70015 research outputs found
Sort by
Predictors of frequency and success of wild meat hunting trips and carcass prices in an African biodiversity hotspot
Get PDFHunting wild animals for food and income, which is pervasive across tropical regions, drives biodiversity loss. Interventions to promote sustainable wild meat harvesting require information on hunter behavior. Here we monitored the hunting activities of 33 hunters in SE Nigeria over three years (1,106 hunter-months) to identify correlates of (a) the probability of initiating a hunting trip on any given day; (b) trip success – whether an animal was caught, and if so, how many; and (c) carcass price. We found a higher probability of initiating a trip during periods with bright moon phases and in peak agriculture season. Hunters were more likely to catch at least one animal when there was less rainfall and on shorter hunting trips. However, among successful trips, the number of animals caught increased with trip duration. Taken together, these results suggest hunters set themselves a minimum target of not returning empty-handed rather than optimally adjusting their hunting effort. Lastly, the carcass price per kilogram of a species’ meat increased with its palatability but decreased with mass, with the fall in price observed to be greater for rarely caught, smaller-bodied animals than more frequently caught animals. Our results provide deeper insights into the behavioral plasticity of wild meat hunters
Analysis of the Sustainable Development Pathway of Urban–Rural Integration from the Perspective of Spatial Planning: A Case Study of the Urban–Rural Fringe of Beijing
Get PDFThis study employs various comprehensive research methods to thoroughly analyze the relationship between urban–rural integration and sustainable development, proposing corresponding optimization pathways. First, a literature review method systematically examines existing theories on urban–rural integration and sustainable development. Then, it identifies the main problems and challenges in the current process of urban–rural integration, thereby laying a theoretical foundation for the study. Second, a case study approach is adopted, selecting Caoqiao Village in Fengtai District, Zhenggezhuang Village in Changping District, and Xihoujie Village in Majuqiao Town, Tongzhou District of Beijing, as typical cases. These cases are analyzed in depth to explore their implementation outcomes and validate the practical results of different development pathways. Subsequently, based on specific data from Beijing’s urban–rural fringe, this study utilizes data analysis methods to conduct an in-depth examination of land use changes, ecological environment status, and influencing factors, with a focus on analyzing relevant data from 2009 to 2023. This analysis reveals the dynamic relationship between urban–rural integration and sustainable development. Regression analysis is adopted to quantify the effect of urban–rural integration on sustainable development, thus exploring the correlation between urban–rural integration, spatial planning, economic development, financial development, and sustainable development. Finally, targeted management recommendations and policy optimization plans are proposed based on the principles of ecological protection red lines and urban development boundaries. The results indicate a significant positive correlation between urban–rural integration and sustainable development levels, with a regression coefficient of 0.48, demonstrating its role in promoting sustainable development. The levels of spatial planning and economic development also positively affect sustainable development, with coefficients of 0.32 and 0.27, respectively. Moreover, financial development and social investment levels show a certain positive relationship. It is noteworthy that although the correlation between foreign trade and sustainable development levels is the lowest, the interconnections between other variables further emphasize the key position of urban–rural integration in overall sustainable development. This study offers a theoretical basis and empirical support for spatial planning in the urban–rural fringe of Beijing, ecological environment protection, and scientific policy formulation, thus advancing sustainable urban development
Investigating The usability of iconography via brain and key-press responses
Get PDFThe purpose of this research is to better understand the circumstances in which icons are most effective in use cases that pertain to memory. To achieve this, the project aimed to investigate the research question: "How do the pre-existing icon classifications discussed by Arledge and Nielsen impact memory performance?". The classifications in use are Solid and Outline & Resemblance and Abstract icons. Current research on the usability of these classifications suggests there is no clear superior classification in either family regarding the speed or accuracy in which they are identified. Despite this, current research regarding both the N-Back task and electroencephalogram (EEG) indicates that these tools remain effective in providing quantitative insights on memory load and the timing of memory recall. Twelve students at the University of Kent participated in an experiment where each participant conducted separate 1-back and 2-back tests (variants of the N-Back test) for each of the four permutations for both families of classifications (Solid Resemblance, Solid Abstract, Outline Resemblance, Outline Abstract). Eight tests were conducted by each participant. During each test, the reaction times of participants (in milliseconds) and the participant's EEG data was recorded in parallel. A 32-channel wireless brain-computer interface (BCI) was used to collect the EEG recordings. The statistical analysis conducted on the data provided by the N-Back tasks and ERPs employed the Wilcoxon signed rank test to test for significant differences between the stimuli classifications used. The comparisons made on the N-Back reaction times show that Arledge's two icon styles Solid and Outline failed to outperform each other significantly in terms of the speed at which they were recalled. Contrary to previous work favouring resemblance icons due to their suggested superior usability over abstract icons, comparisons showed that abstract icons were recalled significantly faster than resemblance icons, this observation was made when the two semantic groups were in an outline style. Dually, a comparison made showed an instance in which abstract icons in an outline style elicited a significantly smaller amplitude compared to resemblance icons in the same style, indicating an increased memory load compared to resemblance icons. User interface design principles of previous work, would suggest that Outline Resemblance stimuli had better usability than Outline Abstract stimuli, due to a significantly lower memory load. Despite these observations, overall, the ERP analysis did not reveal any significant distinction between the memory loads of the icon classifications used. To observe significant differences between the icon classifications used, this study suggests that future work explores the impact of various dimensions on memory performance
Beyond Echo Chambers and Rabbit Holes: Algorithmic Drifts and the Limits of the Online Safety Act, Digital Services Act, and AI Act
Get PDFThis paper uses Karen Barad's concepts of intra-action and diffraction to argue that dominant models of understanding algorithmic radicalisation (echo chambers, filter bubbles, rabbit holes) and how it is addressed in the UK and EU, are inadequate for deep neural network systems like YouTube. These models assume users seek out bad content and get stuck in reinforcing loops. However, these systems are far more dynamic and relational, constantly probing, drifting, experimenting, seeking rewards to keep users engaged. Consequently, current laws (e.g., Online Safety Act, Digital Services Act, and Artificial Intelligence Act) are ill-equipped to deal with this. They focus on removing 'bad' content and punishing bad actors, missing the deeper, systemic processes driving radicalisation even when no ‘bad’ content or actors are necessarily involved. While not claiming to have the answers to solve this complex problem, the paper argues for better questions that account for the complexity of these systems and their relational, intra-active, and diffractive nature
Rethinking EU Food Law: Addressing Slow Harm in the Age of Ultra-Processed Foods
Get PDFEU food law prioritises acute risks and informed choice but overlooks the chronic harms of ultra-processed foods. This post calls for a shift towards recognising ‘slow harm’, rebalancing regulatory duties, and embedding structural safeguards in food safety policy
Development and validation of the S-TIMHSS: A quality metric to inform and evaluate interventions to (re)build trust
Get PDFPublic acceptance of health messaging, recommendations, and policy is heavily dependent on the public’s trust in doctors, health systems and health policy. Any erosion of public trust in these domains is thus a concern for public health as it can no longer be assumed that the public will follow official health recommendations. In response, the health policy and health services communities have emphasized a commitment to (re)building trust in healthcare. As such, measures of trust that can be used to develop and evaluate interventions to (re)build trust are highly valuable. In 2024, the Trust in Multidimensional Health System Scale (TIMHSS) was published, providing the first measure of trust in healthcare that includes doctors, the system and health policy within a single measure. This measure can effectively facilitate research on trust across diverse populations. However, it is limited in its application because results cannot be directly added together for a total trust score. Further, at 38-items, it is burdensome for respondents and analysts, particularly when being used as a repeat measure in an applied setting. The aim of the present work was to develop a shortened measure of trust in healthcare for use in applied settings. Survey data were collected (N=512; in Sept 2024) to reduce the number of items and to test if the factor structure was consistent with the original TIMHSS. Several statistical criteria were used to support item reduction (i.e., correlated errors, measurement invariance, inter-item correlations, factor loadings and communalities, item-total correlation, and skewness), as well as an exercise testing the content validity ratio (CVR). We then tested a three-factor model based on the 18 items that remained following the CVR and statistical test metrices to finalize the measure. The result is the S-TIMHSS, an 18-item scale that allows for direct scoring of trust items for applied research. We recommend the measure be used by health policy makers and practitioners as a quality metric to inform and evaluate interventions which aim to (re)build trust in doctors, health systems and health policy
Exploring the Cybercrime Potential of LLMs: A Focus on Phishing and Malware Generation
Get PDFLanguage Large Models (LLMs) are revolutionizing various sectors by automating complex tasks, enhancing productivity, and fostering innovation. From generating human-like text to facilitating advanced research, LLMs are increasingly becoming integral to societal advancements. However, the same capabilities that make LLMs so valuable also pose significant cybersecurity threats. Malicious actors can exploit these models to create sophisticated phishing emails, deceptive websites, and malware, which could lead to substantial security breaches. In response to these challenges, our paper introduces a comprehensive framework to assess the robustness of six leading LLMs (Gemini API, Gemini Web, GPT-4o API, GPT-4o Web, Llama 3 70B, and Mixtral 8x7B) against both direct and elaborate malicious prompts to generate phishing and malware attacks. This framework not only measures the ability – or the lack thereof – of LLMs to resist being manipulated into performing harmful actions, but also provides insights into enhancing their security features to safeguard against such prompt injection attempts. Our findings reveal that even direct prompt injections can successfully compel all tested LLMs to generate phishing emails, websites, and malware. This issue becomes particularly pronounced with elaborate malicious prompts, which achieve high rates of malicious compliance, especially in scenarios involving phishing. Specifically, models such as Llama 3 70B, Gemini API, and Gemini Web show high compliance in generating convincing phishing content under elaborate instructions, while GPT-4o models (both the API and Web versions) excel in creating phishing webpages even when presented with direct prompts. Finally, local models demonstrate nearly perfect compliance with malware generation prompts, underscoring the critical need for sophisticated detection methods and enhanced security protocols tailored to mitigate such elaborate threats. Our findings contribute to the ongoing discussion about ensuring the ethical use of Artificial Intelligence (AI) technologies, particularly in cybersecurity contexts
How does counterfactual imagination affect our ability to remember the past? EEG and behavioural evidence
No full textVivid imagination is a known source of false memories. However, often when we imagine the past we do not create new memories but instead we imagine counterfactual versions of past events and consider how our memories could have been different. Flexible and reconstructive episodic memory processes have many adaptive purposes; however, they are also known to leave the original memory susceptible to modification and updating. Previous studies have shown that counterfactual reimagined versions of the past can be mistaken for memories of real events but only a limited number of studies have considered how counterfactual imagination may affect our ability to remember the veridical version of what really happened. This research question was therefore the focus of the current thesis. In an initial series of behavioural experiments, I investigated how counterfactual imagination could induce memory errors in the context of eyewitness testimony. In subsequent experiments with combined EEG and behavioural measures, I considered more commonplace memory decisions in our everyday life; when we attempt to remember if we performed simple actions, such as whether we locked our front door or only imagined this action. The findings across these experiments indicate that counterfactual imagination could either strengthen or impair our memory of the past depending on different contextual factors. I also used EEG to investigate the neurocognitive mechanisms that are engaged when we make these kinds of decisions during memory retrieval. Analysis of both ERPs and oscillatory EEG power showed separable effects related to the reactivation of the memory and subsequent post-retrieval processes. Lastly, I conducted representational similarity analysis on the EEG data to further explore the processes engaged when recalling a memory associated with an imagined counterfactual. Taken together, the research in this thesis provides a behavioural and neurocognitive explanation for how counterfactual imagination affects our ability to remember the past
The use of individualised, media based sleep hygiene education for professional female footballers
Get PDFSleep hygiene can be defined as practicing habits that facilitate sleep; poor sleep hygiene is common among elite athletes, and improving this can be one way to enhance sleep indices. Given the large inter-individual variability of sleep, there is a need for further investigation into individualised sleep hygiene for elite female athletes, with consideration for the practical application of the method. Using a self-controlled time series design with repeated measures, n = 16 professional female footballers completed a 9-week study during mid-season. Monitoring of sleep (actigraphy, self-report) occurred at week 1, 4, 7 and 9—a control period occurred at week 2 and 3, and a subsequent intervention period occurred at weeks 5 and 6. Based on baseline sleep monitoring, media-based messages were designed with the purpose of giving a singular sleep hygiene message; all participants received these individualised messages daily across the 2-week intervention period at a standardised time of 8.00 p.m., with the intention of them actioning the sleep hygiene point. One-way analysis of variance with repeated measures was conducted to assess the differences between control period, intervention period and follow-up for each measured variable. Significant differences were observed post-intervention for sleep efficiency (p < 0.001) and sleep latency (p < 0.001), whereas the athlete sleep behaviour questionnaire score significantly improved in the follow-up period (week 9) post intervention (p = 0.039). This is the first study to present this novel method of individualised sleep hygiene education for elite female athletes and is also the first study to demonstrate the use of sleep hygiene interventions to improve sleep factors for female athletes' mid-season. This demonstrates a promising, time-efficient approach to sleep hygiene education, with a potentially wide scope of application, as well as demonstrating there is indeed potential for elite female athletes to gain sleep improvements mid-season
A Greedy Global Framework for Lattice Reduction Using Deep Insertions
Get PDFLLL-style lattice reduction algorithms iteratively employ size reduction and reordering on ordered basis vectors to find progressively shorter, more orthogonal vectors. DeepLLL reorders the basis through deep insertions, yielding much shorter vectors than LLL. DeepLLL was introduced alongside BKZ, however, the latter has received greater attention and has emerged as the state-of-the-art. We first show that LLL-style algorithms work with a designated measure of basis quality and iteratively improves it; specifically, DeepLLL improves a sublattice measure based on the generalised Lovász condition. We then introduce a new generic framework X-GG for lattice reduction algorithms that work with a measure X of basis quality. X-GG globally searches for deep insertions that minimise X in each iteration. We instantiate the framework with two quality measures – basis potential (Pot) and squared sum (SS) – both of which have corresponding DeepLLL algorithms. We prove polynomial runtimes for our X-GG algorithms and also prove their output to be X-DeepLLL reduced. Our experiments on non-preprocessed bases show that X-GG produces better quality outputs whilst being much faster than the corresponding DeepLLL algorithms. We also compare SS-GG and the FPLLL implementation of BKZ with LLL-preprocessed bases. In small dimensions (40 to 210), SS-GG is significantly faster than BKZ with block sizes 8 to 12, while simultaneously also providing better output quality in most cases. In higher dimensions (250 and beyond), by varying the threshold for deep insertion, SS-GG offers new trade-offs between the output quality and runtime. On the one hand, it provides significantly better runtime than BKZ-5 with worse output quality; on the other hand, it is significantly faster than BKZ-21 while providing increasingly better output quality after around dimension 350
Access Repository Dashboard
Do you manage Kent Academic Repository? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!