Friday, February 21, 2025

Iberian nailed head ritual was more complex than expected


The isotope analysis of the Puig Castellar and Ullastret sites point to different mobility patterns in these individuals, who would not have been randomly selected.



Universitat Autonoma de Barcelona

Iberian nailed head ritual was more complex than expected 

image: 

One of the severed heads found at Ullastret (Girona, Spain). © Museu d'Arqueologia de Catalunya (MAC)-Ullastret a De Prado, 2015

view more 

Credit: © Museu d'Arqueologia de Catalunya (MAC)-Ullastret a De Prado, 2015




The nailed heads ritual did not correspond to the same symbolic expression among the Iberian communities of the northeast of the Iberian Peninsula, but rather a practice that differed in each settlement. In some, external individuals were used as symbols of power and intimidation, while other settlements could have given priority to the veneration of members of the local community.

This is the conclusion reached by a study led by the Universitat Autònoma de Barcelona (UAB) which analyses the mobility patterns of these human communities existing in the Iron Age of the last millenium BCE. Researchers studied seven nailed skulls of men found in two sites dating back to this period: the city of Ullastret (found in the same town of the province of Girona) and the settlement of Puig Castellar (Santa Coloma de Gramenet, Barcelona).

The study, published in the Journal of Archaeological Science: Reports, was coordinated by researchers from the Department of Animal Biology, Plant Biology and Ecology of the UAB and also included the collaboration of researchers from the Museum of Archaeology of Catalonia (MAC), the Museum Torre Balldovina and the universities of Lleida, Bordeaux (France), and Tübingen (Germany).

Severed heads: more than just simple war trophies

Severed heads were a unique funerary practice within the Iberian world and represent an exceptional opportunity to analyse these communities, of which very little archaeological record exists since cremation was the predominant burial ritual. This practice consisted of the public exhibition of the skulls of certain individuals, subjected to a post-mortem treatment. Some of these skulls have been recovered with signs of nailing and in some cases with an iron nail still in place.

 “Who were these individuals and for what were their heads used?” Traditionally, archaeologists have debated whether the skulls were war trophies — to intimidate their enemies — or venerated relics of important community members. These hypotheses, however, based on oral and ethnographic sources, have not yet been verified, nor has there been in-depth studies on the relationship between these groups and the land they inhabited. “Our premise in approaching the study was that if they were war trophies they would not come from the sites analysed, while if they were venerated individuals, these would most likely be local”, explains Rubén de la Fuente-Seoane, archaeologist at the UAB and first author of the study.

“Our results reveal that the individuals from Puig Castellar and Ullastret would not have been randomly selected. There would have been a homogeneous trend towards men in these rituals. However, the mobility and localisation patterns suggest a greater diversity, which could also imply social and cultural differences among the individuals of the two communities,” says the UAB researcher.

Isotope analysis reveals differences between the sites

To carry out the study, the research team combined bioarchaeology and the analysis of stable strontium and oxygen isotopes in the dental enamel of seven severed skulls of men recovered from Puig Castellar and Ullastret, together with archaeozoological data and a detailed sampling of sediment and vegetation collected in the vecinity of the sites. The results of the strontium isotopes of the sediment and vegetation allowed researchers to define the reference range of the strontium in the area near each site (bioavailable strontium). This in turn made it possible to discern which individuals coincided or not with this range and, therefore, identify whether they were local or not.

“At Puig Castellar the isotope values of three of the four individuals differ significantly from the local strontium reference, which suggests that they were probably not from the local community. In contrast, Ullastret revealed a mixture of local and non-local origins. This result suggests that the practice of severed heads was applied in a different way at each site, which seems to rule out a homogeneous symbolic expression. But more research is needed to be sure”, says de la Fuente-Seoane.

The fact that in Puig Castellar the skulls were exposed in an area such as the wall makes the researchers opt for the hypothesis that the reason for their exposure was aimed at the demonstration of power and coercion, both for internal repression and towards a group outside the community. In the case of Ullastret, the two local individuals were found in a street, in the middle of the city, which suggests that they were exhibited on a wall or doorway of the adjacent houses. This fact would provide support to the hypothesis suggesting that they could have belonged to important people of this community, venerated or vindicated by its inhabitants. A third Ullastret skull, of possible foreign origin, was found in one of the external walls of the settlement, which could represent a war trophy.

New tools to help understand the Iberian society

The results of the study reveal for the first time direct evidence of human mobility patterns during the Iron Age in the northeastern Iberian Peninsula, and provide new perspectives on the territorialisation contexts of northeastern Iberia.

Previous research on Iberian territorial management suggested differences in how these societies exploited the resources surrounding them. With this study researchers were able to see that the skulls found at Puig Castellar and Ullastret also show different mobility patterns, given that the values of the humans and their relationship with the values of the area are completely different in each site. The faunal samples also reveal a very differentiated resource management, in coherence with the typology of each of the settlements.

“This differentiation reflects a dynamic and complex society with important local and external interactions. Our study is a first approach to this archaeological problem using a method that is revolutionising the way we study mobility in the past. At the same time, it suggests that the selection of individuals for the severed heads ritual was more complex than initially thought”, indicates Rubén de la Fuente-Seoane.

The study underlines the importance of integrating bioarchaeological and isotope data to improve the understanding of social structures and human interactions in the past. “We have established a local strontium reference based on a rigorous protocol, applying in humans a pioneering methodology in Catalonia that, moreover, serves as a first step towards the creation of a Catalan map of bioavailable strontium. This will favour other future studies and the group of archaeologists studying mobility”, concludes the UAB researcher.

 

Why GPT can’t think like us




Universiteit van Amsterdam





Artificial Intelligence (AI), particularly large language models like GPT-4, has shown impressive performance on reasoning tasks. But does AI truly understand abstract concepts, or is it just mimicking patterns? A new study from the University of Amsterdam and the Santa Fe Institute reveals that while GPT models perform well on some analogy tasks, they fall short when the problems are altered, highlighting key weaknesses in AI’s reasoning capabilities.

Analogical reasoning is the ability to draw a comparison between two different things based on their similarities in certain aspects. It is one of the most common methods by which human beings try to understand the world and make decisions. An example of analogical reasoning: cup is to coffee as soup is to ??? (the answer being: bowl)

Large language models like GPT-4 perform well on various tests, including those requiring analogical reasoning. But can AI models truly engage in general, robust reasoning or do they over-rely on patterns from their training data? This study by language and AI experts Martha Lewis (Institute for Logic, Language and Computation at the University of Amsterdam) and Melanie Mitchell (Santa Fe Institute) examined whether GPT models are as flexible and robust as humans in making analogies. ‘This is crucial, as AI is increasingly used for decision-making and problem-solving in the real world’, explains Lewis.

Comparing AI models to human performance

Lewis and Mitchell compared the performance of humans and GPT models on three different types of analogy problems:

  1. Letter sequences – Identifying patterns in letter sequences and completing them correctly.
  2. Digit matrices – Analyzing number patterns and determining the missing numbers.
  3. Story analogies – Understanding which of two stories best corresponds to a given example story.


A system that truly understands analogies should maintain high performance even on variations

In addition to testing whether GPT models could solve the original problems, the study examined how well they performed when the problems were subtly modified. ‘A system that truly understands analogies should maintain high performance even on these variations’, state the authors in their article.

GPT models struggle with robustness

Humans maintained high performance on most modified versions of the problems, but GPT models, while performing well on standard analogy problems, struggled with variations. ‘This suggests that AI models often reason less flexibly than humans and their reasoning is less about true abstract understanding and more about pattern matching’, explains Lewis.

In digit matrices, GPT models showed a significant drop in performance when the position of the missing number changed. Humans had no difficulty with this. In story analogies, GPT-4 tended to select the first given answer as correct more often, whereas humans were not influenced by answer order. Additionally, GPT-4 struggled more than humans when key elements of a story were reworded, suggesting a reliance on surface-level similarities rather than deeper causal reasoning.

On simpler analogy tasks, GPT models showed a decline in performance decline when tested on modified versions, while humans remained consistent. However, for more complex analogical reasoning tasks, both humans and AI struggled.

Weaker than human cognition

This research challenges the widespread assumption that AI models like GPT-4 can reason in the same way humans do. ‘While AI models demonstrate impressive capabilities, this does not mean they truly understand what they are doing’, conclude Lewis and Mitchell. ‘Their ability to generalize across variations is still significantly weaker than human cognition. GPT models often rely on superficial patterns rather than deep comprehension.’

This is a critical warning for the use of AI in important decision-making areas such as education, law, and healthcare. AI can be a powerful tool, but it is not yet a replacement for human thinking and reasoning.

Article details

Martha Lewis and Melanie Mitchell, 2025, ‘Evaluating the Robustness of Analogical Reasoning in Large Language Models’, In: Transactions on Machine Learning Research.

 

Measuring poverty better to strengthen tuberculosis research





Boston University School of Medicine





(Boston—Tuberculosis (TB) has long been recognized as a disease of poverty, yet most TB research does not measure poverty in a meaningful way. A new review in the journal BMC Global and Public Health examines existing methods for assessing socioeconomic status in TB studies and highlights their shortcomings. The authors call for better, standardized poverty metrics to improve research and policy.

Led by researchers from Boston University Chobanian & Avedisian School of Medicine, Brown University and Oxford University, the review explores different ways poverty has been measured in TB studies—including income-based measures, wealth indices and multidimensional poverty indices (MPIs). The authors found that many commonly used tools fail to capture key aspects of deprivation relevant to TB risk and treatment, such as food insecurity, overcrowded housing and access to healthcare.

“Poverty isn’t just background context—it’s central to understanding who develops TB and who struggles with treatment,” says corresponding author Pranay Sinha, MD, assistant professor of medicine at Boston University. “Yet, many TB studies rely on outdated or oversimplified measures of socioeconomic status, limiting what we can learn and how we respond.”

As part of the review, the researchers also assessed how frequently observational TB studies accounted for poverty and found that of the 100 most recently published observational TB studies, nearly 70% did not include any measure of socioeconomic status. According to the authors, this lack of measurement can lead to misinterpretations of TB risk factors and policies that fail to reach the most vulnerable.

“If we don’t measure deprivation well, we can’t address it effectively,” said co-first author Chelsie Cintron, MPH, a third-year doctoral student at Brown University and a senior research study coordinator at Boston Medical Center. “We found that most TB research does not include even basic socioeconomic data. That’s a missed opportunity because poverty affects every stage of the TB journey—from exposure to treatment outcomes.”

The researchers argue that adopting tools like the global multidimensional poverty index – which has been refined and applied in various contexts by co-senior author Jakob Dirksen MSc, MPP, a research and policy officer at Oxford Poverty and Human Development Initiative – could lead to more targeted interventions and stronger policy recommendations."

This research was supported by the National Institutes of Health, the Warren Alpert Foundation, the Burroughs Wellcome Fund/American Society for Tropical Medicine and Hygiene, and the Government of India’s Department of Biotechnology and Indian Council of Medical Research.

 

PFAS: Found 180 times more ‘forever chemicals’ in birds



New technology allows for much more precise measurements of these chemicals in organisms.


Norwegian University of Science and Technology

PFAS toxins 

image: 

New methods show that birds absorb more PFAS toxins than previously demonstrated.

view more 

Credit: Photo: Louis Westgeest, NTNU





Researchers studying birds and the food they eat are now finding much larger volumes of the toxic PFAS chemicals than before. The substances either never break down or degrade very slowly, which is why they are called ‘forever chemicals’.

PFAS (per- and polyfluoroalkyl substances) are a large group of synthetic environmental toxins, and you are most likely full of them too. Forever chemicals do not break down; instead, they accumulate in the natural environment and inside your body.

“PFAS have received a lot of attention in recent years. This is because they are so widely used in industry, at the same time as these substances can also be harmful to many different organisms,” said postdoctoral fellow Junjie Zhang, who was recently affiliated with the Norwegian University of Science and Technology (NTNU).

He is the lead author of an article that addresses new findings regarding the toxins. PFAS contain fluorine, and they have received particular attention in Norway because they are used in ski wax, Teflon and fire retardants.

Possible and confirmed harmful effects include various forms of cancer, liver damage, cholesterol disorders, reduced fertility, hormone disorders, developmental disorders in children, and a weakened immune system.

Finding more PFAS than before

Ideally, you do not want these substances in your body, but in practice, it is virtually impossible for humans and many other living organisms to avoid them.

Recent research and a new method for detecting PFAS bring both bad and good news. The bad news is that we are finding PFAS in places we have not previously found them. The good news is that this means we have become better at detecting these substances.

“The biggest increase is in the livers of wading birds. We found up to 180 times more PFAS than previously,” said Zhang.

Some of the increase may be due to a new analysis method.

“This suggests that previous methods have not been good enough at detecting certain types of PFAS,” said Zhang.

He was affiliated with the Department of Chemistry at NTNU during the study and collaborated with Professor Veerle Jaspers at the Department of Biology on a project funded by the Research Council of Norway (COAST IMPACT). He is now a postdoctoral fellow at the University of Copenhagen.

Investigating migratory birds and their food

The international research group analyzed samples from migratory birds and the shellfish they eat.

“The East Asian–Australasian Flyway is an important route for millions of migratory birds, including wading birds,” said Jaspers.

As the name suggests, this migration route extends between Siberia and East Asia and large parts of Australia.

However, the populations of many bird species along this migration route are rapidly declining. The researchers wanted to find out whether exposure to environmental toxins could be a contributing factor.

The researchers took samples from 25 wading birds. In addition, they collected samples from 30 shellfish found in areas of China where migratory birds often stop to feed. This is because it is common for birds – and humans for that matter – to ingest PFAS through food and water.

Easier to detect substances

The researchers took both liver and blood samples from the birds. They used a new method to analyze the samples, called the Total Oxidizable Precursor (TOP) assay, developed by co-author Lara Cioni. This method makes it easier to detect certain types of PFAS.

A lot of research has been done on one group of PFAS called PFAAs (perfluoroalkyl acids), but little is known about the substances that can be converted into PFAAs. PFAAs are formed when other substances break down, and it is these substances that are more easily detected using TOP.

“The TOP results show a significant increase in several types of harmful substances in all of the samples,” said Zhang.

Some of the findings suggest that many forever chemicals originate from sources we are not yet aware of, which is not particularly good news.

According to the researchers, the findings highlight how important it is to conduct more research on the substances that PFAS come from.

“We need to find out more about the sources, but also about the effects PFAS have on wading birds, other animals and humans,” said Jaspers.

Reference: Junjie Zhang, Lara Cioni, Veerle L.B. Jaspers, Alexandros G. Asimakopoulos, He-Bo Peng, Tobias A. Ross, Marcel Klaassen, Dorte Herzke. Shellfish and shorebirds from the East-Asian Australian Flyway as bioindicators for unknown per- and polyfluoroalkyl substances using the total oxidizable precursor assay, Journal of Hazardous Materials, Volume 487, 2025, 137189, ISSN 0304-3894. https://doi.org/10.1016/j.jhazmat.2025.137189


Ruddy turnstones ingest PFAS both through food and water. 

Credit

Photo: Louis Westgeest, NTNU.