Social networks and natural selection in changing societies

Informazioni relative al progetto

KinSocieties

ID dell’accordo di sovvenzione: 101098266

DOI

10.3030/101098266

Data della firma CE 28 Agosto 2023

Data di avvio 1 Gennaio 2024

Data di completamento 31 Dicembre 2028

Finanziato da

• European Research Council (ERC)

Costo totale

€ 2 499 971,00

Contributo UE

€ 2 499 971,00

Coordinato da

TURUN YLIOPISTO
Finland

Periodic Reporting for period 1 - KinSocieties (Social networks and natural selection in changing societies)

Periodo di rendicontazione: 2024-01-01 al 2026-06-30

Sintesi del contesto e degli obiettivi generali del progetto

Social interactions can improve support, resources and protection, but can also increase disease, stress and conflict. Consequently, for group-living species inc. humans, traits as diverse as personality, residence patterns, family-living and disease resistance all evolve in response to pros and cons of sociality. However, the direct links between sociality, health, fitness outcomes and ultimately natural selection are not well-known. KinSocieties reveals, for the first time, the benefits and costs of sociality accrued by individuals and whole societies in two complementary study species - humans and Asian elephants - facing current drastic changes in social structures due to break-down of kin networks. We use rare longitudinal data and two previously unstudied natural experiments to investigate effects of translocation to new social environments, addressed in 6 WPs:

How have human social networks transformed with the modernisation of societies and associated with reproduction, cause of death and lifespan at different times?
How does population structuring in humans affect reproduction, cause of death and lifespan?
How does population structuring in Asian elephants affect reproduction, cause of death and lifespan?
How is translocation to new area with/without kin or friends in humans related to subsequent integration, social networks, reproduction, lifespan and cause of death? How is translocation to new working units with or without kin, friends or social group in Asian elephants associated with stress, health, and friendship formation?

This research boldly combines approaches from social sciences, conservation, evolutionary demography and biology. The results have key theoretical importance in several fields as well as practical consequences, making also critical contributions to public health by revealing concrete costs and benefits tied to social relationships and their changes in the rapidly changing world of today.

Lavoro eseguito dall’inizio del progetto fino alla fine del periodo coperto dalla relazione e principali risultati finora ottenuti

To construct the data used in this project, we have built a fruitful collaboration with natural language processing experts to develop large language models to digitalise a historical collection of 89,339 brief Finnish language interviews of refugee families relocated post-WWII from Finnish Eastern Karelia. We have extracted social organizations and hobbies from the free text of the interviews, separately for each family member. These can act as a proxy variable indicating the degree of social integration of refugees in their new environment. At a second step, we used a highly innovative method to process the extracted 350K mentions of leisure time activities and organizational memberships from these interviews, totalling 71K unique activity and organization names—far too many to analyze directly. We developed a categorization framework that captures key aspects of participation (the kind of activity/organization, how social it typically is, how regularly it happens, and how physically demanding it is). We annotated a goldstandard set to allow for a reliable evaluation, and then tested whether large language models can apply the same schema at scale. Using a simple voting approach across multiple model runs, we found that an open-weight LLM can closely match expert judgments. Finally, we applied the method to label the 350K entities, producing a structured resource for downstream studies of social integration and related outcomes. We are now using this novel pipeline to also analyse the elephant logbooks in a highly interdisciplinary manner. We have also applied a deep-learning pipeline to create a structured dataset of internal migration in Finland between 1800 and 1920 using digitized church moving records that includes over six million entries extracted from approximately 200,000 images of handwritten migration records.
Using these data we have so far quantified the long-term changes in the structure of family networks, and how they vary across lifespan using unique longitudinal kin networks data that covers over 300 years of economic and demographic development in Finland. We have shown how such variation in network structures is associated with reproduction parameters and survival in both childhood and later-life, highlighting differences between maternal and paternal kin.

We have shown socioeconomic differences in movement probability and vaccination coverage and quantified how lifespan and cause-specific mortality vary across the country. Our first results of the social drivers of epidemics show that dialect groups explained spatial variation of pertussis, administrative regions for smallpox, while measles showed no broader scale spatial variation.

We have also thus far shown in elephants that the availability of allomothers within the social unit (working group) affects the stress levels and body condition of both mother, calf and allomother.

Progressi oltre lo stato dell’arte e potenziale impatto previsto (incluso l’impatto socioeconomico e le implicazioni sociali più ampie del progetto fino ad ora)

The most novel and unconventional outcome from KinSocieties are the multi-faceted, unique datasets that we are currently building for the use of cross-disciplinary research. These are achieved through highly interdisciplinary developments and the work to utilise them is under way.