A major goal of sensory neuroscience is to understand how neurons can together extract complex features from natural scenes. Even as early as in the retina, this is still an open challenge. Models built from the responses of sensory neurons to simple stimuli do not generalize to predict how complex, natural scene are processed. Between photoreceptors, the input, and ganglion cells (RGC), the retinal output, there are several layers of cells shaping the retinal computations. It is still unclear how they contribute to retinal processing. The purpose of our project is to decompose the retinal circuit that processes natural images, to understand how cell types in the intermediate layers contribute to retinal computations. Accessing intermediate layers with classical tools is challenging. We will develop for this project novel optical tools, holographic and silent substitution stimulation combined with optogenetics, to manipulate precisely the intermediate layers of the retina, and overcome the challenge of manipulating with light a tissue that is intrinsically light-sensitive. We will model the results with a deep convolutional neural network (DCNN), which is a neural network model composed of several layers, and show which cell type in the intermediate layers performs the computations of each layer of the neural network model. Despite DCNNs recent success modeling visual processing, we do not know if the computational mechanisms generating neuronal responses are the same as those in the brain: DCNNs act as a “black box” and it is difficult to relate the intermediate layers, modeled as hidden variables, to specific cell types. Here, we aim to provide the first full correspondence between a DCNN and a sensory circuit. We ask: how do intermediate layers of a DCNN map into specific circuit components of the retina? We propose that optogenetic manipulations of specific intermediate cells in combination with a new modeling strategy will reveal the identity of the inner nodes of a biologically compatible DCNN. This will be a major step towards understanding how the biological components of a sensory circuit embody complex computations. Beyond the retina, the methods and analyses that we will develop will pave the way for similar strategies to study other circuits in the retina, or in other parts of the brain. In general, we hope to establish a process to build biophysically realistic models when an intermediate hidden layer is perturbed and while a downstream area is being recorded. Such a method should be broadly applied to any sensory area in the brain.

Despite the fact that the World's 10 largest rivers drain almost one fifth of the global continental land area and deliver about one third of the terrestrial sediment supplied to oceans, we know relatively little about how such large rivers function. This is both surprising and problematic given that they impact directly on a wide range of environmental, social and economic issues (e.g. flooding, bank erosion, loss of land and infrastructure collapse) and ultimately create deposits that host some of the World's most lucrative mineral and fossil fuel reserves. Present understanding of large rivers is based almost entirely upon the findings of studies conducted in small channels. However, recent research gives us good reason to expect that transferring this knowledge to large rivers may not be straight-forward. Consequently, there is an urgent need to develop an improved quantitative understanding of the interactions between river processes, channel morphology and subsurface sedimentology in the World's largest rivers. Addressing this knowledge gap represents a significant challenge because it involves developing methods that can be used to investigate process-product relationships that operate across a wide range of time and space scales (from decimetres/minutes up to kilometres/millennia). This research will bring together a multi-disciplinary team of leading UK and overseas researchers in order to achieve this goal. In this project we will investigate one of the World's largest rivers, the Paraná-Paraguay in Argentina to understand: (1) what controls water and sediment movement and river channel changes over time; and (2) what this means for the formation and preservation of river sedimentary deposits. We will address these issues by implementing a research strategy that involves three key elements. First, we will use state-of-the-art field instrumentation to map river bed morphology and its evolution through time, and measure the three-dimensional patterns of water and sediment movement around and over channel bars. Second, we will take advantage of recent developments in Ground Penetrating Radar technology to map the three-dimensional sedimentary structure of braid-bar deposits, both within the current river and in formerly active areas that have been abandoned over the past few thousand years. Third, we will develop new numerical modelling approaches to investigate and quantify the interactions between water and sediment transport processes, bar formation, evolution of channel morphology and the subsurface sedimentology of deposits. The latter will involve combining, for the first time, Computational Fluid Dynamics models that provide a sophisticated representation of the physics governing water and sediment movement, with innovative Reduced-Complexity models capable of simulating how these processes interact to determine channel evolution and deposit sedimentology over periods of centuries to millennia. The result of this work will be the World's first comprehensive database on how the morphology of a large river changes through time, obtained concurrently with data on what drives those changes and what this means for the formation of sedimentary deposits. This will allow us to develop new models of how these rivers work and to use these models to address practical questions concerning large river resources and their management.

<< Objectives >>SMARTER develops smart supply chain knowledge competences as well as transversal skills through experiential learning applications. 7 universities from Spain, Austria, Ireland, Finland, Croatia, Romania and Argentina as well as 2 business development organisations (IRE, ESP) deliver digital learning material, innovation created in industry-university collaboration, as well as formative assessment tools for sustainable learning methods realising the 'Internationalisation at home' principle.<< Implementation >>University students will conduct a digital maturity gap analysis within the target businesses to create the framework for the core learning activity - the SMARTER Student Challenge, organised in Ireland in April 2024. Work packages 4-6 will prepare the Smart Material, Money, and Info flow to support the student challenge by using simulations, serious gaming, webinars and online workshops. The in-person workshop will be realised as a collaborative industry-university knowledge creation event.<< Results >>The final conference in Croatia in June 2025 will present the research publication of the digital maturity gap analysis, smart supply chain learning material, the experiential learning tools, the best practices learned from the student challenge, and feedback and reflection documents from student, educators and target group representatives as well as formal curricula integration opportunities. SMARTER achievements are also published in Social Media throughout the progress of the SMARTER project.

Successful previous LA and EU HE cooperation revealed differences between HESystems, academic recognition principles and practices. A long path is still ahead in terms of streamlined mobility recognition with fair credit transfer and grade conversion. Most LA countries have no credit system nationally applying to HEIs and the majority has regulations to frame mobility abroad, but handles recognition on individual basis and equivalence without grading is very common, negatively impacting students. RecMat joins partners from AR, BR and EU with the core aim of contributing to promote mobility between EU and LA, by reducing barriers related to academic recognition and building LA HEIs capacity to implement a fairer recognition process. Partners will work at two levels, linking the policy and practical dimensions behind academic recognition. Unlike previous initiatives, RecMat targets not only International Officers, but teachers who are the main decision makers in LA HEIs in what recognition is concerned, who are distant from Bologna rationale and usually show high resistance to recognition. This will be achieved by involving teachers in blended-training and in piloting concrete case studies, to evidence practical successful processes. Through a peer-to-peer approach, RecMat will raise teachers’ awareness about the importance of ensuring full recognition and stimulating fair grade conversion. RecMat activities will capacitate HEIs to formally frame recognition and build a linkage with IT teams by providing training to IT staff and enabling LA HEIs to outline concrete technical solutions to ease recognition. Through the organisation of public (inter-)national events and policy forums, RecMat will bring the topic to wide discussion, encouraging a high number of HEIs to adopt similar processes and sharing with them the project's innovative outcomes (MOOC, Digital Compendium and Conclusions Paper) towards the improvement and transparency of academic recognition processes.

Democratisation of HE in LA has helped to ensure a growing trend of increasing enrolment of students with disability, although it is still not significant enough in terms of potential numbers (Mexico: disabled people access in HE lower than 5%, Chile: 6,6% from which, only 2,66% complete their studies, Argentina: disabled students represents 0,08% of the students population). The adaptation of HE to cater for disability is of major importance from an economic, political and social point of view. Only through this can the employability of disabled persons be enhanced, and public policies focusing on the promotion of work, income security, poverty prevention and social exclusion can be supported. In this context, Chile, Argentina and Mexico goverments and HEIs have taken some measures to provide the necessary legal frameworks for the inclusion of disabled students within society however there are enormous voids between the law, public policies, and actual practices.The MUSE project proposal “Disability and Modernity: Ensuring Quality Education for Disabled Students” is an ambitious initiative with a wide range of stakeholders, activities and operational goals to cope with complex challenges. Taking into account the substantial expertise and advancement in this domain in recent years by European universities, the project aims to raise awareness on inclusive education through exchange of good-practices between EU and LA HEIs. The main objective constitutes the development of a sustainable knowledge base and support structure allowing for a coherent implementation of disability initiatives, and directional strategic plans in LA HEIs. A network will engage stakeholders in an educational and social discussion for the inclusion of disabled students in HEIs and a new external relation framework will be created to facilitate their economic integration.