Search

The ability to learn novel motor skills is essential for patients with Parkinson's disease (PD) to regain activities of daily living. However, the underlying mechanisms of motor learning in PD remain unclear. To identify motor features that are distinctively manifested in PD during motor learning, we quantified a rich set of variables reflecting various aspects of the learning process in a virtual throwing task. While the performance outcome improved similarly over 3 days of practice for both PD patients and age-matched controls, further analysis revealed distinct learning processes between the two groups. PD patients initially performed with a slow release velocity and gradually increased it as practice progressed, whereas the control group began with an unnecessarily rapid release velocity, which they later stabilized at a lower value. Performance characteristics related to the timing of ball release and the inter-release interval did not show significant group differences, although they were modulated across practice in both groups. After one week, both groups retained the performance outcomes and underlying kinematics developed over practice. This study underscores the importance of analyzing the multi-faceted learning process to characterize motor skill learning in PD. The findings may provide insights into PD pathophysiology and inform rehabilitation strategies. [ABSTRACT FROM AUTHOR]

Facing the growing popularity of autonomous online learning, students require more agentic engagement (AE) to construct a supportive learning environment for themselves. Existing studies based on self-determination theory (SDT) state that teachers' motivating style can predict students' engagement, and students' competence beliefs can moderate the predicting effect. However, few studies have discussed the correlation between teaching motivating styles and AE in middle- and high-school online courses, considering the students' self-regulated learning (SRL) ability as a moderator. To address this gap, this study examined the effect of autonomy-supportive motivating style (ASMS) and controlling motivating style (CMS) on AE, taking into account students' SRL ability as a potential moderator. A total of 425 valid questionnaires were collected to evaluate the model. The results revealed that teachers' ASMS positively predicted students' AE; CMS positively predicted students' AE; and SRL ability positively predicted AE. However, SRL ability did not significantly moderate the correlation between teachers' motivating styles and AE. These findings provide educators with insights into how to improve students' AE, and suggest that teachers should be involved as much as possible in online teaching interactions, and explore strategies to enhance the effectiveness of students' AE. [ABSTRACT FROM AUTHOR]

While discourse competence (DC) has been identified as the core competence in English as a foreign language (EFL) learning and communication, only a few empirical studies have focused on DC in speaking assessments. This study investigates the discourse features (i.e., cohesion and coherence) of Taiwanese candidates' performances in an argumentative exposition task in the General English Proficiency Test (GEPT) by analyzing the construction of cohesion and coherence across four band scores (i.e., Bands 2, 3, 4, and 5). Cohesion analysis showed that all the three types of cohesion devices (i.e., grammatical, logical, and lexical) develop linearly with increasing band score levels. The higher-score test-takers (Bands 4 and 5) adopted grammatically and semantically more appropriate references to establish an accurate relation with the referents. They also tended to use a higher frequency of overall conjunctions (particularly additive) and repetition to promote logical and lexical cohesion. Coherence analysis found that high-score candidates were more equipped to produce a response that complies with the expected generic structure of argumentative tasks. While more complex theme-rheme patterns appeared more frequently in the discourse of higher-score candidates, theme reiteration patterns were only observed in Band 2. These results illustrate how Taiwanese EFL learners demonstrate DC with cohesive and coherence strategies in speaking test performance across band scores. Pedagogical implications for language teaching and learning are then discussed. [ABSTRACT FROM AUTHOR]

Outfit compatibility modeling is an increasingly important task that has garnered much attention from researchers, and graph neural network (GNN)-based methods have become the mainstream approach to address this task. Despite significant progress achieved by existing research, most of them have overlooked the importance of low-order connectivity in graph data and the relationship between individual fashion items and overall outfit style. To address this issue, we propose an outfit compatibility modeling scheme based on outfit-level relationships. This scheme consists of two key components: FCSA-GNN and GOR. The former consists of multiple layers of Fashion Items Relationship Propagation, where the weights of the graph edges are adapted based on the Category Co-occurrence Matrix. This approach avoids potential imbalances in the quantities of different fashion item categories within the dataset. Subsequently, by connecting the outputs of GNN at each layer, it explores the relationships between fashion item visual signals at different levels. The latter integrates the overall outfits style and further enhances the model's ability to learn a comprehensive representation of the overall outfits through the proposed outfit-level relationships. Experimental results on two real datasets, Polyvore outfit-ND and Polyvore outfit-D, demonstrate that our method outperforms existing state-of-the-art methods when considering only visual information. [ABSTRACT FROM AUTHOR]

This paper presents a new cognitive agent design approach integrating spatial knowledge representation and reasoning in agent-based modeling dedicated to land use simulations. A deep motivation for our agent-centric contribution is the ever-increasing development of spatially explicit agent simulation platforms. We build on this technological evolution and topology theory to endow the agent with human's spatial representation and reasoning following a Belief–Desire–Intention architecture. A pilot implementation of the methodology with simulation experiments on a hunting model was carried out in GAMA platform to assess agent performances. Simulations display a consistent trend of animal population dynamics and also confirm a high model sensitivity to the integration of spatial knowledge and reasoning in agent-based models of human actor. These results demonstrate a successful implementation and the importance of spatial dimension in the expressive power and the validity of agent-based models. Future research efforts should therefore emphasize on designing human knowledge representation and incorporating learning abilities to improve models efficiency. [ABSTRACT FROM AUTHOR]

Objective Methods Results Significance Application of cluster analytic procedures has advanced understanding of the cognitive heterogeneity inherent in diverse epilepsy syndromes and the associated clinical and neuroimaging features. Application of this unsupervised machine learning approach to the neuropsychological performance of persons with juvenile myoclonic epilepsy (JME) has yet to be attempted, which is the intent of this investigation.A total of 77 JME participants, 19 unaffected siblings, and 44 unrelated controls, 12 to 25 years of age, were administered a comprehensive neuropsychological battery (intelligence, language, memory, executive function, and processing speed), which was subjected to factor analysis followed by K‐means clustering of the resultant factor scores. Identified cognitive phenotypes were characterized and related to clinical, family, sociodemographic, and cortical and subcortical imaging features.Factor analysis revealed three underlying cognitive dimensions (general ability, speed/response inhibition, and learning/memory), with JME participants performing worse than unrelated controls across all factor scores, and unaffected siblings performing worse than unrelated controls on the general mental ability and learning/memory factors, with no JME vs sibling differences. K‐means clustering of the factor scores revealed three latent groups including above average (31.4% of participants), average (52.1%), and abnormal performance (16.4%). Participant groups differed in their distributions across the latent groups (p < 0.001), with 23% JME, 22% siblings, and 2% unrelated controls in the abnormal performance group; and 18% JME, 21% siblings, and 59% unrelated controls in the above average group. Clinical epilepsy variables were unassociated with cluster membership, whereas family factors (lower parental education) and abnormally increased thickness and/or volume in the frontal, parietal, and temporal‐occipital regions were associated with the abnormal cognition group.Distinct cognitive phenotypes characterize the spectrum of neuropsychological performance of patients with JME for which there is familial (sibling) aggregation. Phenotypic membership was associated with parental (education) and imaging characteristics (increased cortical thickness and volume) but not basic clinical seizure features. [ABSTRACT FROM AUTHOR]

Interfaith initiatives are seen as promising sites for societal change and personal transformation. However, many questions about the actual outcomes of such initiatives have remained unanswered. Therefore, a systematic review was conducted, focusing on 1) the outcomes of interfaith initiatives and 2) the methods used to evaluate these outcomes. This review analyses 47 empirical studies on interfaith initiatives. The review of the initiatives' outcomes indicates that interfaith initiatives can positively impact participants' knowledge about other worldviews, communication skills and appreciation for other worldviews. At the same time, it can be challenging for interfaith initiatives to positively impact participants' knowledge of socio-political challenges, awareness of power imbalance and privilege and behaviour as interfaith leaders. The analysis of the methods used to evaluate the impact of interfaith initiatives shows that most studies used qualitative methods. For studies using quantitative or mixed methods designs, a variety of measurement instruments were used, whereby there was little overlap or consensus regarding the concepts being used. The findings of this article will stimulate scholars to focus more on learning activities and evaluation methods that take societal learning outcomes into consideration. It emphasises the need to build on previous evaluation and research efforts. [ABSTRACT FROM AUTHOR]

In this paper, the authors describe an investigation of Kennesaw State University's inaugural MOOC, a professional development opportunity designed to enhance educator skills in designing blended and online learning environments for their K-12 learners. Additionally, the authors discuss design aspects of the MOOC in regard to learning impact and first steps taken to evaluate the learners' perceptions of the instructional strategies used during the MOOC. The methods included an analysis of quantitative survey data, as well as a qualitative analysis of both open-ended survey questions and discussion forum responses. The findings from the investigation are presented in terms of design factors contributing to the students' learning, as well as the strategies perceived as barriers to their learning. Results of the study indicate a successful MOOC design from the learners' perspective. [ABSTRACT FROM AUTHOR]

The eusocial "paper wasp" Polistes dominula is an invasive species that has successfully established worldwide. In the present study, we investigate the foraging behaviour of P. dominula wasps when revisiting previously gathered resources. We recorded the number of learning flights and food manipulation efficiency performed by a single target wasp during several consecutive collecting visits. Additionally, we analysed P. dominula relocating behaviour when dealing with changes in food position. We assessed whether wasps choose the previously visited dish containing food or opt for a novel baited dish placed 60 cm away. Then, we trained wasps to collect food either once or three times and compared the time taken to find the novel food position when it was displaced 60 cm away. This is the first study to demonstrate relocating behaviour in P. dominula in which wasps rapidly learned to return to certain food sites. A significant reduction in learning flights after only one collecting visit was observed. Likewise, wasps learned to manipulate the resource spending less time after each experience. Moreover, when two food sources were offered, after just one visit, the majority of wasps collected the resource from the previously rewarded dish. Furthermore, when food was displaced, foragers found the new food location more rapidly after one visit than after three consecutive collecting experiences. Our findings contribute to the understanding of P. dominula behavioural plasticity while collecting food sources in anthropized environments, which highlight the importance of considering such learning experiences. [ABSTRACT FROM AUTHOR]

Alternative reproductive tactics (ARTs) offer a unique window into how sexual selection can shape cognition and nonreproductive behaviour. Sexual selection more broadly has been shown to shape cognitive traits but has rarely been investigated in the context of ARTs. Furthermore, the divergent morphology and sexual behaviour associated with ARTs may lead to different socioecological challenges (e.g. predation, space use) that further shape nonreproductive behaviours. We take advantage of the behavioural diversity in alternative reproductive tactics among male El Abra swordtails, Xiphophorus nigrensis , to ask how cognition and behaviour diverge between male tactics. We tested courting (large), coercive (small) and mixed-strategy (intermediate-sized) male phenotypes in four assays to evaluate boldness, aggression, exploration, spatial learning and spatiotemporal learning. We found that the courting-only large male phenotype had faster latencies to the reward in the spatial learning assay than the other two male phenotypes and also had greater accuracy than chance. Furthermore, we found that courting phenotypes (large and intermediate-sized males) had greater increases in accuracy over successive trials than the coercive-only small male phenotype. We also found that small, coercive male phenotypes were bolder than large and intermediate-sized males. We found no differences in aggressive behaviour (directed at a mirror) or spatiotemporal learning (shuttle box assay) across the three phenotypes. We suggest that courting males' superior spatial learning abilities may reflect their defence of females at ephemeral foraging patches as well as their need for reliable escape routes from predators. Coercive males' increased boldness may bolster their persistent behaviour and reflect reduced predation pressure. These results suggest that sexual selection for alternative reproductive strategies affects nonreproductive behaviour and cognition in directions that may align with divergent socioecological challenges associated with each mating type. • We examine how cognition and behaviour diverge in male swordtail mating tactics. • Courtship was associated with greater performance in a spatial learning task. • Coercive-only phenotypes were more bold. • Mating tactic did not explain differences in spatiotemporal learning or aggression. [ABSTRACT FROM AUTHOR]

The intrinsic functional network architecture accounts for task-evoked brain activity changes and variabilities in cognitive performance. Relationships between the intrinsic functional network architecture and task performance or learning ability have been previously reported. However, the relationships between learning benefits and the characteristics of intrinsic functional network architecture for different types of learning methods remain unclear. In this study, we used graph theoretical analysis to examine the relationships between intrinsic functional network connectivity and learning benefits in two well-known learning methods in the field of cognitive rehabilitation—errorless learning (EL learning) and trial-and-error learning (T&E learning). We focused on the default mode network (DMN) as a task-relevant network, which can differentiate between EL and T&E learning and was found to be more important for T&E learning in a previous study. Participants performed a color–name association task with both learning methods. The graph metrics used were within-network connectivity and efficiency for the DMN. Within-DMN connectivity and DMN efficiency showed a significantly weak positive correlation with T&E scores but not with EL scores. These findings show that the intrinsic integration strength within the DMN relates to individuals' learnability through the T&E method. [ABSTRACT FROM AUTHOR]

Existing works relied on subjective interviews in clinical settings, which does not detect mental health problems at early stages. With the advancement in learning strategies, research are trying to find a better way for detecting mental disorders and dysfunctions at early stages. EEG signal measurements is one of the prolific ways for identifying mental disorders and dysfunction in a non-invasive and ubiquitous way. However, the label scarcity and multiclass classification concerning EEG signal measurements has been a challenge for the research community that hinders the realization of automated mental dysfunction identification using EEG signals. Data imbalance is another issue that indirectly relates to the data scarcity issue. To tackle this challenge, we propose a novel Transformer-based Self-Contrastive and Multitask Learning (SCAM-Learning) framework for mental dysfunction classification using EEG signals. The SCAM-Learning framework uses Transformer networks, self-supervised contrastive learning paradigm, and multitask learning strategy to improve the classification performance. The multitask learning is accomplished by utilizing simple and complex data augmentation strategies to train the network for pretext task. The self-supervised contrastive learning helps in dealing with data and label scarcity issues. We also propose a novel cross contrastive loss that helps in improving interdependent correlation matrix for improving the classification performance. Our experimental results on a publicly available dataset reveal that the proposed method can achieve up to 11.89% performance gain in comparison to the existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Knowledge representation learning is to project entities and relations into low-dimensional vector spaces, and efficiently calculate entities, relations, and complex semantic associations between them. However, the translation models that fuse multi-source information fail to specially pay attention to the semantic information of important sentences in the textual descriptions of an entity, which lead to the low quality of embedded representation of textual descriptions, and fail to utilize the primary neighbor of the entity to further constrain the entity representation, resulting in the weak knowledge representation learning ability of the models. To address these issues, we propose a joint knowledge representation learning model of sentence vectors weighting and primary neighbor constraints called SW &NC-Joint. First, the weight distribution of the sentences is calculated according to the frequency of the neighbors in each sentence, and then the textual descriptions are represented by the sentence vectors weighting. Secondly, the concept of neighbor association degree is put forward, and the primary neighbor is filtered out for each entity according to the association degree. Finally, the weighted textual description information and triple structure information are jointly trained, the most crucial thing that the model is constrained by using the distance between the entity and its primary neighbor in the training process, so as to improve the representation learning ability of the model. The link prediction and triple classification experiments are carried out on FB15K and WN18 datasets. The experimental results demonstrate that our model significantly outperforms existing state-of-the-art text-enhanced knowledge representation learning model for link prediction and triple classification tasks. [ABSTRACT FROM AUTHOR]

In order to ensure that students have a deeper and more thorough understanding of knowledge and skills, the author takes deep learning as the direction and integrates it into the real-time evaluation of basketball classrooms, organically combining the two. Deep learning provides a new theoretical perspective and driving force for the optimization of real-time evaluation, and points out the direction for promoting evaluation effectiveness; The optimization of real-time evaluation provides a feasible path for achieving deep learning for students. After the implementation of the real-time evaluation plan between the experimental group and the control group, the comparison results of basketball skills tests were obtained: before and after the experiment, the experimental group's 60 second self throwing and self grabbing, vigorous dribbling, comprehensive dribbling, teaching competitions, and teaching lectures all showed significant differences in test indicators; Before and after the experiment, the control group showed significant differences in the four test indicators of 60 second self throwing and self grabbing, vigorous dribbling, comprehensive dribbling, and teaching lectures. The teaching competition indicators also showed significant differences, all of which were improved. Comparison results of deep learning abilities: Before and after the experiment, there were significant differences in the dimensions of experimental combination ability and learning perseverance, while there were significant differences in the dimensions of communication ability, self-learning ability, and total score of deep learning ability; Before and after the experiment, there was a significant difference in the dimensions of control group work ability, while there was no significant difference in the dimensions of self-directed learning ability, learning perseverance, and communication ability. Overall, there was a very significant difference in deep learning ability. Overall, the experimental group showed a more significant improvement in deep learning ability compared to the control group. The real-time evaluation plan for sports education professional basketball classrooms based on deep learning is more effective in improving students' basketball skills and deep learning abilities than the conventional real-time evaluation plan for sports education professional basketball classrooms. The design of the plan has certain effectiveness and feasibility. [ABSTRACT FROM AUTHOR]

Computational intelligence is used to create artificially intelligent systems with the ability to learn, adapt, and solve problems. Learners' computational abilities can be improved with the help of Cognitive Neuroscience techniques. Computational intelligence refers to the ability of a learner based on the complex operations of the brain in the four relevant dimensions: Visual (V), Aural (A), Kinematic (K), and Reading/Writing (R/W). Our proposed framework consists of electroencephalogram (EEG) data acquisition, signal extraction, and EEG signal categorization to assess students' cognitive learning abilities. Our proposed approach uses an EEG device equipped with a microprocessor, a Think Gear (TGAM) EEG sensor, and a PCB of 16 dry electrodes. The EEG device and the remote processing device are connected by Bluetooth. The EEG signal provides the students with neurofeedback on their cognitive learning capability. The feedback obtained through the learning process will be endowing to improvise computational intelligence. The statistical derivative, Pearson Co-efficient, is used to find the correlation among the derived attributes. The attributes considered are the learner's gender, stream, age, and geographical region. The results obtained highlight that gender, stream, and age have no correlation with the detectable learning index, and the most accurate learners are kinesthetic ones. Bi-modal learners, who can maintain focus while reading and using their kinematic abilities, had the secondhighest learning capacity. [ABSTRACT FROM AUTHOR]

In the context of drilling operations, the study investigated the ability of a combination of rigid mineral particles and composite plugging agents to seal simulated cracks effectively. The study used a neural network model to predict the outcomes of experiments using this combination, based on data collected during the research. Initially, a backpropagation (BP) neural network was used to establish the prediction model, which was later optimized using the particle swarm optimization (PSO) algorithm to improve its accuracy, stability, and learning abilities. As a result, the optimized prediction model was found to be capable of providing accurate and compliant drilling plugging formulas quickly. This feature helped guide targeted formula experiments and significantly reduced experimental time and costs. In five practices in a well area in the southern Sichuan region of China, the application success rate was as high as 60%, and the time spent on plugging was reduced by an average of 36%. Overall, this study contributes to the development of effective and efficient drilling techniques, which are essential in the exploration and production of hydrocarbon resources. [ABSTRACT FROM AUTHOR]

Malware poses a significant threat to cybersecurity, continuously evolving to evade traditional detection methods. Deep learning (DL) offers a promising avenue for addressing this challenge by leveraging its ability to learn complex patterns from large amounts of data. This paper introduces a novel deep learning architecture that combines convolutional neural network (CNN), long short-term memory network (LSTM), and radial basis function network (RBF) to extract discriminative features from malware images. Our model captures both low-level visual details and high-level semantic information for robust and accurate malware classification. The Softmax layer assigns probabilities to different malware classes, enabling precise classification. We evaluated the model on three diverse datasets: Malimg, MaleVis, and Microsoft Big 2015, each differing in malware families, image characteristics, and collection methods. The model achieved accuracy rates of 99.06%, 93.55%, and 97.52% on these datasets, respectively, demonstrating its effectiveness and potential in malware image classification. [ABSTRACT FROM AUTHOR]

Cortical neurons are versatile and efficient coding units that develop strong preferences for specific stimulus characteristics. The sharpness of tuning and coding efficiency is hypothesized to be controlled by delicately balanced excitation and inhibition. These observations suggest a need for detailed co-tuning of excitatory and inhibitory populations. Theoretical studies have demonstrated that a combination of plasticity rules can lead to the emergence of excitation/inhibition (E/I) co-tuning in neurons driven by independent, low-noise signals. However, cortical signals are typically noisy and originate from highly recurrent networks, generating correlations in the inputs. This raises questions about the ability of plasticity mechanisms to self-organize co-tuned connectivity in neurons receiving noisy, correlated inputs. Here, we study the emergence of input selectivity and weight co-tuning in a neuron receiving input from a recurrent network via plastic feedforward connections. We demonstrate that while strong noise levels destroy the emergence of co-tuning in the readout neuron, introducing specific structures in the non-plastic pre-synaptic connectivity can re-establish it by generating a favourable correlation structure in the population activity. We further show that structured recurrent connectivity can impact the statistics in fully plastic recurrent networks, driving the formation of co-tuning in neurons that do not receive direct input from other areas. Our findings indicate that the network dynamics created by simple, biologically plausible structural connectivity patterns can enhance the ability of synaptic plasticity to learn input-output relationships in higher brain areas. Author summary: Many studies of learning in biological neural networks have focused on how plausible plasticity rules shape individual connections between neurons in a recurrent network or in feed-forward projections. However, in the latter case, the presynaptic network properties, such as clustered connectivity, strongly influence population dynamics and, thus, the learning process of the projections. Here, we aim to close this gap by showing how non-plastic network structure can strongly influence the outcomes of synaptic learning. We show that unstructured recurrent connectivity and the presence of noise can significantly reduce the ability of synaptic plasticity to separate presynaptic input populations and coordinate excitatory and inhibitory connection development, while the introduction of overlapping clustered structures in fixed or plastic recurrent connectivity can boost synaptic learning. Using Bayesian inference, we identify the optimal connectivity structures for a recurrent network and demonstrate the strong effects that relatively simple connectivity patterns can have on the ability of a network to learn via local plasticity. [ABSTRACT FROM AUTHOR]

Knowledge tracing is a technology that models students' changing knowledge state over learning time based on their historical answer records, thus predicting their learning ability. It is the core module that supports the intelligent education system. To address the problems of sparse input data, lack of interpretability and weak capacity to capture the relationship between exercises in the existing models, this paper build a deep knowledge tracing model DKVMN&MRI based on the Dynamic Key-Value Memory Network (DKVMN) that incorporates multiple relationship information including exercise-knowledge point relations, exercise-exercise relations, and learning-forgetting relations. In the model, firstly, the Q-matrix is utilized to map the link between knowledge points and exercises to the input layer; secondly, improved DKVMN and LSTM are used to model the learning process of learners, then the Ebbinghaus forgetting curve function is introduced to simulate the process of memory forgetting in learners, and finally, the prediction strategies of Item Response Theory (IRT) and attention mechanism are used to combine the similarity relationship between learners' knowledge state and exercises to calculate the probability that learners would correctly respond during the subsequent time step. Through extensive experiments on three real-world datasets, we demonstrate that DKVMN&MRI has significant improvements in both AUC and ACC metrics contrast with the latest models. Furthermore, the study provides explanations at both the exercise level and learner knowledge state level, demonstrating the interpretability and efficacy of the proposed model. [ABSTRACT FROM AUTHOR]

Purpose: This study aimed to examine the mechanisms influencing metacognitive awareness among nursing students, encompassing educational environment, learning motivation, emotional intelligence, and self-directed learning ability. These findings could be a theoretical reference for formulating metacognitive awareness intervention strategies specifically designed for nursing students. Method: Through stratified sampling, 600 nursing students were recruited from two universities in China. Data collection involved five self-report instruments: the Metacognitive Awareness Inventory, Medical Education Environment Evaluation Scale, Learning Motivation Scale, Emotional Intelligence Scale, and Self-Directed Learning Ability Scale. Confirmatory factor analysis was utilized to test the measurement model, while structural equation modeling utilized multiple indices to assess model fit. Bootstrapping was conducted to analyze total, direct, and indirect effects. Results: All study variables differed significantly according to age and year level (p < 0.01). Metacognitive awareness positively correlated with the educational environment, learning motivation, emotional intelligence, and self-directed learning ability (r = 0.493–0.729; all p < 0.01). The test of the measurement model shows good reliability and validity. The model depicting factors influencing metacognitive awareness demonstrated a good fit (χ2/df = 2.42, SRMR = 0.048, RMSEA = 0.049, CFI = 0.947, IFI = 0.935). Path analyses revealed that educational environment (EE), learning motivation (LM), and emotional intelligence (EI) directly impacted self-directed learning ability (SDLA) (βEE = 0.189, βLM = 0.436; βEI = 0.215; all p < 0.05). Moreover, educational environment and self-directed learning ability directly influenced metacognitive awareness (βEE = 0.224, βSDLA = 0.520; all p < 0.05). Lastly, the educational environment, learning motivation, and emotional intelligence influenced metacognitive awareness by self-directed learning ability. Conclusion: Metacognitive awareness in nursing students rises with age and year level. Females exhibit more potent self-directed learning abilities than males. Educational environment, motivation, and emotional intelligence indirectly influence metacognitive awareness via self-directed learning ability. This finding has enriched the ways and perspectives of metacognitive awareness training for nursing students. Highlights: This study profoundly explores the characteristics and interconnections between metacognitive awareness, self-directed learning ability, educational environment, emotional intelligence, and learning motivation in undergraduate nursing students in China. By examining the role of self-directed learning ability as a mediator, the research offers theoretical and practical insights that could guide the design of intervention strategies to enhance metacognitive awareness in nursing education. [ABSTRACT FROM AUTHOR]

Background: Self-regulated learning (SRL) is the core purpose of lifelong learning and the precondition for medical staff to continuously adapt to and excel in their current and future positions. Good SRL abilities positively impact the academic performance and clinical skills of medical students. Despite its importance, the factors influencing SRL among medical students in China remain under-explored. This study investigates these factors among Chinese medical undergraduates. Methods: This cross-sectional survey was conducted among a convenience sample of undergraduate students at all four medical universities in Guangxi, using a questionnaire developed and tailored by Zude Zhu. Responses from 2,179 Chinese medical undergraduate students were included in the analyses. The data were collected using the College Student SRL Questionnaire which consisted of 69 items rated on a 6-point Likert scale, ranging from 1 (very inconsistent) to 6 (very consistent), and scored by calculating the mean and standard deviation. The data were analyzed using the Statistical Package for Social Sciences, version 22.0 (IBM SPSS Corp). Results: Univariate analysis showed that registered residence, specialty, educational system, parental educational background, parenting style, annual household income, parental occupation, grade point average (GPA), engagement in SRL activities, and participation in SRL training were significantly associated with SRL scores (p < 0.05 for all). The results of the multiple linear regression analysis revealed that educational system ("5 + 3"-year) (p < 0.001; 95% Confidence Interval (CI), 9.842 to 15.686), mother's occupation (staff of government agency/institution) (p < 0.001; 95% CI, − 3.798 to − 2.081), GPA (p < 0.001; 95% CI, − 10.009 to − 5.712), engagement in SRL activities (p = 0.001; 95% CI, − 11.903 to − 3.162), and participation in SRL training (p < 0.001; 95% CI, − 15.966 to − 9.972) were positively related to SRL abilities in medical students. Conclusions: Further studies are necessary to identify additional external (e.g., academic environment, parents' expectations for their children) and internal factors (e.g., student cadres or not, interest in medicine or not, whether there is a clear career planning) affecting SRL abilities. Initiating relevant training early in the college career could enhance SRL capabilities in medical students. [ABSTRACT FROM AUTHOR]

Introduction: Online self-regulated learning (OSRL) is crucial for online learners' success and lifelong learning. This study investigated the OSRL characteristics of K-12 teachers in China, who embody the dual roles of learners and educators. It also analyzed the differences in OSRL abilities across different genders, education stages, and school locations, and examined the correlation between education qualifications, years of service, and OSRL abilities. Methods: A self-report measure was used to assess K-12 teachers' OSRL, with data collected from 1,443 K-12 teachers (394 males and 1,049 females) in northeastern China. Descriptive statistical analysis was utilized to explore the characteristics of their OSRL. Independent t-tests and ANOVA were employed to investigate differences in OSRL among different genders, regions, and educational stages. Correlation analysis was conducted to examine the relationships between education qualifications, years of service, and OSRL among K-12 teachers. Results: The data analysis revealed that K-12 teachers scored the lowest in Online Learning Self-Efficacy (OLSE), followed by Online Learning Management Strategies (OLMS), and then Online Learning Resource Management (OLRM). Notably, urban K-12 teachers exhibited higher OLSE abilities than their rural counterparts, while high school teachers scored higher in OLSE and OLMS compared to primary school teachers. Furthermore, a positive correlation was observed between education qualifications and K-12 teachers' OLSE, whereas a negative correlation was identified between years of service and K-12 teachers' OLSE. Conclusion: The findings reveal an uneven development across various dimensions of online self-regulated learning among K-12 teachers, necessitating support for the advancement of OLSE, OLMS, and OLRM. Particular attention should be given to the OLSE of teachers with longer teaching years and rural teachers. Encouraging K-12 teachers with lower education qualifications to pursue further education is also recommended. This study provides evidence and a novel perspective for teacher educators to develop online professional development programs, which is significant for optimizing online learning experiences and enhancing educational outcomes. [ABSTRACT FROM AUTHOR]

Copyright of Food & Fermentation Industries is the property of Food & Fermentation Industries and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Objective: To investigate the value of 7E instructional model in the teaching of nursing students in nursing clinical probation. Methods: One hundred and fifty-nine nursing undergraduates of the class of 2018 who were on probation in our hospital from November 2020 to June 2021 were selected as the control group by convenience sampling, and the traditional teaching model was adopted; an additional 171 nursing undergraduates of the class of 2019 who were on clinical probation in the same hospital from November 2021 to June 2022 were selected as the experimental group, and the 7E instructional model was adopted, including elicit, engagement, exploration, explain, elaborate, evaluate and extend. Results: After the intervention, the experimental group was higher than the control group in theoretical knowledge scores (59.47 ± 4.51 VS 54.11 ± 8.46), practical skills scores (19.62 ± 1.36 VS 19.14 ± 1.24) and total course scores (90.13 ± 2.98 VS 84.67 ± 4.47); total scores of learning motivation (30.57 ± 5.24 VS 29.41 ± 4.77), learning cooperation ability (20.35 ± 4.41 VS 19.02 ± 4.25), information literacy (23.14 ± 4.36 VS 21.12 ± 3.95) and self-regulated learning ability (114.00 ± 19.72 VS 109.07 ± 18.61); total scores of viewpoint acquisition (43.58 ± 5.21 VS 42.16 ± 5.10), emotional care (33.41 ± 4.54 VS 32.09 ± 4.47), transposition thinking (13.04 ± 2.97 VS 11.56 ± 2.43) and empathy ability (89.03 ± 13.87 VS 85.81 ± 13.55); and total scores of positive attitudes toward communication skills (55.39 ± 7.03 VS 51.54 ± 6.54), with a statistically significant difference (P < 0.05). Conclusion: The 7E instructional model contributes to improved course grades, self-regulated learning and empathy, and positive attitudes toward communication skills among nursing students in their nursing clinical probation. [ABSTRACT FROM AUTHOR]

SUMMARY The use of deep learning techniques to identify grape leaf diseases relies on large, high‐quality datasets. However, a large number of images occupy more computing resources and are prone to pattern collapse during training. In this paper, a depth‐separable multifeature generative adversarial network (DMFGAN) was proposed to enhance grape leaf disease data. First, a multifeature extraction block (MFEB) based on the four‐channel feature fusion strategy is designed to improve the quality of the generated image and avoid the problem of poor feature learning ability of the adversarial generation network caused by the single‐channel feature extraction method. Second, a depth‐based D‐discriminator is designed to improve the discriminator capability and reduce the number of model parameters. Third, SeLU activation function was substituted for DCGAN activation function to overcome the problem that DCGAN activation function was not enough to fit grape leaf disease image data. Finally, an MFLoss function with a gradient penalty term is proposed to reduce the mode collapse during the training of generative adversarial networks. By comparing the visual indicators and evaluation indicators of the images generated by different models, and using the recognition network to verify the enhanced grape disease data, the results show that the method is effective in enhancing grape leaf disease data. Under the same experimental conditions, DMFGAN generates higher quality and more diverse images with fewer parameters than other generative adversarial networks. The mode breakdown times of generative adversarial networks in training process are reduced, which is more effective in practical application. [ABSTRACT FROM AUTHOR]

Investigators in neuroscience have turned to Big Data to address replication and reliability issues by increasing sample size. These efforts unveil new questions about how to integrate data across distinct sources and instruments. The goal of this study was to link scores across common auditory verbal learning tasks (AVLTs). This international secondary analysis aggregated multisite raw data for AVLTs across 53 studies totaling 10,505 individuals. Using the ComBat-GAM algorithm, we isolated and removed the component of memory scores associated with site effects while preserving instrumental effects. After adjustment, a continuous item response theory model used multiple memory items of varying difficulty to estimate each individual's latent verbal learning ability on a single scale. Equivalent raw scores across AVLTs were then found by linking individuals through the ability scale. Harmonization reduced total cross-site score variance by 37% while preserving meaningful memory effects. Age had the largest impact on scores overall (− 11.4%), while race/ethnicity variable was not significant (p > 0.05). The resulting tools were validated on dually administered tests. The conversion tool is available online so researchers and clinicians can convert memory scores across instruments. This work demonstrates that global harmonization initiatives can address reproducibility challenges across the behavioral sciences. [ABSTRACT FROM AUTHOR]

Mobile technologies are expected to provide many benefits for teacher reflection during teaching practice. However, empirical studies on these outcomes are still scarce. In an experimental field trial, we tested the effects of different variants of a mobile portfolio app on the development of pre-service teachers’ self-reported reflective ability, their orientations to learning to teach (e.g. developing ideas through discussion), and the interaction between their use of the mobile portfolio and orientation with regard to the development of reflective ability. Although there were only small direct effects of using the mobile app to foster reflective ability, the use of the app seemed to enhance the willingness to develop ideas for teaching practice through discussion and communication with mentors. These orientations were positively correlated with reflective ability; however, they showed no moderating effect on reflective ability when using the app. [ABSTRACT FROM AUTHOR]

In research on building a one-shot learning neural network without pre-training using mass data, the limitation on the information obtained from a single training sample downgrades the performance of the network. In order to improve performance and take full advantage of the support set, in this study, we design three kinds of shadow nodes and propose a structure-based training method for a correlation-coefficient-based neural network. This training strategy focuses on branches that are not activated or inactivated as expected. In contrast to existing networks that optimize the parameters using back-propagation, the training method proposed in this paper optimizes the structure of the correlation-coefficient-based network by correcting its pixel errors. For the shadow nodes and training process based on this strategy, the intersection over union (IOU) of a detected target increases by 4.83% in the experiments when using the Fashion-Mnist dataset, increases by 4.02% when using the Omniglot dataset, and increases by 3.89% when using the Cifar-10 dataset. The samples in category "7" wrongly classified as "1" decreased by 27.32% when using the Mnist dataset after training. This training strategy, along with shadow nodes, makes the correlation-coefficient-based network a more practical model and enables the network to develop during the accumulation of reliable samples, thus making it more suitable for simple target detection projects that collect samples over time. Moreover, the shadow nodes and training method proposed in this paper supplement the non-gradient-based parameter-gaining strategy. Additionally, it is a new attempt to explore the imitation of a human's ability to learn a new pattern from a low number of references. [ABSTRACT FROM AUTHOR]

Ocean deoxygenation and standing levels of hypoxia are shrinking fundamental niches, particularly in coastal areas, yet documented repercussions on species development and behaviour are limited. Here, we tackled the impacts of deoxygenation (7 mg O2 l−1), mild hypoxia (nocturnal 5 mg O2 l−1) and severe hypoxia (nocturnal 2 mg O2 l−1) on cuttlefish (Sepia officinalis) development (hatching success, development time, mantle length), cognition (ability to learn individually and socially) and behaviour (ability to camouflage and to explore its surroundings spatially). We found that hypoxia yielded lower survival rates, smaller body sizes and inhibited predatory (increased latency to attack the prey) and anti-predator (camouflage) behaviours. Acute and chronic exposure to low oxygen produced similar effects on cognition (inability to socially learn, increased open-field activity levels, no changes in thigmotaxis). It is thus expected that, although cuttlefish can withstand oxygen limitation to a certain degree, expanding hypoxic zones will diminish current habitat suitability. [ABSTRACT FROM AUTHOR]

Identifying the ash accumulation generated on the water-cooled walls of the waste incinerator is essential for the cleanup by the robotic arm. This paper improves a new algorithm based on YOLOv8n, which can identify the ash accumulation position on the water-cooled wall quickly and accurately. Firstly, the multi-scale fusion image enhancement algorithm is used to improve the sharpness and contrast of the image and enrich the details of the image. Secondly, the backbone feature extraction network of YOLOv8n is replaced by Mobilenetv3 network, which reduces the parameters in the model greatly. Finally, this paper improves a new attention mechanism AWGAM (Add Weight Global Attention Mechanism) based on GAM (Global Attention Mechanism), which can better integrate the feature information between different dimensions and improve the learning ability of the model. AWGAM is added to the backbone of the model. The experimental results show that compared with the original YOLOv8n model, the improved YOLOv8n model has 59.9% fewer parameters, 4.4% higher precision, 8.8% higher recall, 3.2% higher mAP50 (mean Average Precision) and 8.8% higher mAP50-95. This model has made remarkable progress on the basis of the original algorithm, and has strong competitiveness compared with other advanced target detection models. The lightweight and high accuracy of ash accumulation detection offered by the proposed model presents promising applications in ash accumulation detection tasks of water-cooled walls. [ABSTRACT FROM AUTHOR]

AbstractAs the domain of artificial intelligence has advanced, the integration of deep learning techniques into terrain and landform analysis has become more prevalent. Nevertheless, many existing methods are fully supervised and designed for specific tasks; thus, their transferability is limited and massive annotated samples are required. This study introduces a geomorphic pretrained model (GeomorPM) capable of performing multiple tasks. First, an architecture was designed that combined a convolution-based Vector Quantised-Variational Autoencoder (VQVAE) with a Transformer-based masked autoencoder (MAE) framework, allowing it to autonomously learn local details and global patterns from large-scale digital elevation model (DEM) data. Subsequently, GeomorPM, based on the VQMAE architecture, was pretrained on massive DEM data and fine-tuned for three specific tasks: DEM void filling, DEM superresolution, and landform classification. GeomorPM outperformed the traditional and other deep learning methods in all three tasks, demonstrating the superior learning ability and transferability of the model. This study provides a practical framework for developing pretrained models based on DEMs that can be expanded to other continuous geoscientific data. [ABSTRACT FROM AUTHOR]

The use of virtual reality (VR) training in medical school education has increased over the last decade partly due to coronavirus-19 (COVID-19) and due to technology advancements in this integrative area. As medical students are being trained in new pedagogical interventions, it is important to drill-down and focus on the areas of most translational concern for assessing their competency and proficiency. Thus, the present pilot study sought to evaluate a group of fourth-year medical students' ability to learn a modified medical school curriculum on implementing a subdermal drain by first learning how to stitch on a silicone model (i.e., proper stitching only) and then being randomly assigned to either a third-person point-of-view (TP-POV) or first-person point-of-view (FP-POV) immersive VR group to learn how to conduct the surgical procedure (i.e., conceptually and procedurally). Then, the participants' surgical procedural motor skills were assessed on a single attempt to demonstrate their competency and proficiency on a cadaver. The results showed that the pedagogical learning differences in FP-POV immersive VR improved the cleanliness and accuracy of the surgical dissection movement and drain placement with 100% proficiency in the first attempt at implementing this surgical technique on a cadaver. Additionally, there was statistically significant improvement in the dissection movement (p < 0.05*) and the tension test (p < 0.01**) in the FP-POV group compared to the TP-POV group, even with a small sample size in the present pilot study. These findings offer unique training opportunities to be considered when designing future VR curriculum with the suggested use of repeated-measures designs on the surgical procedural motor skills to assess learning curves and their potential translations in reducing operating room errors, costs, planning and scheduling times, and surgery overlaps, thereby positively contributing to a more efficient and safer OR surgical experience by staff and patients in the near future. [ABSTRACT FROM AUTHOR]

Instance segmentation of Real-time video is a crucial step in the object identification and classification process. Object detection is the task of finding different types of information about the object in a video by masking and bounding a rectangular box on the object's position in the image. Deep learning advances in the field of object identification by utilizing its excellent feature learning ability. Numerous researchers have employed various deep-learning methods to perform object detection with the goal of improving the precision of feature extraction. Due to the poor extraction features of the video frame, the higher and lower-level features of an object from the video frame are not extracted properly. Hence, the Feature Pyramid Network (FPN) integrated Modified Mask RCNN based on Inception-ResNetV2 is employed to extract the higher and lower level features from the video in order to solve this problem. In this designed model, the video dataset is converted to frames, and the features from the lower and higher level of the video frame are extracted using the FPN and backbone (Inception-ResNetV2) of the designed model. The automatic selective approach of the regions for the detection of an object is made by using the Regional Proposal Network, and the selected region is aligned using Region of Interest. From the aligned image, the fully convoluted layer is used for boxing and class detection of the object. Then, the convoluted layers are used for masking the detected object. In order to evaluate object detection on real-time video using Modified Mask RCNN, the performance metrics such as Accuracy, Precision, and Recall attained by the proposed model for the CoCo dataset are 0.98, 0.93, 0.94, which results in better values than the existing approaches including RCNN, SWINV2-G, Mask RCNN, SWINV2-L, and Fast RCNN. As a result, the developed model accurately and rapidly differentiates the object from the real-time video. [ABSTRACT FROM AUTHOR]

This research aimed to analyze the influence of teacher professional competence on teacher performance, and workplace spirituality. The sample consisted of 182 high school teachers in Malang City, with a standard error of 7%. The analysis tool used was Intelligent PLS. This research shows that professional competence can facilitate teacher mastery of teaching materials and a spiritual attitude that supports the learning process. The results of this research can inform education managers about how to improve teachers' abilities in teaching and learning process activities and mastery of teaching materials. They can also be a policy consideration for the education department to develop teacher performance in Malang City. [ABSTRACT FROM AUTHOR]

It is well known that dogs are capable of following human verbal instructions. However, very little is known about the equivalent ability in cats. In this study, we used a switched stimuli task to examine whether cats rapidly form picture-word association, which is a fundamental ability for word learning. We presented cats with two meaningless picture-word combinations, in the habituation phase. Then, on half of the trials we switched the combination (switched condition), but the other half of the trials remained as before (non-switched condition). If cats rapidly form picture-word association, they were expected to look at the monitor for longer in the switched condition, reflecting detection of the change. We used human speech as stimuli in Exp.1, and mechanical sounds (electronic sounds) in Exp.2. Cats expressed detection of the switched combination in Exp.1, where human speech and objects were paired. However, in Exp.2 where non-social sounds and objects were paired, there was no statistical difference between switched and non-switched conditions, although there was a main effect of condition when the data from the two experiments were pooled. These results demonstrate that cats can rapidly form picture-word association. Further research should investigate whether domestication has played a role in this ability. [ABSTRACT FROM AUTHOR]

Background: Which mammals show vocal learning abilities, e.g., can learn new sounds, or learn to use sounds in new contexts? Vocal usage and comprehension learning are submodules of vocal learning. Specifically, vocal usage learning is the ability to learn to use a vocalization in a new context; vocal comprehension learning is the ability to comprehend a vocalization in a new context. Among mammals, harbor seals (Phoca vitulina) are good candidates to investigate vocal learning. Here, we test whether harbor seals are capable of vocal usage and comprehension learning. Results: We trained two harbor seals to (i) switch contexts from a visual to an auditory cue. In particular, the seals first produced two vocalization types in response to two hand signs; they then transitioned to producing these two vocalization types upon the presentation of two distinct sets of playbacks of their own vocalizations. We then (ii) exposed the seals to a combination of trained and novel vocalization stimuli. In a final experiment, (iii) we broadcasted only novel vocalizations of the two vocalization types to test whether seals could generalize from the trained set of stimuli to only novel items of a given vocal category. Both seals learned all tasks and took ≤ 16 sessions to succeed across all experiments. In particular, the seals showed contextual learning through switching the context from former visual to novel auditory cues, vocal matching and generalization. Finally, by responding to the played-back vocalizations with distinct vocalizations, the animals showed vocal comprehension learning. Conclusions: It has been suggested that harbor seals are vocal learners; however, to date, these observations had not been confirmed in controlled experiments. Here, through three experiments, we could show that harbor seals are capable of both vocal usage and comprehension learning. [ABSTRACT FROM AUTHOR]

Networks of interconnected neurons communicating through spiking signals offer the bedrock of neural computations. Our brain's spiking neural networks have the computational capacity to achieve complex pattern recognition and cognitive functions effortlessly. However, solving real-world problems with artificial spiking neural networks (SNNs) has proved to be difficult for a variety of reasons. Crucially, scaling SNNs to large networks and processing large-scale real-world datasets have been challenging, especially when compared to their non-spiking deep learning counterparts. The critical operation that is needed of SNNs is the ability to learn distributed representations from data and use these representations for perceptual, cognitive and memory operations. In this work, we introduce a novel SNN that performs unsupervised representation learning and associative memory operations leveraging Hebbian synaptic and activity- dependent structural plasticity coupled with neuron-units modelled as Poisson spike generators with sparse firing (~1Hz mean and ~100 Hz maximum firing rate). Crucially, the architecture of our model derives from the neocortical columnar organization and combines feed forward projections for learning hidden representations and recurrent projections for forming associative memories. We evaluated the model on properties relevant for attractor-based associative memories such as pattern completion, perceptual rivalry, distortion resistance, and prototype extraction. [ABSTRACT FROM AUTHOR]

Objective: According to Bronfenbrenner's bioecological model of human development, focus on exploring the mechanism of person characteristics (achievement motivation) in their own development (improvement of educational practice ability). Method: A survey was conducted on 1,225 pre-service teachers in Anhui Province, China, using the Achievement Motivation Scale, Professional Identity Scale, Learning Engagement Scale, and Educational Practice Ability Scale. Results: (1) Achievement motivation can significantly and positively predict preservice teachers' educational practice ability; (2) Achievement motivation can indirectly affect pre-service teachers' educational practice ability through the mediating effects of professional identity and learning engagement; (3) Professional identity and learning engagement play a chain mediated role in the impact of achievement motivation on pre-service teachers' educational practice ability. [ABSTRACT FROM AUTHOR]

Speaking ability is an important skill that must be possessed by students, especially in the field of tourism. This study aims to explore the effect of role play and learning motivation on the speaking ability of first-year students of the Rooms Division at Palembang Tourism Polytechnic. This research method used a factorial design with a population of first-year students of the Hospitality Department. The research sample consisted of 32 participants, consisting of 16 people in the Experiment Group and 16 people in the Control Group. Data were collected through questionnaires and two measurement instruments, then analyzed using Two-Way Analysis of Variance. The results showed that there was a significant effect of role play on students' speaking ability, both those with high and low learning motivation. In addition, conventional teaching techniques also had a significant effect on students' speaking ability, depending on their level of learning motivation. The use of role play proved to be more effective than the conventional teaching technique in improving students' speaking ability. This finding shows that there is an interaction between role play and learning motivation that affects students' speaking ability. Therefore, this finding emphasizes the importance of applying innovative methods and high motivation in the learning process to improve students' speaking ability. The implications of this study suggest that educators prioritize interactive and motivational teaching strategies to improve speaking skills, which are crucial for career readiness in the tourism industry. In addition, role play can be a key component in the curriculum that aims to improve communication skills at various levels of student motivation. [ABSTRACT FROM AUTHOR]

Depicting landmarks on mobile maps is an increasingly popular countermeasure to the negative effect that navigation aids have on spatial learning – landmarks guide visual attention and facilitate map-to-environment information matching. However, the most effective method to visualize landmarks on mobile map aids remains an open question. We conducted a real-world navigation study outdoors to evaluate the influence of realistic vs. abstract 3D landmark visualization styles on wayfinders' navigation performance, visual attention, and spatial learning. While navigating with realistic landmarks, low-spatial-ability wayfinders focused more on the landmarks in the environment and demonstrated improved knowledge of directions between landmarks. Our findings emphasize the importance of visual realism when enriching navigation aids with landmarks to guide attention and enhance spatial learning for low-spatial-ability wayfinders. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the learning abilities of linear recurrent neural networks (RNNs) trained using Gradient Descent. We present a theoretical guarantee demonstrating that these linear RNNs can effectively learn any stable linear dynamical system with polynomial complexity. Importantly, our derived generalization error bound is independent of the episode length. For any stable linear system with a transition matrix C characterized by a parameter ρ C related to the spectral radius, we prove that despite the non-convexity of the parameter optimization loss, a linear RNN can learn the system with polynomial sample and time complexity in 1 1 - ρ C , provided that the RNN has sufficient width. Notably, the required width of the hidden layers does not depend on the length of the input sequence. This work provides the first rigorous theoretical foundation for learning linear RNNs. Our findings suggest that linear RNNs are capable of efficiently learning complex dynamical systems, paving the way for further research into the learning capabilities of more general RNN architectures. [ABSTRACT FROM AUTHOR]

Objectives: In the land cover classification study from multi-source remote sensing images, domain adaptation method can align images or extracted image features from source and target images, thus improves the generalization ability of deep learning models and plays an important role in intelligent remote sensing image interpretation. Methods: A self-training domain adaptation method based on information entropy uncertainty estimation for pseudo label correction is proposed, its core is an entropy uncertainty loss function for land cover classification between cross source remote sensing images. First, a land cover classification model is pretrained on the source domain training set with ground truth, and is applied on the target domain images without ground truth labels to generate pseudo labels. Then, the pseudo labels are used to further train the model, the information entropy of the prediction result is calculated and used as the uncertainty estimation of the pseudo labels to further correct the pseudo labels with self-training, so as to obtain weights of the classification model more suitable for the target domain dataset. Finally, a cross domain classification experiment was conducted on three data sets, namely, the WHU building change detection data set, the ISPRS 2D semantic annotation competition data set, and the Wuhan land cover classification data set. Results: Experimental results show that:(1) The proposed method improved the mean intersection over union(mIoU) and overall accuracy(OA) of semantic segmentation network by 0.3%-3.1% and 1.2%-4.5%, respectively. (2) Compared with the traditional self-training method, the proposed method can improve the mIoU and OA by 0.1%-1.5%. (3) Compared with the most recent uncertainty estimation method based on Kullback-Leibler divergence, the proposed method can improve the mIoU and OA about 0.6% in average. Conclusions: The proposed method can further improve the performance of a trained segmentation model for the target domain images without the requirement of target labels. There is also no need to introduce additional modules or parameters on the existing segmentation model. [ABSTRACT FROM AUTHOR]

Simple Summary: This study explores the potential of honeybees to detect and differentiate the odor of invasive red imported fire ants (Solenopsis invicta) using olfactory conditioning. By training bees with deceased ants, the results showed that bees quickly learned to recognize ant odors, with stronger responses to higher odor intensities. Bees generalized well across different worker castes and female alates and could recognize live ants after being conditioned with deceased ones. Discrimination varied by species but improved with a latent inhibition procedure. The findings suggest that honeybees may serve as a valuable tool for S. invicta detection and surveillance. Invasive red imported fire ants (Solenopsis invicta) create a serious threat to public safety, agriculture, biodiversity, and the local economy, necessitating early detection and surveillance, which are currently time-consuming and dependent on the inspector's expertise. This study marks an initial investigation into the potential of honeybees (Apis mellifera) to detect and discriminate the odor of S. invicta through the olfactory conditioning of proboscis extension responses. Deceased S. invicta were used as conditioned stimuli to ensure relevance to non-infested areas. The results showed that the bees rapidly learned to respond to deceased ant odors, with response levels significantly increasing at higher odor intensities. Bees exhibited generalization across the odors of 25 minor workers, 21 median workers, 1 major worker, and 1 female alate. When conditioned with deceased ant odors, bees effectively recognized live ants, particularly when trained on a single minor worker. Discrimination abilities varied by species and were higher when S. invicta was paired with Polyrhachis dives and Nylanderia yaeyamensis, and lower with S. geminata, Pheidole rabo, and Pheidole fervens. Notably, discrimination improved significantly with the application of latent inhibition. These findings suggest that trained honeybees have the potential to detect S. invicta. Further refinement of this approach could enhance its effectiveness for detection and surveillance. [ABSTRACT FROM AUTHOR]

To address the difficulties in manual counting for the uneven distribution and severe occlusion of watermelons in natural environments, this study utilizes drones and smartphones to collect videos and images, combined with manual annotation to establish a dataset for Sanbai melons and Ningxia selenium sand melons. A watermelon video automatic counting method based on the YOLOv7-GCSF model and an improved DeepSORT algorithm is proposed. The lightweight YOLOv7 model with GhostConv is enhanced with GBS modules, G-ELAN modules, and G-SPPCSPC modules to increase the model’s detection speed. Some ELAN modules are replaced with the C2f module from YOLOv8 to reduce redundant information. The SimAM attention mechanism is introduced into the MP module of the feature fusion layer to construct the MP-SimAM module, which is used to enhance the model's feature extraction capability. The CIoU loss function is replaced with the fasterconverging, lower-loss Focal EIoU loss function to increase the model's convergence speed. In video tracking and counting, a mask collision line mechanism is proposed for more accurate counting of Sanbai melons and Ningxia selenium sand melons. The results show that in terms of object detection: the four improvements to the YOLOv7-GCSF model have all enhanced the model’s performance to some extent. Specifically, compared to the YOLOv7 model, the construction of the MP-SimAM module increased accuracy by 1.5 percentage points, indicating a greater focus on Sanbai melons and Ningxia selenium sand melons. The addition of GhostConv reduced the model size by 28.1MB, demonstrating that the construction of GBS, G-ELAN, and G-SPPCSPC modules effectively reduced the model size and improved detection speed. The incorporation of the C2f module reduced the model's floating-point operations (FLOPs) by 77.5 billion, indicating that the model has eliminated most of the redundant information. The addition of the Focal EIoU loss function significantly increased the model’s convergence speed, indicating further enhancement of the model's learning ability. The improved YOLOv7-GCSF model achieved an accuracy (P) of 94.2% and a mean average precision (mAP0.5) of 98.2%, which is 5.0, 2.3, 21.9, and 14.9 percentage points higher in accuracy and 3.7, 0.3, 4.6, and 9.3 percentage points higher in mean average precision compared to YOLOv5, YOLOv7, Faster RCNN, and SSD, respectively. In terms of model lightweighting, the YOLOv7-GCSF model has seen a decrease of 1.18M and 0.11M in the number of parameters compared to the YOLOv4-Ghostnet and YOLOv7-Slimneck models, respectively. Compared to the original YOLOv7, the YOLOv7-GCSF model has reduced the parameter count and model size by 0.57M and 18.88MB, respectively. In terms of object tracking: the improved DeepSORT multi-object tracking accuracy is 91.2%, and the multi-object tracking precision is 89.6%, which is 5.0 and 13.7 percentage points higher in tracking accuracy and 3.7 and 13.1 percentage points higher in tracking precision compared to Tracktor and SORT, respectively. Comparing the improved model with manual counting results, the determination coefficient for the counting results of Sanbai melons and Ningxia selenium sand melons is 0.93, the average counting accuracy is 96.3%, and the average absolute error is 0.77, indicating that the error between the improved model and manual counting is small. This approach, by enabling effective counting of watermelons in agricultural fields, provides a technical methodology for the forecasting of watermelon yields. [ABSTRACT FROM AUTHOR]

This paper presents a fault diagnosis method for a vacuum contactor using the generalized Stockwell transform (GST) of vibration signals. The objective is to solve the problem of low diagnostic performance efficiency caused by the inadequate feature extraction capability and the redundant pixels in the graph background. The proposed method is based on the time-frequency graph optimization technique and ShuffleNetV2 network. Firstly, vibration signals in different states are collected and converted into GST time-frequency graphs. Secondly, multi-resolution GST time-frequency graphs are generated to cover signal characteristics in all frequency bands by adjusting the GST Gaussian window width factor λ. The OTSU algorithm is then combined to crop the energy concentration area, and the size of these time-frequency graphs is optimized by 68.86%. Finally, considering the advantages of the channel split and channel shuffle methods, the ShuffleNetV2 network is adopted to improve the feature learning ability and identify fault categories. In this paper, the CKJ5-400/1140 vacuum contactor is taken as the test object. The fault recognition accuracy reaches 99.74%, and the single iteration time of model training is reduced by 19.42%. [ABSTRACT FROM AUTHOR]

Intelligent Personal Assistants (IPAs) are becoming more prevalent in daily and educational contexts, increasing the possibility of using them as learning partners that can provide more personalized and learner-centric learning opportunities. However, research has primarily focused on educational advantages that IPAs may provide, overlooking investigation on students' acceptance of IPAs. To address this gap, this study empirically explored factors influencing students' intention to use IPAs for learning. A conceptual research model was designed based on the technology acceptance model. Data were collected from 1044 South Korean students, who were experienced in using IPAs. Results were analyzed using structural equation modeling. The findings indicate that the proposed model sufficiently explains students' IPAs acceptance for their learning. In addition, students' computer self-efficacy and self-management of learning ability were prominent factors influencing their intention. Moreover, their trust toward IPA and feeling of social presence on IPAs were found to be significant factors to be considered as well. Based on these findings, future recommendations for fostering students' active use of IPAs for learning are discussed. Additionally, both theoretical and practical implications are highlighted. [ABSTRACT FROM AUTHOR]

Background: Cooperative learning provides an opportunity for the development and improvement of social skills by emphasizing the importance of social interaction in fostering both academic and social intelligence. This study examined the application of cooperative learning in teaching traditional Chinese culture at Chinese universities using a mixed methods approach. Methods: This study adopted a mixed methods approach inclusive quantitative phase involved a true experimental design with pre- and post-tests with control group. The study focused on Chinese university students, and a cluster sampling approach was utilized to select 60 participants. Quantitative assessments were conducted using performance tests, the Motivated Strategies for Learning Questionnaire (MSLQ) by Paul Pintrich, and the Social Skills Inventory (SSI) by Riggio, while qualitative insights were gathered through post-experiment interviews with 10 participants. The qualitative data in this study were collated and analyzed using NVivo, integrating interviews, group discussions, questionnaires, and more. Quantitative data were analyzed by mixed analysis of variance (ANOVA) method. Data were analyzed with SPSS software. Results: Mixed analysis revealed that main effect of time in performance score was found to be statistically significant [F(1) = 19.25, P < 0.001]. This indicates a significant change in scores from the pre-test to the post-test across both groups. Additionally, the main effect of the group was also found to be significant [F(1) = 5.81, P = 0.019], suggesting a notable difference in scores between the control and experimental groups. Regarding motivation, it is evident that the main effect of time had a significant impact [F(1) = 59.60, P < 0.001], signifying a substantial change in motivation levels from pre-test to post-test across both groups. In social skills, it is evident that the main effect of time had a significant impact [F(1) = 25.67, P < 0.001], indicating a considerable change in skill levels from pre-test to post-test across both groups. Conclusion: These findings underscore the potential of cooperative learning to mobilize students' enthusiasm for learning and enhance their independent learning ability, cooperative participation, and competitive consciousness, thus effectively improving the teaching quality of Chinese traditional culture in universities. [ABSTRACT FROM AUTHOR]

Image restoration in adverse weather conditions is a difficult task in computer vision. In this paper, we propose a novel transformer-based framework called GridFormer which serves as a backbone for image restoration under adverse weather conditions. GridFormer is designed in a grid structure using a residual dense transformer block, and it introduces two core designs. First, it uses an enhanced attention mechanism in the transformer layer. The mechanism includes stages of the sampler and compact self-attention to improve efficiency, and a local enhancement stage to strengthen local information. Second, we introduce a residual dense transformer block (RDTB) as the final GridFormer layer. This design further improves the network's ability to learn effective features from both preceding and current local features. The GridFormer framework achieves state-of-the-art results on five diverse image restoration tasks in adverse weather conditions, including image deraining, dehazing, deraining & dehazing, desnowing, and multi-weather restoration. The source code and pre-trained models will be released. [ABSTRACT FROM AUTHOR]

A growing number of international doctoral students choose to study in China, a non-traditional learning destination. However, relatively few studies have investigated these students' academic writing practices while undertaking their studies in China. This study draws upon Bourdieu's concepts of field, habitus, and capital, and the notion of global-national-local imbrications, to explore 20 international doctoral students' language-engagement experiences in a Chinese university. Our analysis found that English and Chinese co-exist to varying and sometimes shifting degrees in doctoral students' academic writing practices. As a result, some students potentially developed a hybrid, 'in-between', cosmopolitan habitus. Notably, however, other students felt disempowered in the Chinese HE 'sub-field', with its unique logics of practice. Students' experiences indicate that multifarious language practices potentially create a heavily hybridised 'sub-field' characterised by a multitude of imbricating global, national, and local influences, and highlight the need to ensure that the language of instruction is indeed oriented to student learning – a language for learning. [ABSTRACT FROM AUTHOR]

Many real-world problems have an uneven distribution of data over different classes. The imbalanced nature of data impacts the performance of classifiers. The higher counts of majority class samples influence the learning abilities of well-known classifiers. Genetic programming (GP) algorithm based on natural evolution also impacts if the data's nature is imbalanced. The fitness function plays a pivotal role and impacts almost each building block of the GP framework. GP with the standard fitness function produces under-fitted and biased classifiers. Therefore, this paper has proposed a new fitness function in GP to classify the imbalanced data. The proposed method is used to classify nine imbalanced problems: ABL-18, ABL-9-18, BAL, YEAST2, YEAST1, ABL-9, ION, WDBC, and SPECT. The imbalanced factor of benchmark problems varies from 99:1 to 59:41. The proposed method's performance is compared with K-Nearest-Neighbourhood (KNN) and the standard fitness function-based GP methods. The GP with newly proposed fitness function gives average AUC values for ABL-18(99:1), ABL-9-18(94:6), BAL(92:8), YEAST2(89:11), YEAST1(84:16), ABL-9(83:17), ION(64:36), WDBC(63:37), and SPECT(59:41) as 0.714, 0.812, 0.975, 0.916, 0.768, 0.654, 0.872, 0.939, and 0.704, respectively, which are higher than KNN and the standard fitness function-based GP methods. The result outcomes prove the superiority of the proposed method. [ABSTRACT FROM AUTHOR]