Many practitioners, such as pilots, frequently face working memory (WM) demands under acute stress environments, while the effect of acute stress on WM has not been conclusively studied because it is moderated by a variety of factors. The current study investigated how acute stress affects pilots' WM under different memory load conditions. There are 42 pilots conducting the experiments, consisting of 21 stress group participants experiencing the Trier Social Stress Test (TSST) and 21 control group participants experiencing the controlled TSST (C-TSST). Subsequently, both groups performed N-back tasks under three memory load conditions (0-back, 1-back, and 2-back). State Anxiety Inventory (S-AI), heart rate (HR), and salivary cortisol concentrations (SCC) were collected to analyze acute stress induction. The results revealed that (1) the TSST could effectively induce acute stress with higher S-AI, HR, and SCC; (2) higher memory load reduces WM accuracy (ACC) and delays response times (RT); (3) acute stress increases WM ACC under moderate load conditions (1-back task). These results suggest that acute stress may not necessarily impair WM and even improve WM performance under certain memory load conditions. Potential mechanisms of acute stress effects on WM and alternative explanations for the modulatory role of memory load consistent with the emotion and motivation regulation theory are discussed. These findings not only provide insight into the field of acute stress and WM but are also beneficial for pilot training and the development of stress management strategies.
The regulation of fatty acid metabolism is crucial for milk flavor and quality. Therefore, it is important to explore the genes that play a role in fatty acid metabolism and their mechanisms of action. The RNA-binding protein Musashi2 (MSI2) is involved in the regulation of numerous biological processes and plays a regulatory role in post-transcriptional translation. However, its role in the mammary glands of dairy cows has not been reported. The present study examined MSI2 expression in mammary glands from lactating and dry milk cows. Experimental results in bovine mammary epithelial cells (BMECs) showed that MSI2 was negatively correlated with the ability to synthesize milk fat and that MSI2 decreased the content of unsaturated fatty acids (UFAs) in BMECs. Silencing of Msi2 increased triglyceride accumulation in BMECs and increased the proportion of UFAs. MSI2 affects TAG synthesis and milk fat synthesis by regulating fatty acid synthase (FASN). In addition, RNA immunoprecipitation experiments in BMECs demonstrated for the first time that MSI2 can bind to the 3′-UTR of FASN mRNA to exert a regulatory effect. In conclusion, MSI2 affects milk fat synthesis and fatty acid metabolism by regulating the triglyceride synthesis and UFA content through binding FASN.
Feature description has an important role in image matching and is widely used for a variety of computer vision applications. As an efficient synthetic basis feature descriptor, SYnthetic BAsis (SYBA) requires low computational complexity and provides accurate matching results. However, the number of matched feature points generated by SYBA suffers from large image scaling and rotation variations. In this paper, we improve SYBA’s scale and rotation invariance by adding an efficient pre-processing operation. The proposed algorithm, SR-SYBA, represents the scale of the feature region with the location of maximum gradient response along the radial direction in Log-polar coordinate system. Based on this scale representation, it normalizes all feature regions to the same reference scale to provide scale invariance. The orientation of the feature region is represented as the orientation of the vector from the center of the feature region to its intensity centroid. Based on this orientation representation, all feature regions are rotated to the same reference orientation to provide rotation invariance. The original SYBA descriptor is then applied to the scale and orientation normalized feature regions for description and matching. Experiment results show that SR-SYBA greatly improves SYBA for image matching applications with scaling and rotation variations. SR-SYBA obtains comparable or better performance in terms of matching rate compared to the mainstream algorithms while still maintains its advantages of using much less storage and simpler computations. SR-SYBA is applied to a vision-based measurement application to demonstrate its performance for image matching.
Pedestrian detection (P-D) is an significant part in the field of machine vision. The development of P-D is allong with the whole process of target detection. The P-D's objective is to accurately find the moving suspected people in the video, return its classification results. Due to the particularity of its detection targets, for the past few years, P-D technology has been broadly applied in monitoring equipment, unmanned driving, smart home systems and other fields, which has attracted widespread attention from the academic community and society. However, the human body itself is a non-rigid target, and when a limb movement occurs, the extracted target features are time-varying. This leads to the machine learning method based on hand-designed features, which has the disadvantages of robustness and low noise immunity.
Weakly-supervised text classification trains a classifier using the label name of each target class as the only supervision, which largely reduces human annotation efforts. Most existing methods first use the label names as static keyword-based features to generate pseudo labels, which are then used for final classifier training. While reasonable, such a commonly adopted framework suffers from two limitations: (1) keywords can have different meanings in different contexts and some text may not have any keyword, so keyword matching can induce noisy and inadequate pseudo labels; (2) the errors made in the pseudo label generation stage will directly propagate to the classifier training stage without a chance of being corrected. In this paper, we propose a new method, PIEClass, consisting of two modules: (1) a pseudo label acquisition module that uses zero-shot prompting of pre-trained language models (PLM) to get pseudo labels based on contextualized text understanding beyond static keyword matching, and (2) a noise-robust iterative ensemble training module that iteratively trains classifiers and updates pseudo labels by utilizing two PLM fine-tuning methods that regularize each other. Extensive experiments show that PIEClass achieves overall better performance than existing strong baselines on seven benchmark datasets and even achieves similar performance to fully-supervised classifiers on sentiment classification tasks.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)