarities and differences and define their characteristics. We also considered other cities of the same type for comparison.

Andrea Palladio's Renaissance villas are amongst the most famous and widely studied examples of domestic architecture ever produced. The majority of past research about Palladio's architecture employed historical, mathematical and computational methods to analyse their complex proportional systems and rules. In contrast, this paper examines three of Palladio's arguments about his villas plans which relate to their spatial properties and topological connections. Specifically, this paper uses a computational method – the justified plan graph (JPG) – to test two arguments about the location and significance of the primary salon on the plan, and a third about the extent to which the rooms in Palladio's plans are, as he claims, flexible enough to contain alternative functions. Using ten of Palladio's piano nobile (main floor) plans from I Quattro Libri Dell'Architettura as cases, this paper develops mathematical data to test three hypotheses framed around Palladio's plans and theories.

Objective The purpose of this study was to compare the effects of linear (LP), nonlinear (NLP), and differential (DL) motor learning methods on the emergence of creative actions in soccer. Method Sixty-six novice players were randomly allocated to one of three groups, which practiced soccer skills for three months, according to different motor learning methods. In the LP-group, attention was focused on acquiring movement patterns largely through prescriptive instructions and feedback and frequent repetitions. By contrast, in the NLP-group, the aim was to guide the players’ search for movement solutions. This was done by manipulating task constraints to induce practice variability. Finally, in the DL-group, prescriptive instructions were again used, but now to maximize random variation in the practice of movement patterns. Results Behavioral analysis after the three-month training showed that the number of different actions (i.e., variability) was highest in the NLP-group, followed by DL-group, whereas the LP-group showed the lowest number of different actions. Similarly, the NLP-group and DL-group demonstrated more original and creative actions than the LP-group. Conclusion In sum, the NLP and DL motor learning methods, which aim to promote motor variability in practice, are most conducive for the emergence of creative actions. Coaches are advised to design training environment that help learners increase their adaptive motor repertoire and creative actions.

In the field of neurobiology of language, neuroimaging studies are generally based on stimulation paradigms consisting of at least two different conditions. Designing those paradigms can be very time-consuming and this traditional approach is necessarily data-limited. In contrast, in computational linguistics analyses are often based on large text corpora, which allow a vast variety of hypotheses to be tested by repeatedly re-evaluating the data set. Furthermore, text corpora also allow exploratory data analysis in order to generate new hypotheses. By drawing on the advantages of both fields, neuroimaging and corpus linguistics, we here present a unified approach combining continuous natural speech and MEG to generate a corpus of speech-evoked neuronal activity.

We re-examine the uncertainty of the El Niño–Southern Oscillation (ENSO) teleconnection to the North Atlantic following the investigation of Deser et al. (2017) (DES2017). Our analyses are performed on the November–December (ND) and January–February (JF) means separately and for a geographical area that covers a larger extent in the midlatitude North Atlantic than DES2017. The motivation for splitting the cold season in this way arises from the fact that the teleconnection patterns and underlying physical mechanisms are different in late fall compared to midwinter. As in DES2017, our main technique in quantifying the uncertainty is bootstrap resampling. Amplitudes and spatial correlations of the bootstrap samples are presented together effectively using Taylor diagrams. In addition to the confidence intervals calculated from Student's t tests and the percentiles of anomalous sea level pressure (SLP) values in the bootstrap samples, we also investigate additional confidence intervals using techniques that are not widely used in climate research but have different advantages. In contrast to the interpretation by DES2017, our results indicate that we can have confidence (at the 5 % significance level) in the patterns of the teleconnected SLP anomalies. The uncertainties in the amplitudes remain large, with the upper-percentile anomalies at up to 2 times those of the lower percentiles in the North Pacific and 2.8 times in the North Atlantic.

Graphical abstract, As a novel lactate-derived post-translational modification (PTM), lysine lactylation (Kla) is involved in diverse biological processes, and participates in human tumorigenesis. Identification of Kla substrates with their exact sites is crucial for revealing the molecular mechanisms of lactylation. In contrast with labor-intensive and time-consuming experimental approaches, computational prediction of Kla could provide convenience and increased speed, but is still lacking. In this work, although current identified Kla sites are limited, we constructed the first Kla benchmark dataset and developed a few-shot learning-based architecture approach to leverage the power of small datasets and reduce the impact of imbalance and overfitting. A maximum 11.7% (0.745 versus 0.667) increase of area under the curve (AUC) value was achieved in contrast to conventional machine learning methods. We conducted a comprehensive survey of the performance by combining 8 sequence-based features and 3 structure-based features and tailored a multi-feature hybrid system for synergistic combination. This system achieved >16.2% improvement of the AUC value (0.889 versus 0.765) compared with single feature-based models for the prediction of Kla sites in silico. Taken few-shot learning and hybrid system together, we present our newly designed predictor named FSL-Kla, which is not only a cutting-edge tool for Kla site profile but also could generate candidates for further experimental approaches. The webserver of FSL-Kla is freely accessible for academic research at http://kla.zbiolab.cn/.

The purpose of this study is to empirically compare and contrast the relationship between several dimensions of idiosyncratic deals (i-deals) and employee affective commitment among Asian countries. Data were collected from 909 employees in Singapore, Thailand and Japan. Results of the study show that there is a positive relationship between several i-deals and affective commitment. However, the effect size of i-deals on each dimension varies across the three countries. In particular, the effect size of task and work responsibilities i-deals on the affective commitment of Japanese employees is stronger than that of Thai employees. Simultaneously, the effect size of the i-deals on affective commitment of Singaporean employees is weaker than that of Thai employees. On the contrary, for the dimensions of schedule and location flexibility and financial incentives, the effect size of i-deals on affective commitment is weaker for Singaporean than for Japanese and Thai, and the effect sizes for Japanese and Thai are the same. The results of this study question the traditionally and universally naive conclusion that i-deals are effective. Although i-deals result in, to some extent, some positive consequences, the results of the study suggest that not in all cases do they increase employee affective commitment. Empirical evidence of the study also demonstrate that the effects i-deals had on employee affective commitment are inconsistent even across cultures with close cultural origin.

As the basis for the static security of the power grid, power load forecasting directly affects the safety of grid operation, the rationality of grid planning, and the economy of supply-demand balance. However, various factors lead to drastic changes in short-term power consumption, making the data more complex and thus more difficult to forecast. In response to this problem, a new hybrid model based on Vari-ational mode decomposition (VMD) and Long Short-Term Memory (LSTM) with seasonal factors elimination and error correction is proposed in this paper. Comprehensive case studies on four real-world load datasets from Singapore and the United States are employed to demonstrate the effectiveness and practicality of the proposed hybrid model. The experimental results show that the prediction accuracy of the proposed model is significantly higher than that of the contrast models. Index Terms-Power grid security, short-term load forecasting , seasonal factors elimination, error correction.

Objective: To evaluate the performance of gender detection tools that allow the uploading of files (e.g., Excel or CSV files) containing first names, are usable by researchers without advanced computer skills, and are at least partially free of charge.Methods: The study was conducted using four physician datasets (total number of physicians: 6,131; 50.3% female) from Switzerland, a multilingual country. Four gender detection tools met the inclusion criteria: three partially free (Gender API, NamSor, and genderize.io) and one completely free (Wiki-Gendersort). For each tool, we recorded the number of correct classifications (i.e., correct gender assigned to a name), misclassifications (i.e., wrong gender assigned to a name), and nonclassifications (i.e., no gender assigned). We computed three metrics: the proportion of misclassifications excluding nonclassifications (errorCodedWithoutNA), the proportion of nonclassifications (naCoded), and the proportion of misclassifications and nonclassifications (errorCoded).Results: The proportion of misclassifications was low for all four gender detection tools (errorCodedWithoutNA between 1.5 and 2.2%). By contrast, the proportion of unrecognized names (naCoded) varied: 0% for NamSor, 0.3% for Gender API, 4.5% for Wiki-Gendersort, and 16.4% for genderize.io. Using errorCoded, which penalizes both types of error equally, we obtained the following results: Gender API 1.8%, NamSor 2.0%, Wiki-Gendersort 6.6%, and genderize.io 17.7%.Conclusions: Gender API and NamSor were the most accurate tools. Genderize.io led to a high number of nonclassifications. Wiki-Gendersort may be a good compromise for researchers wishing to use a completely free tool. Other studies would be useful to evaluate the performance of these tools in other populations (e.g., Asian).

Background Incremental value (IncV) evaluates the performance change between an existing risk model and a new model. Different IncV metrics do not always agree with each other. For example, compared with a prescribed-dose model, an ovarian-dose model for predicting acute ovarian failure has a slightly lower area under the receiver operating characteristic curve (AUC) but increases the area under the precision-recall curve (AP) by 48%. This phenomenon of disagreement is not uncommon, and can create confusion when assessing whether the added information improves the model prediction accuracy. Methods In this article, we examine the analytical connections and differences between the AUC IncV (ΔAUC) and AP IncV (ΔAP). We also compare the true values of these two IncV metrics in a numerical study. Additionally, as both are semi-proper scoring rules, we compare them with a strictly proper scoring rule: the IncV of the scaled Brier score (ΔsBrS) in the numerical study. Results We demonstrate that ΔAUC and ΔAP are both weighted averages of the changes (from the existing model to the new one) in separating the risk score distributions between events and non-events. However, ΔAP assigns heavier weights to the changes in higher-risk regions, whereas ΔAUC weights the changes equally. Due to this difference, the two IncV metrics can disagree, and the numerical study shows that their disagreement becomes more pronounced as the event rate decreases. In the numerical study, we also find that ΔAP has a wide range, from negative to positive, but the range of ΔAUC is much smaller. In addition, ΔAP and ΔsBrS are highly consistent, but ΔAUC is negatively correlated with ΔsBrS and ΔAP when the event rate is low. Conclusions ΔAUC treats the wins and losses of a new risk model equally across different risk regions. When neither the existing or new model is the true model, this equality could attenuate a superior performance of the new model for a sub-region. In contrast, ΔAP accentuates the change in the prediction accuracy for higher-risk regions.

In all spoken languages, speakers use demonstratives - words like this and that e to refer to entities in their immediate environment. But which factors determine whether they use one demonstrative (this) or another (that)? Here we report the results of an experiment examining the effects of referent visibility, referent distance, and addressee location on the production of demonstratives by speakers of Ticuna (isolate; Brazil, Colombia, Peru), an Amazonian language with four demonstratives, and speakers of Dutch (Indo-European; Netherlands, Belgium), which has two demonstratives. We found that Ticuna speakers' use of demonstratives displayed effects of addressee location and referent distance, but not referent visibility. By contrast, under comparable conditions, Dutch speakers displayed sensitivity only to referent distance. Interestingly, we also observed that Ticuna speakers consistently used demonstratives in all referential utterances in our experimental paradigm, while Dutch speakers strongly preferred to use definite articles. Taken together, these findings shed light on the significant diversity found in demonstrative systems across languages. Additionally, they invite researchers studying exophoric demonstratives to broaden their horizons by cross-linguistically investigating the factors involved in speakers' choice of demonstratives over other types of referring expressions, especially articles. (C) 2021 Elsevier B.V. All rights reserved.

Sensory processing necessitates discarding some information in service of preserving and reformatting more behaviorally relevant information. Sensory neurons seem to achieve this by responding selectively to particular combinations of features in their inputs, while averaging over or ignoring irrelevant combinations. Here, we expose the perceptual implications of this tradeoff between selectivity and invariance, using stimuli and tasks that explicitly reveal their opposing effects on discrimination performance. We generate texture stimuli with statistics derived from natural photographs, and ask observers to perform two different tasks: Discrimination between images drawn from families with different statistics, and discrimination between image samples with identical statistics. For both tasks, the performance of an ideal observer improves with stimulus size. In contrast, humans become better at family discrimination but worse at sample discrimination. We demonstrate through simulations that these behaviors arise naturally in an observer model that relies on a common set of physiologically plausible local statistical measurements for both tasks., Visual processing necessitates both extracting and discarding information. Here, the authors use a specialized set of stimuli and two complementary discrimination tasks to demonstrate the opposing perceptual implications of these two aspects of information processing.

A common issue in trade network analysis is missing data, as some countries do not report trade flows. This paper explores what constitutes suitable data, how to deal with missing data, and demonstrates the results using key network measures. All-to-all potential connectivity of trade between countries is considered as a starting point, in contrast to the common approach of analyzing trade networks using only the countries that actually report trade flows. In order to fill the gap between the two approaches, a more complete dataset than just the dataset of trade between reporting countries is reconstructed and the robustness of studying this bigger dataset is examined. The difference between imputed and actual network adjacency matrices is evaluated based on several centrality measures. The results are illustrated using ten commodity groups from the United Nations Database, which demonstrate that under the proposed reconstruction procedure the ranks of the countries do not change significantly as the size of the imputed network becomes bigger or smaller. Further, the degree distributions of networks based on reporting countries and trading partners are the same to within their uncertainties. So, it is robust to study the imputed bigger network that provides richer insights into trade relations, particularly for nonreporting countries.

Feature extraction is an integral part in speech emotion recognition. Some emotions become indistinguishable from others due to high resemblance in their features, which results in low prediction accuracy. This paper analyses the impact of spectral contrast feature in increasing the accuracy for such emotions. The RAVDESS dataset has been chosen for this study. The SAVEE dataset, CREMA-D dataset and JL corpus dataset were also used to test its performance over different English accents. In addition to that, EmoDB dataset has been used to study its performance in the German language. The use of spectral contrast feature has increased the prediction accuracy in speech emotion recognition systems to a good degree as it performs well in distinguishing emotions with significant differences in arousal levels, and it has been discussed in detail.

Drones and machine learning‐based automated detection methods are being used by ecologists to conduct wildlife surveys with increasing frequency. When traditional survey methods have been evaluated, a range of factors have been found to influence detection probabilities, including individual differences among conspecific animals, which can thus introduce biases into survey counts. There has been no such evaluation of drone‐based surveys using automated detection in a natural setting. This is important to establish since any biases in counts made using these methods will need to be accounted for, to provide accurate data and improve decision‐making for threatened species. In this study, a rare opportunity to survey a ground‐truthed, individually marked population of 48 koalas in their natural habitat allowed for direct comparison of the factors impacting detection probability in both ground observation and drone surveys with manual and automated detection. We found that sex and host tree preferences impacted detection in ground surveys and in manual analysis of drone imagery with female koalas likely to be under‐represented, and koalas higher in taller trees detected less frequently when present. Tree species composition of a forest stand also impacted on detections. In contrast, none of these factors impacted on automated detection. This suggests that the combination of drone‐captured imagery and machine learning does not suffer from the same biases that affect conventional ground surveys. This provides further evidence that drones and machine learning are promising tools for gathering reliable detection data to better inform the management of threatened populations., Wildlife surveys using drones are becoming increasingly popular, but little is known about the factors influencing detection of threatened species. We show a range of factors can influence the detections of koalas in a natural setting using traditional survey methods and drone surveys using manual evaluation of the imagery. Drone surveys combined with AI algorithms were not affected by these factors, suggesting that they provide a less biased approach to threatened species detection.

Statistical analyses of randomized controlled trials (RCTs) yield a causally valid estimate of the overall treatment effect, which is the contrast between the outcomes in two randomized treatment groups commonly accompanied by a confidence interval. In addition, the trial investigators may want to examine whether the observed treatment effect varies across patient subgroups (also called 'heterogeneity of treatment effects'), i.e. whether the treatment effect is modified by the value of a variable assessed at baseline. The statistical approach for this evaluation of potential effect modifiers is a test for statistical interaction to evaluate whether the treatment effect varies across levels of the effect modifier. In this article, we provide a concise and nontechnical explanation of the use of simple statistical tests for interaction to identify effect modifiers in RCTs. We explain how to calculate the test of interaction by hand, applied to a dataset with simulated data on 1,000 imaginary participants for illustration. (C) 2021 The Author(s). Published by Elsevier Inc.

This paper explores how the manipulation of vowel duration as a perceptual cue influences listeners’ perceptual ability. Four native speakers of Najdi Arabic, a well-known variety of Arabic in the Arabian Peninsula, were tested on the perception of /a/ vs. /ɛ/ vowels. Listeners’ identification and discrimination rates along each vowel continuum showed a clear effect of duration on the perception of /a/-/ɛ/ contrast. In each vowel continuum, listeners were more inclined to classify stimuli as belonging to one vowel or the other based on relative proximity to the steady-state vowel duration. Perceptibility naturally improved as duration approximated the normal duration of either vowel. Listeners’ perceptual judgments in the identification and discrimination of the vowels were swayed by their aural sensitivity to perceptual shifts (/a/-/ɛ/ at 185-195ms; /ɛ/-/a/ at 195-205ms). Moreover, findings of the identification task followed predictably from the discrimination task; this could be taken as evidence for the existence of categorical perception. Results aggregately indicate that perception of the two Najdi Arabic vowels proceeded as a function of duration.

In their fundamental paper, Luce, Steingrimsson, and Narens (2010, Psychological Review, 117, 1247-1258) proposed that ratio productions constituting a generalization of cross-modality matching may be represented on a single scale of subjective intensity, if they meet “cross-dimensional commutativity.” The present experiment is the first to test this axiom by making truly cross-modal adjustments of the type: “Make the sound three times as loud as the light appears bright!” Twenty participants repeatedly adjusted the level of a burst of noise to result in the desired sensation ratio (e.g., to be three times as intense) compared to the brightness emanating from a grayscale square, and vice versa. Cross-modal commutativity was tested by comparing a set of successive ×2×3 productions with a set of ×3×2 productions. When this property was individually evaluated for each of 20 participants and for two possible directions, i.e., starting out with a noise burst or a luminous patch, only seven of the 40 tests indicated a statistically significant violation of cross-modal commutativity. Cross-modal monotonicity, i.e. checking whether ×1, ×2, and ×3 adjustments are strictly ordered, was evaluated on the same data set and found to hold. Multiplicativity, by contrast, i.e., comparing the outcome of a ×1×6 adjustment with ×2×3 sequences, irrespective of order, was violated in 17 of 40 tests, or at least once for all but six participants. This suggests that both loudness and brightness sensations may be measured on a common ratio scale of subjective intensity, but cautions against interpreting the numbers involved at face value.

Background: There is no precise description of changes of gait during curve walking. Research in curve walking is exclusively performed in clinical settings. Research question: Is there a difference in foot placement variables between the inner- and the outer leg during curve walking in a natural environment? And are these differences correlated with time or the curvature of the path?Method: During this observational study, camera footage was shot on a crossing where pedestrians were not aware of being filmed. Participants (n = 21, male, 18?40 yrs) were selected from this video footage. Using the software package ?Movieprocessing?, the kinematic variables (time, curvature, stride length, step length, step width and relative foot angle (RFA)) were extracted from the collected data. A MANOVA and Pearson correlation test were performed to explore the data.Results: MANOVA showed no significant differences in stride length and step length between inner- and outer leg. In contrast, a significant difference between the inner (M = 0.06, SD = 0.05) and outer leg (M = 0.10, SD = 0.06, F(20,256) = 3.577, p < .001) for the step width, and the inner (M = 11.72, SD = 7.99) and outer leg (M = 11.30, SD = 8.07, F(20,256) = 4.542, p < .001) for RFA was found. Pearson correlation was significant for curvature and step width for both legs pooled (r = .28, p < .01) and the outer leg (r = .64, p < .01), as well for time and RFA in the inner (r= -.25, p < .01) and outer leg (r = .213, p < .01). Significance: This research funds further research in curve walking in natural conditions, since curve walking is found to be performed non-symmetrically and not determined by geometrics but by choice. Foot placement variables change gradually and differently for both legs during walking a curve.

Background Analytic thinking skills are important to the development of physicians. Therefore, educators and licensing boards utilize multiple-choice questions (MCQs) to assess these knowledge and skills. MCQs are written under two assumptions: that they can be written as higher or lower order according to Bloom’s taxonomy, and students will perceive questions to be the same taxonomical level as intended. This study seeks to understand the students’ approach to questions by analyzing differences in students’ perception of the Bloom’s level of MCQs in relation to their knowledge and confidence. Methods A total of 137 students responded to practice endocrine MCQs. Participants indicated the answer to the question, their interpretation of it as higher or lower order, and the degree of confidence in their response to the question. Results Although there was no significant association between students’ average performance on the content and their question classification (higher or lower), individual students who were less confident in their answer were more than five times as likely (OR = 5.49) to identify a question as higher order than their more confident peers. Students who responded incorrectly to the MCQ were 4 times as likely to identify a question as higher order than their peers who responded correctly. Conclusions The results suggest that higher performing, more confident students rely on identifying patterns (even if the question was intended to be higher order). In contrast, less confident students engage in higher-order, analytic thinking even if the question is intended to be lower order. Better understanding of the processes through which students interpret MCQs will help us to better understand the development of clinical reasoning skills.

Modern investors face a high-dimensional prediction problem: thousands of observable variables are potentially relevant for forecasting. We reassess the conventional wisdom on market efficiency in light of this fact. In our equilibrium model, N assets have cash flows that are linear in J characteristics, with unknown coefficients. Risk-neutral Bayesian investors learn these coefficients and determine market prices. If J and N are comparable in size, returns are cross-sectionally predictable ex post. In-sample tests of market efficiency reject the no-predictability null with high probability, even though investors use information optimally in real time. In contrast, out-of-sample tests retain their economic meaning.

We develop a unifying econometric framework for the analysis of heterogeneous panel data models that can account for both spatial dependence and common factors. To tackle the challenging issues of endogeneity due to the spatial lagged term and the correlation between the regressors and factors, we propose the CCEX-IV estimation procedure that approximates factors by the cross-section averages of regressors and deals with the spatial endogeneity using the internal instrumental variables. We develop the individual and Mean Group estimators, and establish their consistency and asymptotic normality. By contrast, the Pooled estimator is shown to be inconsistent in the presence of parameter heterogeneity. Monte Carlo simulations confirm that the finite sample performance of the proposed estimators is quite satisfactory. We demonstrate the usefulness of our approach with an application to the house price growth for Local Authority Districts in the UK over 1997Q1–2016Q4.

We examined firm-level and country-level antecedents of R&D internationalization strategies, focusing on differences between enterprises in emerging and advanced economies. Previous research often focuses on the relative importance of home-base-exploiting versus home-base-augmenting knowledge transfer strategies. We suggest that country-level and firm-level effects differ for the two strategies, and hence, we examined each strategy independently. Collecting data in China, India, the United States, and Germany, we demonstrated that firms' relative technological position as a firm-level characteristic can explain differences in home-base-exploiting strategies between emerging and advanced economies. In contrast, home-base-augmenting is more closely related to exploratory institutional environments, a country-level factor. Thus, either firm- or country-level antecedents can gain a dominant role, depending on the strategy implemented.

Here, we argue that polygenic risk scores (PRSs) are different epistemic objects as compared to other biomarkers such as blood pressure or sodium level. While the latter two may be subject to variation, measured inaccurately or interpreted in various ways, blood flow has a pressure and sodium is available in a concentration that can be quantified and visualised. In stark contrast, PRSs are calculated, compiled or constructed through the statistical assemblage of genetic variants. How researchers frame and name PRSs has consequences for how we interpret and value their results. We distinguish between the tangible and inferential understanding of PRS and the corresponding languages of measurement and computation, respectively. The conflation of these frames obscures important questions we need to ask: what PRS seeks to represent, whether current ways of ‘doing PRS’ are optimal and responsible, and upon what we base the credibility of PRS-based knowledge claims.

Anthropogenic macrolitter (>0.5 cm) in rivers is of increasing concern. It has been found to have an adverse effect on riverine ecosystem health, and the livelihoods of the communities depending on and living next to these ecosystems. Yet, little is known on how macrolitter reaches and propagates through these ecosystems. A better understanding of macrolitter transport dynamics is key in developing effective reduction, preventive, and cleanup measures. In this study, we analyzed a novel dataset of citizen science riverbank macrolitter observations in the Dutch Rhine-Meuse delta, spanning two years of observations on over 200 unique locations, with the litter categorized into 111 item categories according to the river-OSPAR protocol. With the use of regression models, we analyzed how much of the variation in the observations can be explained by hydrometeorology, observer bias, and location, and how much can instead be explained by temporal trends and seasonality. The results show that observation bias is very low, with only a few exceptions, in contrast with the total variance in the observations. Additionally, the models show that precipitation, wind speed, and river flow are all important explanatory variables in litter abundance variability. However, the total number of items that can significantly be explained by the regression models is 19% and only six item categories display an R2 above 0.4. This suggests that a very substantial part of the variability in macrolitter abundance is a product of chance, caused by unaccounted (and often fundamentally unknowable) stochastic processes, rather than being driven by the deterministic processes studied in our analyses. The implications of these findings are that for modeling macrolitter movement through rivers effectively, a probabilistic approach and a strong uncertainty analysis are fundamental. In turn, point observations of macrolitter need to be planned to capture short-term variability., Environmental Science & Technology, 55 (8), ISSN:0013-936X, ISSN:1520-5851

DNA methylation variability arises due to concurrent genetic and environmental influences. Each of them is a mixture of regular and noisy sources, whose relative contribution has not been satisfactorily understood yet. We conduct a systematic assessment of the age-dependent methylation by the signal-to-noise ratio and identify a wealth of “deterministic” CpG probes (about 90%), whose methylation variability likely originates due to genetic and general environmental factors. The remaining 10% of “stochastic” CpG probes are arguably governed by the biological noise or incidental environmental factors. Investigating the mathematical functional relationship between methylation levels and variability, we find that in about 90% of the age-associated differentially methylated positions, the variability changes as the square of the methylation level, whereas in the most of the remaining cases the dependence is linear. Furthermore, we demonstrate that the methylation level itself in more than 15% cases varies nonlinearly with age (according to the power law), in contrast to the previously assumed linear changes. Our findings present ample evidence of the ubiquity of strong DNA methylation regulation, resulting in the individual age-dependent and nonlinear methylation trajectories, whose divergence explains the cross-sectional variability. It may also serve a basis for constructing novel nonlinear epigenetic clocks.

While recent advancements in computation and modelling have improved the analysis of complex traits, our understanding of the genetic basis of the time at symptom onset remains limited. Here, we develop a Bayesian approach (BayesW) that provides probabilistic inference of the genetic architecture of age-at-onset phenotypes in a sampling scheme that facilitates biobank-scale time-to-event analyses. We show in extensive simulation work the benefits BayesW provides in terms of number of discoveries, model performance and genomic prediction. In the UK Biobank, we find many thousands of common genomic regions underlying the age-at-onset of high blood pressure (HBP), cardiac disease (CAD), and type-2 diabetes (T2D), and for the genetic basis of onset reflecting the underlying genetic liability to disease. Age-at-menopause and age-at-menarche are also highly polygenic, but with higher variance contributed by low frequency variants. Genomic prediction into the Estonian Biobank data shows that BayesW gives higher prediction accuracy than other approaches., Few genome-wide association studies have explored the genetic architecture of age-of-onset for traits and diseases. Here, the authors develop a Bayesian approach to improve prediction in timing-related phenotypes and perform age-of-onset analyses across complex traits in the UK Biobank.

Language learning can be stressful for learners when they are required to communicate with classmates using the target language. In such situations, high-anxious learners often feel frustrated compared to low-anxious learners. Feelings of language anxiety may obstruct learner’s willingness to communicate (WTC). However, not many studies have elaborated on the straight-forward relationships between these two variables. Thus, this study seeks to ascertain: 1) the correlations between language anxiety and WTC, 2) the differences of WTC between high-anxious and low-anxious learners, and 3) learner willingness to use the four skills in English. A total of 145 university students participated in this study. The results indicate a significant negative correlation between language anxiety and WTC. The result of the t-test indicates significant differences in WTC scores between high-anxious and low-anxious learners. Thus, the negative relationships between language anxiety and WTC as well as a clear contrast in high- and low-anxious learners’ WTC are established.

We propose multimodal machine translation (MMT) approaches that exploit the correspondences between words and image regions. In contrast to existing work, our referential grounding method considers objects as the visual unit for grounding, rather than whole images or abstract image regions, and performs visual grounding in the source language, rather than at the decoding stage via attention. We explore two referential grounding approaches: (i) implicit grounding, where the model jointly learns how to ground the source language in the visual representation and to translate; and (ii) explicit grounding, where grounding is performed independent of the translation model, and is subsequently used to guide machine translation. We performed experiments on the Multi30K dataset for three language pairs: English–German, English–French and English–Czech. Our referential grounding models outperform existing MMT models according to automatic and human evaluation metrics.