Virtual reality (VR) is a transformative technology for studying language processing and learning. It offers realistic and naturalistic environments while maintaining experimental control, supporting the exploration of social and sensorimotor cues that have often been overlooked in traditional psycholinguistic research. It can also be integrated with behavioral and neurocognitive measures to examine how learning processes influence learning outcomes. These VR characteristics have brought new opportunities and contributed to the expansion of theoretical frameworks on language processing and learning. Through reviewing recent empirical findings, we establish VR as a novel methodology for investigating how embodied and social experiences shape language processing and learning. We emphasize that realizing this potential fully requires real-time multimodal tracking of learning processes, systematic examination of individual differences and development of richer virtual environments and VR-suitable assessment tasks. This review provides methodological insights for future VR research on the neurocognitive bases of language processing and learning.

Bees, as vital pollinators, are currently confronting numerous survival threats. Balanced amino acid nutrient have a positive impact on helping bees to resist pathogens and pesticides. This study aims to explore the interaction between leucine (Leu) and isoleucine (Ile) on key physiological parameters of Apis mellifera ligustica (Apidae: Hymenoptera) colony. The experiment was designed into four distinct dietary groups, including control diet (CK): 6.8 g/kg Leu + 3.9 g/kg Ile; high Ile diet: 6.8 g/kg Leu + 21.9 g/kg Ile; high Leu diet: 43.8 g/kg Leu + 3.9 g/kg Ile; diet with both high L-I: 43.8 g/kg Leu + 21.9 g/kg Ile. The results showed that Leu and Ile, to a considerable extent, promote the growth and vitality of the bee colony, with Ile mitigating Leu-induced absorption inhibition. Leu and Ile synergistically reduced haemolymph triglyceride via upregulated activities of fatty acid synthase and acetyl-CoA carboxylase, and expression of Lipase 1 and adipokinetic hormone receptor compared with CK. Furthermore, L-I treatment increased key activities of immune enzymes like phenol oxidase, lysozyme, glucose oxidase, and altered expression of antimicrobial peptides (abaecin, apidaecin, and defensin 1), indicating coordinated immunomodulation. In conclusion, within the context of this study, 43.8 g/kg Leu and 21.9 g/kg Ile in the artificial pollen substitute feed prove to be beneficial for the nutrient deposition and immunity of individual bees. This work provides a framework for optimising artificial pollen diets, offering experimental guidance towards more effective strategies in sustainable beekeeping.

Conventional active ankle exoskeletons are often bulky and heavily reliant on external power sources. This study presents a lightweight and flexible passive ankle exoskeleton (LFPA-EXO) aimed at reducing metabolic cost of walking. The LFPA-EXO features a gait-adaptive clutch (GA-clutch) and a super-elastic composite booster (SC-booster). By matching the walking gait, it stores gravitational potential energy and converts it into elastic energy through the booster, thereby reducing the metabolic cost of human locomotion. Mechanical and biomechanical evaluations demonstrate that the GA-clutch achieves less than 5% interference and over 85% assistance, indicating that the LFPA-EXO operates within the natural ankle joint range of motion without disrupting normal gait patterns. It delivers a peak assisting moment of 24.56 Nm during normal walking. Notably, it decreases the activation of the soleus muscle while moderately reducing the activation of the gastrocnemius muscle, with minimal impact on the tibialis anterior muscle. The LFPA-EXO achieves a 12.22% reduction in metabolic cost and an 11.17% decrease in average heart rate, underscoring its effectiveness in reducing metabolic cost during walking.

Radio recombination line (RRL) maser is a useful tool to study massive star formation regions with ionised gas close to new born massive stars. Masers often show sharp line profiles and/or extreme narrow widths, and high brightness temperatures. However, RRL masers were rarely detected only in several sources. Here we report the detection of sharp line profiles of the RRL H29$\alpha$, which can be interpreted as maser candidates, in two sources within W49A, a mini-starburst region in our Galaxy. These observations, conducted with high resolution ($\sim0.03''$) using the Atacama Large Millimeter/sub-millimeter Array (ALMA), reveal high brightness temperatures up to $\sim$9 000 K for H29$\alpha$ emission in another two sources, which might also be regarded as maser candidates. Additionally, suggestions for efficiently identifying RRL maser candidates are also provided.

Unmet healthcare needs remain a major barrier to achieving universal health coverage (UHC) globally. The intersection of aging and disability intensifies individual vulnerability and deepens structural health inequalities. Using Andersen’s Behavioural Model of healthcare utilisation as the theoretical framework, this study examines the determinants of healthcare utilisation at the individual and contextual levels among older adults with disabilities living in China. We use a dataset in China from 319 prefectures, with a total sample size of 634,445 individuals. Our findings reveal the presence of higher-income and urban-residence advantages in healthcare utilisation for older adults with disabilities in China. Regional economic development positively affects healthcare utilisation and reduces urban-rural inequality in healthcare utilisation, but its impact on income-based inequality is limited. These results highlight the need for targeted social assistance programmes for low-income groups to promote universal healthcare coverage and social equity.

The emergence of large language models, exemplified by ChatGPT, has garnered growing attention for their potential to generate feedback in second language writing, particularly automated written corrective feedback (AWCF). In this study, we examined how prompt design – a generic prompt and two domain-specific prompts (zero-shot and one-shot) enriched with comprehensive domain knowledge about written corrective feedback (WCF) – influences ChatGPT’s ability to provide AWCF. The accuracy and coverage of ChatGPT’s feedback across these three prompts were benchmarked against Grammarly, a widely used traditional automated writing evaluation (AWE) tool. We find that ChatGPT’s ability in flagging language errors grew considerably with prompt sophistication driven by the integration of domain-specific knowledge and examples. While the generic prompt resulted in substantially lower performance than Grammarly, the zero-shot prompt achieved comparable results to it and the one-shot prompt surpassed it considerably in error detection. Notably, the most pronounced improvement in ChatGPT’s performance was observed in its detection of frequent error categories, including those of word choice or expression, direct translation, sentence structure and pronoun. Nonetheless, even with the most sophisticated prompt, ChatGPT still displayed certain limitations when compared to Grammarly. Our study has both theoretical and practical implications. Theoretically, it lends empirical evidence to Knoth et al.’s (2024) proposition to separate domain-specific AI literacy from generic AI literacy. Practically, it sheds light on the pedagogical application and technical development of AWE systems.