Julian Vicens

This paper explores the integration of citizen science into climate citizens’ assemblies as a pathway to democratise not only decision-making but also knowledge production. While citizens' assemblies enable citizens to deliberate on policy, they typically rely on expert-defined evidence, limiting participants’ epistemic agency. Conversely, citizen science empowers lay publics to generate and interpret data but often lacks deliberative structures for collective reflection. We present the co-creation and testing of a Citizen Science Toolkit for Climate Assemblies, developed through participatory workshops in Living Labs and field deployments across two European climate citizens' assemblies, in Riga and Edermunde. Using mixed methods, including surveys, focus groups, and qualitative coding, we analyse participants’ and organisers’ perspective. Results from those two assemblies show that integrating citizen science fosters knowledge uptake, citizen empowerment, and trust in the process, but also increases organisational demands. We argue that structurally embedding citizen science within deliberative formats expands the epistemic legitimacy of democratic innovation. The paper concludes with design principles for future citizens' assemblies and outlines a framework for citizen science as a climate adaptation governance tool.

Energy production and management face significant political, economic, and environmental challenges. The rise in information consumption through social media can, under certain conditions, reduce the visibility and uptake of reliable knowledge, particularly in environments characterised by high information overload, algorithmic amplification, and low media literacy. This study examines the ideas discussed in the energy transition content on YouTube, assesses the most effective methods of communicating knowledge and information, and identifies the most engaged audiences. We examine videos related to the subject, analysing the themes discussed, the language used, and the emotions conveyed on YouTube, linking language formality to user engagement. To test the relationship experimentally, original content was uploaded to YouTube through two mirror channels containing the same material but using different levels of language formality. We conducted a systematic statistical analysis of engagement data collected from YouTube and the Google Ads platform. The YouTube engagement suggests that the conversational channel reaches a broader audience, although retention across video segments varies. While user retention was higher in the early segments for the conversational content, a higher retention in the mid to late sections of academic videos was found. Data from the Google Ads platform provided a deeper understanding of engagement across user profiles. Younger individuals and women show greater engagement regardless of language style, although women demonstrate relatively greater interest in the academic content.

We study the dynamics and intervention strategies of a rumor using the modified Maki-Thompson model. A key challenge in social networks is distinguishing between natural increases in transmissibility and artificial injections of rumor spreaders, such as through broadcast events or astroturfing. Using stochastic simulations, we compare two scenarios: one with organic growth in transmissibility and another with externally injected spreaders. Although both lead to high autocorrelation, only the organic growth produces oscillatory patterns in autocorrelation at multiple lags, an effect we can analytically explain using the N -intertwined mean-field approximation. This distinction offers a practical tool to identify the origin of rumor virality and also infer its transmissibility. Our approach is validated analytically and tested on real-world data from Twitter during the announcement of the Higgs boson discovery. In addition to detection, we also explore control strategies. We show that the average lifetime of a rumor can be manipulated through targeted interventions: placing spreaders at specific locations in the network. Depending on their placement, these interventions can either extend or shorten the lifespan of the rumor.

Social media platforms like YouTube and Twitter play a key role in disseminating both reliable and unreliable information about climate change. This study analyses the topology of interactions in Twitter and their relation to cross-platform sharing, content discussions and emotional responses. We examined climate change discussions across four topics: the 27th United Nations Climate Change Conference, the Sixth Assessment Report of the United Nations Intergovernmental Panel on Climate Change, Climate Refugees and Doñana Natural Park. While retweets reinforce in-group cohesion in the form of echo chambers, inter-group exposure is significant through mentions, suggesting that exposure to opposing views intensifies polarization, rather than mitigates it. Ideological divides feature content differences accompanied by steeper negative sentiments, especially from right-leaning communities prone to share low-reliability information. We identified a topological and thematic alignment between platforms, indicating that ideological communities are interconnected across them. Our findings show that climate change polarization is multifaceted, involving ideological divides, structural isolation and emotional engagement. These results suggest that effective climate policy discussions must address the emotional and identity-driven nature of public discourse and seek strategies to bridge ideological divides.

Bike Sharing Systems (BSSs) play a crucial role in promoting sustainable urban mobility by facilitating short-range trips and connecting with other transport modes. Traditionally, most BSS fleets have consisted of mechanical bikes (m-bikes), but electric bikes (e-bikes) are being progressively introduced due to their ability to cover longer distances and appeal to a wider range of users. However, the charging requirements of e-bikes often hinder their deployment and optimal functioning. This study examines the spatiotemporal variations in battery levels of Barcelona’s BSS, revealing that bikes stationed in the city’s expansion area (Eixample) tend to have shorter rest periods and lower average battery levels. Additionally, to improve the management of e-bike fleets, a Markov-chain approach is developed to predict both bike availability and battery levels. This research offers a unique perspective on the dynamics of e-bike battery levels and provides a practical tool to overcome the main operational challenges in their implementation.

This study investigates the distinction between fake and real news articles by analysing both linguistic characteristics and the reader's physiological responses during reading. We employed a combination of eye-tracking and peripheral physiological measurements to capture real-time physiological reactions, alongside self-reported responses, as participants engaged with these articles. Our primary objectives were to assess whether physiological responses and eye-tracking patterns differ between fake and real news and to identify linguistic descriptors that most strongly correlate with these responses. The results reveal that specific physiological metrics can differentiate between fake and real news, and we observed notable correlations between certain linguistic features and these physiological responses.

Bike-sharing systems have emerged as a significant element of urban mobility, providing an environmentally friendly transportation alternative. With the increasing integration of electric bikes alongside mechanical bikes, it is crucial to illuminate distinct usage patterns and their impact on maintenance. Accordingly, this research aims to develop a comprehensive understanding of mobility dynamics, distinguishing between different mobility modes, and introducing a novel predictive maintenance system tailored for bikes. By utilising a combination of trip information and maintenance data from Barcelona’s bike-sharing system, Bicing, this study conducts an extensive analysis of mobility patterns and their relationship to failures of bike components. To accurately predict maintenance needs for essential bike parts, this research delves into various mobility metrics and applies statistical and machine learning survival models, including deep learning models. Due to their complexity, and with the objective of bolstering confidence in the system’s predictions, interpretability techniques explain the main predictors of maintenance needs. The analysis reveals marked differences in the usage patterns of mechanical bikes and electric bikes, with a growing user preference for the latter despite their extra costs. These differences in mobility were found to have a considerable impact on the maintenance needs within the bike-sharing system. Moreover, the predictive maintenance models proved effective in forecasting these maintenance needs, capable of operating across an entire bike fleet. Despite challenges such as approximated bike usage metrics and data imbalances, the study successfully showcases the feasibility of an accurate predictive maintenance system capable of improving operational costs, bike availability, and security.

Misinformation in its many forms is a substantial and growing problem for society today. Whether financially or ideologically motivated, purveyors of misinformation do not abide by legal, technical or moral rules. Therefore new, ludic, narrative, gamified and artistic approaches are needed. In this paper we analyse the approaches taken in countering misinformation by 18 AI and machine learning works of art, developed in the MediaFutures project. We examine how these align with existing AI approaches to countering misinformation, and how they address some of the key challenges. We show that AI artists engage with existing debunking and inoculating strategies, including highly technical aspects such as deepfakes, while also utilizing focused strategies of data literacy and collective intelligence. We also find that they are able to integrate hard-to-refute strategies such as narrative and emotion. These findings suggest that data as an art material and AI techniques as art tools are worth of further investigation as to their effectiveness for countering misinformation within society.