Large Language Models (LLMs) and Evolutionary Computation (EC) are increasingly being combined to support automated optimization, algorithm design, and adaptive decision-making. This survey reviews the bidirectional interaction between these two paradigms and examines how their complementary strengths can be leveraged in hybrid intelligent systems. First, we analyze how EC can enhance LLM-based systems through prompt optimization, hyperparameter tuning, and architecture search. Second, we review how LLMs can improve EC by supporting metaheuristic design, surrogate reasoning, adaptive operator control, and heuristic generation. We further discuss emerging co-adaptive frameworks in which LLMs and EC interact through iterative feedback loops. Beyond summarizing recent developments, the survey provides a structured perspective on interaction mechanisms, application patterns, and methodological challenges, including computational cost, reproducibility, interpretability, benchmarking, and generalization. The paper concludes by outlining open research questions and future directions for developing more robust, transparent, and scalable LLM-EC systems.
Literature of person perception suggests the primacy of morality over sociability in interactions. While theories of emotion highlight the importance of adhering to socially acceptable display rules, a contentious debate surrounds whether individuals prioritize moral or sociable communication of emotions. We investigated how people perceive and respond to others who falsely communicate empathic versus truthfully communicate unempathic emotions related to others' (mis)fortunes. Across five preregistered studies (N = 1305), we examined participants' perceptions (sociability, morality, intent) and behavioral responses in a resource-allocation task using multiple paradigms (vignette, retrospective recall, and real-time interaction) and contexts (e.g., promotion, education, and lottery scenarios). Consistently, participants perceived targets who falsely communicated empathic emotions as lower on morality but higher on sociability and having more positive intent than those who truthfully communicated unempathic emotions. Moreover, despite the mixed findings regarding overall impressions, participants were consistently more willing to interact with and allocate resources to targets who falsely communicated empathic emotions. This preference was primarily driven by perceived sociability attributed to the targets. Together, these findings suggest that false empathic communication is favored over truthful unempathic communication, highlighting the relative primacy of sociability (and perceived intent) over morality in shaping responses to emotional communication in the context of others' (mis)fortunes.
We examined the Food Pharmacy in Rotterdam as a hybrid food-health pilot. Designed as a food-based intervention for patients with type-II diabetes, the pilot offers an insight into how institutional boundaries between agrifood and health are reconfigured in practice. We analyzed how institutional logics interact within the Food Pharmacy as well as its transformative potential and perceived pathways for scaling.Building on the institutional logics perspective, multisystem transformation, and debates on scaling, the analysis shows that transformation arises not from replacing old logics but from their tension and blending. Early relational practices created openings for experimentation, yet institutional pressures around efficiency and standardization repeatedly narrowed this space, prompting actors to engage in reactive blending. These oscillations shaped how respondents envisioned scaling: scaling deep drew on relational foundations, scaling out on cross-sector improvisations, and scaling up on alignment with formal structures.The study draws on interviews, observations, and documents, identifying four logics: State, Market, Professional Biomedical, and Community. Each shaped the pilot in dynamic and contested ways. State and Market logics provided resources and legitimacy but often imposed bureaucratic and profit-oriented frames. Community and Professional logics sustained engagement by stressing belonging, fairness, and rights to healthy food.The findings highlight how relational, culturally embedded practices remain difficult to institutionalize within governance frameworks privileging measurable outcomes. The Food Pharmacy illustrates both the promise and fragility of cross-sectoral innovation: new practices and connections can emerge, but their realization depends on governance systems capable of supporting relational work over time.
Worldwide, 10 to 40 million tons of microplastics (MPs) are emitted annually into the environment, redistributing across environmental compartments and affecting organisms. Current life cycle impact assessment (LCIA) methods partially account for their impacts on ecosystem quality by considering a limited number of compartments and particles. This study advances the work of the MarILCA group by introducing the PEBBLE method which integrates physical effects of MPs, tire and road wear particles (TRWPs), and cellulosic microfibers across environmental compartments into life cycle assessment (LCA).Characterization factors (CFs) are developed for 10 conventional polymers, 6 biodegradable polymers, 6 cellulosic materials, and TRWPs, for multiple shapes and sizes. Particle fate is modelled using the SimpleBox object-oriented (SBoo) model for microparticles, computing fate factors (FFs) in air, terrestrial, freshwater, and marine environments at continental and global scales, following USEtox regionalization. SBoo was improved to account for size and shape effects, refined degradation data, regional marine currents, and biofouling.FFs are combined with exposure and effect factors in water, sediments, and terrestrial environments to obtain updated CFs. Uncertainty is quantified using Monte Carlo simulation, and time-horizon-dependent CFs are computed. Results are provided for different LCIA methods: for IMPACT World+ and ReCiPe, CFs representing regional species loss are derived, while for GLAM, the Global Extinction Probability is applied to derive CFs representing global species loss. The results support environmental decision-making for plastics and their alternatives, and aim to be the consolidated version of MarILCA CFs for microparticles, ready for implementation in LCA tools and methods.
Well-being is increasingly recognized as a more meaningful measure of societal progress than economic output, yet most international comparisons continue to emphasize economic indicators. This focus has reinforced the perception that Europe is lagging behind the United States, with the productivity gap shaping the European Commission’s recent competitiveness agenda. Here, we compare European and American development from 1990–2023 using an interdisciplinary well-being framework that integrates five scientific schools of thought. We find that while the US increasingly outperformed the EU on productivity and income, the EU caught up in terms of various well-being indicators, including life satisfaction and education, while increasingly outperforming the US in life expectancy and safety. Europe also exhibits more equitable well-being outcomes and achieves them at lower environmental cost. However, in both regions, the conditions underpinning future well-being are under increasing pressure. These results challenge the dominance of economic metrics in evaluating progress and emphasize the need for inclusive development strategies that place sustainable well-being at their core.
