Researchers in South Africa's Karoo region are redefining the relationship between Indigenous knowledge and modern science, as they study the ethnoastronomical knowledge of local communities. The project, which involves elders and knowledge holders, highlights the dynamic and adaptive nature of traditional knowledge, which continues to...
A mathematical model shows that greater fitness is not enough for a population to succeed; expansion speed and position also play crucial roles. Less fit species can win if they occupy favorable positions, such as peaks or protrusions.
Researchers have developed a new version of the Daydreaming algorithm, which combines learning and cleaning to improve artificial memory systems' reliability even with biased data. The algorithm focuses on differences between pixels, allowing it to work effectively with strongly biased data, similar to real-world conditions.
A team of physicists, including Giorgio Parisi and Francesco Zamponi, collaborated with AI system Claude to solve a decade-old mathematical problem in physics. They found that two different theoretical approaches to jamming lead to the same physical laws, confirming a surprising relationship.
Researchers studied dark matter's potential additional attractive force and found it suppresses structure growth, contrary to expectations. The extra clustering effect is counterbalanced by a decrease in dark matter particles' mass over time, reducing the overall impact on cosmic microwave background observations.
A new study explores how transfer learning reduces computational costs in cosmological simulations while revealing risks of negative transfer, which can hinder learning new physics. Transfer learning can accelerate inference but may also push AI systems toward incorrect interpretations of new effects.
The Science & Cinema format combines short movie clips with expert commentary to reflect on how science is portrayed on screen. The events engaged diverse audiences, including those not already interested in the topic, triggering reflection and emotional connections.
Researchers have developed a simplified mathematical model of learning in neural networks, shedding new light on how these systems produce their responses. The toy model, inspired by physics principles, captures key features of complex systems and offers insights into the surprising efficiency and stability of modern AI systems.
A team of undergraduate students built a cavity detector to search for axions and set new experimental limits on their properties. The result was achieved with relatively limited resources, showcasing the potential for small-scale experiments in precision cosmology.
Researchers propose new method to capture complex human movement in public spaces, addressing limitations of traditional approaches. By standardizing data measurement, they aim to improve crowd safety predictions and design safer spaces.
Researchers suggest that dark matter may consist of multiple particles, whose behavior varies depending on the cosmic environment. This could explain why a signal observed at the center of our galaxy is not seen in dwarf galaxies.
A new study published in JCOM finds that images of GMOs tend to reinforce pre-existing attitudes, amplifying polarization rather than changing minds. The study used a representative sample of the US population and found that even positive images could make people who already supported GMOs more positive, while making skeptics and uncer...
A new study suggests that the ultra-high-energy neutrino may have originated from a population of blazars, which could provide a plausible explanation for the rare phenomenon. The researchers used a combination of simulations and observations from various instruments to test their hypothesis.
A new study published in JCOM found that AI disclosure labels may have the opposite effect of what regulators intend, decreasing the credibility of true scientific information while increasing that of false claims. The study tested 433 participants and found a counterintuitive pattern where the same label reduces credibility in opposit...
A Brazilian research team used non-formal education methods in science museums to improve high school students' interest and confidence in chemistry. The interactive exhibition explained advanced oxidation processes, increasing students' sense of competence and positive perception of the discipline.
A new paper proposes a way to calculate the thermodynamic costs of metabolic processes, ranking them according to their biological efficiency. The method estimates the improbability of a network behaving in a certain way, considering maintenance and restriction costs.
SourceSissa Medialab·JournalJournal of Statistical Mechanics Theory and Experiment·TypeComputational simulation/modeling·DateJan 6, 2026 The Atacama Cosmology Telescope's sixth and final data release confirms the 'Hubble tension' and rules out extended cosmological models, providing new insights into the Universe's evolution and current state. ACT's observations offer a cleaner starting point for future research.
A new study explores how emotions play a crucial role in communicating natural disasters, finding that balancing fear with positive emotions like fascination and confidence can improve disaster preparedness. The research highlights the importance of recognizing and channelling emotions to promote self-efficacy and motivation.
A Portuguese theatre company collaborated with doctors and patients to develop a play about bipolar disorder, which was performed in front of an audience of over 300 people. The play's emotional engagement helped facilitate the assimilation of scientific information and contributed to reducing stigma surrounding mental illness.
Researchers propose a method to distinguish between nanohertz gravitational wave sources using pulsar timing arrays. By searching for beat phenomena in the tiny shifts of pulsars' radio-pulse arrival times, scientists can identify specific, nearby binary supermassive black hole systems.
A new JCOM study reveals that science journalists act as 'knowledge brokers,' negotiating with their audiences to convey accurate information. The researchers found that public trust in science is not uniformly declining, but rather fragmented and dependent on social, political, and media contexts.
Researchers have developed an emulator called Effort.jl that mimics the behavior of large-scale structure models, allowing for fast analysis on standard laptops. The new model delivers similar accuracy as the original, enabling scientists to analyze upcoming data releases from experiments like DESI and Euclid.
The Exoplanet Travel Bureau's poster campaign successfully promoted exoplanets as vacation destinations by leveraging retro-style posters, virtual tours, and coloring books. The synergy between artists and scientists in this campaign revealed that scientists also make imaginative leaps to turn abstract data into concrete planet concepts.
Astronomers have developed a protocol to detect supernovae within 24 hours of their explosion, using high-cadence sky surveys. The method involves rapid searches for candidates based on light signal absence and galaxy location, followed by spectroscopic observations to determine the type of supernova.
Researchers propose 'dark dwarfs' could be key to understanding dark matter, with lithium-7 serving as a unique marker. A discovery of a dark dwarf would provide compelling evidence for WIMPs as a possible dark matter component.
SourceSissa Medialab·JournalJournal of Cosmology and Astroparticle Physics·TypeComputational simulation/modeling·DateJul 7, 2025 A new study reveals that AI systems transition from relying on word positions to meaning-based understanding as they receive enough data for training. The transition occurs abruptly, similar to a phase transition in physical systems, and is driven by the amount of data available.
SourceSissa Medialab·JournalJournal of Statistical Mechanics Theory and Experiment·TypeData/statistical analysis·DateJul 7, 2025 A new study applies network analysis to large-scale data from the French labor market, identifying four main clusters: diffusers, channels, hubs, and condensers. Diffuser occupations offer wide exit opportunities but are hard to enter, while channels and hubs face accessibility issues.
Researchers explore alternative models of black holes without singularities, which could be distinguishable from standard black holes through subtle deviations in predictions. Observational tests using sophisticated instruments and different channels may reveal clues about internal structure.
Researchers found that realistic AI avatars are rated more positively than cartoon-style ones for perceived competence, integrity, and benevolence. However, individual factors such as prior AI knowledge and trust in science moderate perceptions of trustworthiness.
The first phase of Janeway development is complete, enabling a unified publishing platform tailored for science communication and beyond. The system promises to address challenges in open science and AI integration with a modular, expandable architecture.
A study published in JCAP has established upper limits on the strength of quantum gravity effects on neutrino oscillations, providing valuable insights into the long-sought theory. The results show no signs of decoherence, a phenomenon that could be a key indicator of quantum gravity's presence.
A new study in JSTAT introduces a hiring strategy model that suggests dividing candidates into two groups: those to be evaluated and rejected upfront, and those to be selected based on their performance relative to previous hires. The optimal approach depends on the company's objective, balancing quality and speed.
SourceSissa Medialab·JournalJournal of Statistical Mechanics Theory and Experiment·TypeComputational simulation/modeling·DateMar 10, 2025 A new study published in JCOM found that humor can aid scientists' communication efforts, but only if people think they're funny. The research shows that humor can simplify complex concepts, making them more relatable and easier to understand, and has the potential to dispel misinformation in a friendly manner.
Researchers propose a new methodology to test the Universe's isotropy using Euclid space telescope data, aiming to detect potential anisotropies that challenge the Standard Model of Cosmology. If confirmed, these findings would open a new chapter in cosmology, potentially revising our understanding of the Universe's behavior.
A study found that using comics to explain forensic science in court significantly improved jurors' understanding of the evidence. Participants who watched comics discussed the evidence in greater detail, showed increased confidence in their reasoning, and made more explicit references to scientific concepts.
A new paper in JCAP proposes a way to test the anthropic principle, which suggests the universe is fine-tuned for life. The proposal involves confirming three conditions: cosmic inflation, axion existence, and dark matter not being made of axions.
Researchers suggest that starting with smaller neural networks and using curriculum learning can improve performance and reduce the need for massive computing resources. This approach could lead to more resource-efficient and less energy-consuming AI systems.
SourceSissa Medialab·JournalJournal of Statistical Mechanics Theory and Experiment·TypeComputational simulation/modeling·DateNov 19, 2024 A new study published in JCOM identified the characteristics of polarized scientific digital messages, proposing a system to codify and characterize polarized discourses. The research found that simplifying and presenting scientific information as unquestionable can limit critical thinking and hinder scientific progress.
A new special issue of JCOM explores the challenges and opportunities of communicating basic science, highlighting its relevance in everyday life. The issue emphasizes the importance of curiosity and the need to convey the impact of scientific discoveries on individuals.
A recent study suggests that the observation of antihelium nuclei in cosmic rays may be consistent with the existence of WIMP particles, which could make up dark matter. The detection of two distinct isotopes, antihelium-3 and -4, is particularly intriguing as heavier nuclei are unlikely to be produced through natural processes.
A new paper in JCOM describes the development of an automatic translation app for sign language, designed through co-creation with deaf communities. The research team used a theatrical performance and AI tools to gather feedback from audiences, informing the app's design and features.
A new study by an international team, including MIT and CNRS, observed that similarities exist between the behavior of birds in flight and physical systems. The research suggests that the transition from disorder to coordination is not as different between particles and biological elements as previously thought.
SourceSissa Medialab·JournalJournal of Statistical Mechanics Theory and Experiment·TypeComputational simulation/modeling·DateAug 8, 2024 Researchers used hydrogen to track dark matter's presence in the universe, revealing a tension between observations and theoretical predictions. The findings suggest that an unknown particle or new physics may be responsible for this discrepancy.
A new study in JCOM monitored changes in university communication within the German landscape six months after the launch of ChatGPT 3. Practically all responding universities use generative AIs mainly to increase process efficiency, with some caution towards ethical aspects such as data protection.
A mathematical model developed by Alexandre Solon and Eric Bertin describes the movement of particles in situations similar to cars on a road or bacteria attracted to a nutrient source. The model identifies conditions that favor traffic jams, including high vehicle density and driver inertia.
SourceSissa Medialab·JournalJournal of Statistical Mechanics Theory and Experiment·TypeComputational simulation/modeling·DateApr 30, 2024 A new study has sorted through models attempting to solve the cosmological tension, a discrepancy between two ways of calculating the universe's expansion. Three models that were previously viable solutions were excluded by the new data, while others reduced the tension but not solved it.
Researchers applied Bayesian network analysis to a corpus of 142 science fiction works, investigating the representation of extrasolar planets. They found that after the discovery of real exoplanets in 1995, fictional exoplanets became less Earth-like, reflecting the scientific reality of vast numbers of diverse exoplanet types.