Transforming scientific information with everyday thinking: A social representations perspective

Cette thèse examine la circulation et l'intégration des informations scientifiques dans la pensée quotidienne d'après la théorie des représentations sociales (TRS). En tant qu'alternative aux approches traditionnelles de la communication de la science, les transformations survenant entre le discours scientifique et le discours de sens commun sont considérées comme adaptatives. Deux études sur la circulation des informations dans les media (études 1 et 2) montrent des variations dans les thèmes de discours exposés aux profanes, et parmi les discours de ceux-ci, en fonction de différentes sources. Ensuite, le processus d'ancrage dans le positionnement préalable envers la science est étudié, pour l'explication qu'il fournit de la réception et de la transmission d'informations scientifiques dans le sens commun. Les effets d'ancrage dans les attitudes et croyances préexistants sont reportés dans différents contextes de circulation des informations scientifiques (études 3 à 7), incluant des études de type corrélationnel, experimental et de terrain. Globalement, cette thèse procure des arguments en faveur de la pertinence de la TRS pour la recherche sur la communication de la science, et suggère des développements théoriques et méthodologiques pour ces deux domaines de recherche. Drawing on the social representations theory (SRT), this thesis examines the circulation and integration of scientific information into everyday thinking. As an alternative to the traditional approaches of science communication, it considers transformations between scientific and common-sense discourses as adaptive. Two studies, focused on the spreading of information into the media (Studies 1 and 2), show variations in the themes of discourses introduced to laypersons and in the themes among laypersons' discourses, according to different sources. Anchoring in prior positioning toward science is then studied for the explanation it provides on the reception and transmission of scientific information into common sense. Anchoring effects in prior attitudes and beliefs are reported in different contexts of circulation of scientific information (Studies 3 to 7) by using results from correlational, field, and experimental studies. Overall, this thesis provides arguments for the relevance of SRT in science communication research and suggests theoretical and methodological developments for both domains of research.

Traitement de la schistosomiase à S. mansoni : quelle alternative au praziquantel ?

Les schistosomiases sont des maladies parasitaires causées par des helminthes du genre Schistosoma (S.) qui touchent 200 millions de personnes dans le monde, mais restent rares chez le voyageur. Contrairement à S. heamatobium, agent de la bilharziose urinaire, S. mansoni, présent en Afrique subsaharienne, en Egypte ainsi qu'aux Antilles, au Surinam et dans le nordest du Brésil, est responsable des formes hépato-intestinales de la maladie. Les larves, vivant en eaux douces contaminées par des selles infectées, peuvent pénétrer la peau des baigneurs sans que l'individu ne s'en rende compte. Les parasites adultes s'établissent dans le système veineux digestif où ils se reproduisent et excrètent des oeufs qui migreront dans la lumière intestinale. Cette revue systématique évalue les effets des médicaments antibilharziens, utilisés seuls ou en association, pour traiter l'infection à S. mansoni.

Reactive and proactive approaches for introspective CBR

This work investigates applying introspective reasoning to improve the performance of Case-Based Reasoning (CBR) systems, in both reactive and proactive fashion, by guiding learning to improve how a CBR system applies its cases and by identifying possible future system deficiencies. First we present our reactive approach, a new introspective reasoning model which enables CBR systems to autonomously learn to improve multiple facets of their reasoning processes in response to poor quality solutions. We illustrate our model’s benefits with experimental results from tests in an industrial design application. Then as for our proactive approach, we introduce a novel method for identifying regions in a case-base where the system gives low confidence solutions to possible future problems. Experimentation is provided for Zoology and Robo-Soccer domains and we argue how encountered regions of dubiosity help us to analyze the case-bases of a given CBR system.

Spatial modelling of biodiversity at the community level

1. Statistical modelling is often used to relate sparse biological survey data to remotely derived environmental predictors, thereby providing a basis for predictively mapping biodiversity across an entire region of interest. The most popular strategy for such modelling has been to model distributions of individual species one at a time. Spatial modelling of biodiversity at the community level may, however, confer significant benefits for applications involving very large numbers of species, particularly if many of these species are recorded infrequently. 2. Community-level modelling combines data from multiple species and produces information on spatial pattern in the distribution of biodiversity at a collective community level instead of, or in addition to, the level of individual species. Spatial outputs from community-level modelling include predictive mapping of community types (groups of locations with similar species composition), species groups (groups of species with similar distributions), axes or gradients of compositional variation, levels of compositional dissimilarity between pairs of locations, and various macro-ecological properties (e.g. species richness). 3. Three broad modelling strategies can be used to generate these outputs: (i) 'assemble first, predict later', in which biological survey data are first classified, ordinated or aggregated to produce community-level entities or attributes that are then modelled in relation to environmental predictors; (ii) 'predict first, assemble later', in which individual species are modelled one at a time as a function of environmental variables, to produce a stack of species distribution maps that is then subjected to classification, ordination or aggregation; and (iii) 'assemble and predict together', in which all species are modelled simultaneously, within a single integrated modelling process. These strategies each have particular strengths and weaknesses, depending on the intended purpose of modelling and the type, quality and quantity of data involved. 4. Synthesis and applications. The potential benefits of modelling large multispecies data sets using community-level, as opposed to species-level, approaches include faster processing, increased power to detect shared patterns of environmental response across rarely recorded species, and enhanced capacity to synthesize complex data into a form more readily interpretable by scientists and decision-makers. Community-level modelling therefore deserves to be considered more often, and more widely, as a potential alternative or supplement to modelling individual species.

Longitudinal analysis of income-related health inequality: welfare foundations and alternative measures

This paper elaborates the approach to the longitudinal analysis of income-related health inequalities first proposed in Allanson, Gerdtham and Petrie (2010). In particular, the paper establishes the normative basis of their mobility indices by embedding their decomposition of the change in the health concentration index within a broader analysis of the change in “health achievement” or wellbeing. The paper further shows that their decomposition procedure can also be used to analyse the change in a range of other commonly-used incomerelated health inequality measures, including the generalised concentration index and the relative inequality index. We illustrate our work by extending their investigation of mobility in the General Health Questionnaire measure of psychological well-being over the first nine waves of the British Household Panel Survey from 1991 to 1999.

Evaluating sampling strategies and logistic regression methods for modelling complex land cover changes

The role of land cover change as a significant component of global change has become increasingly recognized in recent decades. Large databases measuring land cover change, and the data which can potentially be used to explain the observed changes, are also becoming more commonly available. When developing statistical models to investigate observed changes, it is important to be aware that the chosen sampling strategy and modelling techniques can influence results. We present a comparison of three sampling strategies and two forms of grouped logistic regression models (multinomial and ordinal) in the investigation of patterns of successional change after agricultural land abandonment in Switzerland. Results indicated that both ordinal and nominal transitional change occurs in the landscape and that the use of different sampling regimes and modelling techniques as investigative tools yield different results. Synthesis and applications. Our multimodel inference identified successfully a set of consistently selected indicators of land cover change, which can be used to predict further change, including annual average temperature, the number of already overgrown neighbouring areas of land and distance to historically destructive avalanche sites. This allows for more reliable decision making and planning with respect to landscape management. Although both model approaches gave similar results, ordinal regression yielded more parsimonious models that identified the important predictors of land cover change more efficiently. Thus, this approach is favourable where land cover change pattern can be interpreted as an ordinal process. Otherwise, multinomial logistic regression is a viable alternative.

The Importance of Graduates for the Scottish Economy: A “Micro-to-Macro” Approach

There have been numerous attempts to assess the overall impact of Higher Education Institutions on regional economies in the UK and elsewhere. There are two disparate approaches focussing on: demand-side effects of HEIs, exerted through universities’ expenditures within the local economy; HEIs’ contribution to the “knowledge economy”. However, neither approach seeks to measure the impact on regional economies that HEIs exert through the enhanced productivity of their graduates. We address this lacuna and explore the system-wide impact of the graduates on the regional economy. An extensive and sophisticated literature suggests that graduates enjoy a significant wage premium, often interpreted as reflecting their greater productivity relative to non-graduates. If this is so there is a clear and direct supply-side impact of HEI activities on regional economies through the employment of their graduates. However, there is some dispute over the extent to which the graduate wage premium reflects innate abilities rather than the impact of higher education per se. We use an HEI-disaggregated computable general equilibrium model of Scotland to estimate the impact of the growing proportion of graduates in the Scottish labour force that is implied by the current participation rate and demographic change, taking the graduate wage premium in Scotland as an indicator of productivity enhancement. We conduct a range of sensitivity analyses to assess the robustness of our results. While the detailed results do, of course, vary with alternative assumptions about future graduate retention rates and the size of the graduate wage premium, for example, they do suggest that the long-term supply-side impacts of HEIs provide a significant boost to regional GDP. Furthermore, the results suggest that the supply-side impacts of HEIs are likely to be more important than the expenditure impacts that are the focus of most “impact” studies.

“Policy Scepticism” and the Impact of London-based Higher Education Institutions (HEIs) on the economy of England: Accounting for Alternative Uses of Public Expenditure

This paper replicates the analysis of Scottish HEIs in Hermannsson et al (2010a) for the case of London-based HEIs’ impact on the English economy in order to provide a self-contained analysis that is readily accessible by those whose primary concern is with the regional impacts of London HEIs. A “policy scepticism” has emerged that challenges the results of conventional regional HEI impact analyses. This denial of the importance of the expenditure impacts of HEIs appears to be based on a belief in either a binding regional resource constraint or a regional public sector budget constraint. In this paper we provide a systematic critique of this policy scepticism. However, while rejecting the extreme form of policy scepticism, we argue that it is crucial to recognise the importance of alternative uses of public expenditure, and show how conventional impact analyses can be augmented to accommodate this. While our results suggest that conventional impact studies overestimate the expenditure impacts of HEIs, they also demonstrate that the policy scepticism that treats these expenditure effects as irrelevant neglects some key aspects of HEIs, in particular their export intensity.

Forecasting with Medium and Large Bayesian VARs

This paper is motivated by the recent interest in the use of Bayesian VARs for forecasting, even in cases where the number of dependent variables is large. In such cases, factor methods have been traditionally used but recent work using a particular prior suggests that Bayesian VAR methods can forecast better. In this paper, we consider a range of alternative priors which have been used with small VARs, discuss the issues which arise when they are used with medium and large VARs and examine their forecast performance using a US macroeconomic data set containing 168 variables. We nd that Bayesian VARs do tend to forecast better than factor methods and provide an extensive comparison of the strengths and weaknesses of various approaches. Our empirical results show the importance of using forecast metrics which use the entire predictive density, instead of using only point forecasts.

Establiment de bases científiques per al desenvolupament d'una estratègia integrada per avaluar toxicitat en reproducció

Report for the scientific sojourn at the National Institute for Public Health and the Environment (RIVM), Netherlands, from 2006 to 2008. This project aimed at creating scientific elements that could help deriving an integrated testing strategy for reproductive toxicity. Part of the project focused on the use of alternative tests for regulatory purposes. An in vitro-in vivo extrapolation method for embryotoxicity was proposed and evaluated. In vitro and in vivo dose descriptors were correlated; however, the scatter in the correlation was too large to allow an accurate estimation of an in vivo dose from an in vitro dose. The in vitro-in vivo extrapolation method together with toxicokinetic data was also applied to a category of substances (phthalates). Although based on a limited number of substances, the results suggested that in vitro-in vivo extrapolation for embryotoxicity is possible within a category of compounds if adequate toxicokinetic data is available. Because of the limitations that still remain in the use of alternative tests for reproductive toxicicity, other approaches to reduce animal testing were explored. Thus, a database of reproductive toxicity studies was created to retrospectively evaluate the comparative value of some studies or elements in a particular study. When compared to the subchronic toxicity study, the rat two-generation reproductive toxicity study had a considerable impact on the identification of hazard for reproductive toxicity, but not on the overall NOAEL. Among the two-generation studies included in our database, the second generation affected neither the overall NOAEL nor the critical effect. The rat and the rabbit developmental toxicity studies were, on average, similarly sensitive. However, for certain substances the developmental study in either one of the two species appeared to be more sensitive than in the other species.

Mass spectrometry as a rapid and powerful alternative to antibodies for detecting LPXTG wall-associated proteins of Staphylococcus aureus

Functional characterization of transformed or natively present bacterial virulence proteins can be achieved employing various model systems. A prerequisite is to verify the correct expression of the transformed protein or the presence of the native protein in the microbe. Traditionally, antibodies are raised against the protein or a peptide thereof, followed by Western blot analysis or by fluorescence-activated cell sorting. Alternatively, the protein-coding gene can be fused with a downstream reporter gene, the expression of which reports the simultaneous expression of the upstream recombinant protein. Although being powerful, these methods are time consuming, especially when multiple proteins must be assessed. Here we describe a novel way to validate the expression of Gram-positive surface proteins covalently attached to the peptidoglycan. Eighteen out of the 21 known LPXTG-motif carrying cell wall-associated proteins of Staphylococcus aureus were cloned in Lactoccocus lactis either alone, in combinations or as truncated forms, and their correct expression was assessed by liquid chromatography coupled to mass spectrometry (LC-MS). The method is rapid, sensitive and precise. It can identify multiple proteins in transformed constructs without the time and cost needed for raising and testing multiple sets of antibodies.