University of St Andrews Research Portal
Not a member yet
81329 research outputs found
Sort by
Patterned ground
No full textPatterned ground in periglacial environments is developed through repeated freezing and thawing of moist, frost-susceptible soil, typically forming circles, polygons, irregular nets and stripes. Sorted patterns are defined by fine cells with coarse borders, and nonsorted patterns by vegetation or microrelief. Most patterns evolve though differential frost heave that develops a stable wavelength through self-regulating feedback processes and/or buoyancy-driven soil circulation. Large sorted patterns and nonsorted circles (frost boils) mainly form above permafrost; small sorted patterns are produced by needle-ice heave. Earth hummocks and nonsorted circles may be initiated by differential frost heave, though buoyancy-driven soil circulation and diapirism may contribute to their subsequent development
The individual and situational factors predicting unethical behaviour in the workplace:a direct and conceptual replication of Jones & Kavanagh (1996)
No full textIntentions to act unethically in the workplace are purported to be driven by a number of situational and individual factors. Across two seminal vignette experiments, Jones and Kavanagh reported inconsistent effect sizes for manager and peer influence and locus of control, consistent significant effects for work quality and Machiavellianism, and consistent non-significant effects for gender. Using an innovative multi-site collaboration, the current Registered Report represents a direct replication of these experiments (N = 2218), and adds a longitudinal conceptual replication capturing self-reported unethical work behaviour (N = 1747). Both replications found a consistent small effect of having a more external locus of control and male identity, and a consistent moderate effect of machiavellianism, for increasing unethical intentions and behaviour. The situational factors, whilst consistent in direction with that of the original study, varied more substantively in effect size. Our results highlight the value of multi-site collaborations and different replication types in developing conceptual, methodological, measurement and theoretical clarity to ensure future works can progress more rapidly to minimize the negative impacts of unethical workplace behaviour and improve individual’s working lives. All materials, code and data for this project can be found here: osf.io/d3arx.</p
Multi-parameter control for the (1+(λ,λ))-GA on OneMax via deep reinforcement learning
No full textIt is well known that evolutionary algorithms can benefit from dynamic choices of the key parameters that control their behavior, to adjust their search strategy to the different stages of the optimization process. A prominent example where dynamic parameter choices have shown a provable super-constant speed-up is the (1+(λ,λ)) Genetic Algorithm optimizing the OneMax function. While optimal parameter control policies result in linear expected running times, this is not possible with static parameter choices. This result has spurred a lot of interest in parameter control policies. However, many works, in particular theoretical running time analyses, focus on controlling one single parameter. Deriving policies for controlling multiple parameters remains very challenging. In this work we reconsider the problem of the (1+(λ,λ)) Genetic Algorithm optimizing OneMax. We decouple its four main parameters and investigate how well state-of-the-art deep reinforcement learning techniques can approximate good control policies. We show that although making deep reinforcement learning learn effectively is a challenging task, once it works, it is very powerful and is able to find policies that outperform all previously known control policies on the same benchmark. Based on the results found through reinforcement learning, we derive a simple control policy that consistently outperforms the default theory-recommended setting by 27% and the irace-tuned policy, the strongest existing control policy on this benchmark, by 13%, for all tested problem sizes up to 40,000
Gender, Ukraine and imagining a just, sustainable peace
No full textThis short article seeks to highlight gender blind spots and discuss ways to address them in the context of peace negotiation in bringing the war in Ukraine to a close. This shall be achieved specifically by focusing on women’s experiences during the war, in peace negotiation and mediation to argue that women’s experiences and voices matter for a just and sustainable peace in Ukraine. Moreover, different women combatants have different needs and priorities, especially as it is significant to recall that the positionality and experience of women in war is never monolithic. This involves reckoning with the gendered silences, ideas, and practices that undergird the peace negotiation, peace process, and post-conflict reconstruction and peacebuilding – dimensions that, so far, have been missing from the conversation about peace
Quaternary science for social justice? Reflections on fifteen years of research in Peruvian Amazonia
No full textThis paper presents my experience, as a Quaternary palaeoecologist, of developing an inter- and transdisciplinary research programme on the peatlands and local communities of northwest Peruvian Amazonia. It describes the development of my research from its beginnings as a traditional Quaternary palaeoecology project into a programme which engaged social scientists, conservation and development practitioners, and peatland communities. I discuss some of my learning points, particularly on the findings and benefits emerging from our exploration of local ecological knowledge alongside scientific perspectives. Overall, the paper explores the potential of engaged interdisciplinary research to bring benefits to both scientific and local/indigenous communities, and argues that Quaternary science, with its integrative tradition, can be well placed to lead such research programmes
Prioritising authenticity in live lecture comprehension pre-sessional assessment task design
No full textWhere pre-sessional assessments determine student progression, authenticity is crucial to deliver construct validity and support high-stakes progress decisions. However, replicating the natural processes that occur in academic environments poses challenges relating to domain representation and test-taker engagement as well as meeting stakeholder needs such as benchmarking against commercial tests. The current study reports on a project devised to develop an authentic lecture assessment for a pre-sessional programme, informed by a university-wide needs analysis. In the study, 66 students completed a live lecture comprehension task, and data were analysed using classical test theory and many facet Rasch measurement. In addition, students completed a questionnaire which was analysed using stepwise linear regression. Analysis of the lecture transcript was completed to evaluate the suitability of the content for the target level. Results showed that speech rate and vocabulary were appropriate for the assessment objectives, student self-reported comprehension accounted for some test score variance, and the test was reliable. However, many facet Rasch measurement showed the test was only able to categorise test takers into two distinct levels, which was problematic as the pre-sessional needed to multiple levels for schools to make admissions decisions. Recommendations for balancing these challenges in assessment design are presented
Classical Credentials: Women's Intellectual and Sexual Licence in Sixteenth-Century France
No full textNiall Oddy,<i>Writing Europe in Renaissance France: Travels in Reality and Imagination</i> (Edinburgh: Edinburgh University Press, 2024)
No full textTiming of calcification and environmental variability determine pH proxy fidelity in coastal calcifying macroalgae
No full textLong-lived calcifying marine biota are increasingly used as paleo-archives for reconstructing ocean pH. They enable exploration of the rate and magnitude of ocean acidification in shallow-water ecosystems serving as proxies for environmental pH reconstruction. However, shallow water systems often have highly variable carbonate chemistry, and the impact of this on the accuracy of pH reconstructions from long-lived marine calcifiers is not known. In particular, a better understanding of the timing of calcification with respect to environmental pH cyclicity is needed. To test the fidelity of coastal environmental pH proxies, we assessed the synchronicity between calcification and in situ diel carbonate chemistry in a tropical (One Tree Island, Great Barrier Reef, Australia) and a temperate (Loch Sween, Scotland) location using calcifying macroalgae (rhodolith-forming coralline algae) as a model system. Calcification occurred primarily during daylight hours, meaning a recording bias was introduced when compared to the full diel pH range (< 0.02 pH units). This bias resulted in pH offsets up to 0.043 pH units over the period 1860–2020, representing up to 34% of the projected pH change from 1860 in the tropics and up to 1.8% in temperate latitudes. Therefore, when proxy records are used to extend modern instrumental records of pH, we find that this may lead to bias, indicating daytime, nighttime, and full diel pH records should be assessed separately. We suggest that temporal pH cycles should be characterized at a local scale to enable incorporation of potential biases in the application of calcifying marine macroalgae to reconstruct pH change