Lightnews — Scholar-powered news

Reposted by Ben Kawam

Zoë Goldsborough @zoegoldsborough.bsky.social · 2d

Want to know more about monkeys kidnapping other monkeys?🐒 I had an amazing chat together with @bjjbarrett.bsky.social on @sidedoorpod.bsky.social about the Coiban capuchins and their wild antics. Science really is stranger than fiction! Listen 👂 here: www.si.edu/sidedoor/mon...

3 4

Reposted by Ben Kawam

Aki Vehtari @avehtari.bsky.social · 2d

I'm looking for a doctoral student with Bayesian background to work on Bayesian workflow and cross-validation (see my publication list users.aalto.fi/~ave/publica... for my recent work) at Aalto University.

Apply through the ELLIS PhD program (dl October 31) ellis.eu/news/ellis-p...

ELLIS PhD Program: Call for Applications 2025

The ELLIS mission is to create a diverse European network that promotes research excellence and advances breakthroughs in AI, as well as a pan-European PhD program to educate the next generation of AI...

ellis.eu

32 40

Reposted by Ben Kawam

Ari Strandburg-Peshkin @arispeshkin.bsky.social · 5d

⭐PhD position available!⭐
Come join us @imprs-qbee.bsky.social to study communication and collective behavior in animal groups! We're looking for someone excited to use computational approaches to tackle biological questions, using our full-group tracking datasets
imprs-qbee.mpg.de/121465/analy...

Analysis of communication and collective behavior in animal groups

imprs-qbee.mpg.de

63 42

Ben Kawam @benkawam.bsky.social · 8d

Thanks to everyone involved: @danielredhead.bsky.social, @primbehavecol.bsky.social, and @rmcelreath.bsky.social!

Ben Kawam @benkawam.bsky.social · 8d

Are you studying animal sociality?

My (now published) first PhD chapter proposes a Bayesian + causal framework to infer the factors shaping the social relationships that animals form with one another.

journals.plos.org/ploscompbiol...

A causal framework for the drivers of animal social network structure

Author summary Behavioural ecologists ask mechanistic questions about behaviour—causal questions. When studying animal societies, these questions often concern the drivers of social network structure....

journals.plos.org

1 16 37

Reposted by Ben Kawam

Helen Mylne @helephantmylne.bsky.social · 14d

3yr PDRA at @exeter.ac.uk with @york.ac.uk and Centre for Whale Research, studying how killer whale social relationships change with prey availability, partner loss and social info. A fantastic team & unique longterm dataset. Deadline 19 Oct!
www.jobs.ac.uk/job/DOT336/postdoctoral-research-associate

Postdoctoral Research Associate at University of Exeter

Apply for the Postdoctoral Research Associate role on jobs.ac.uk, the top job board for academic positions in higher education. View details and apply now.

share.google

16 12

Reposted by Ben Kawam

Jean-François Bonnefon @jfbonnefon.bsky.social · 18d

It's hiring season at @iast.fr!

- 2y research postdoc contract
- Full autonomy, you are your own PI
- Awesome multidisciplinary environment
- All social and behavioral sciences welcome
- Seed funding for projects and workshops
- Gorgeous city in the south of France

www.iast.fr/research-fel...

50 61

Reposted by Ben Kawam

Andrew Gelman et al. @statmodeling.bsky.social · 19d

When thinking about causal inference, mechanistic or process models are important. I think that the association of “causal” with black-box models leads to lots of problems.
statmodeling.stat.columbia.edu/2025/09/19/w...

When thinking about causal inference, mechanistic or process models are important. I think that the association of “causal” with black-box models leads to lots of problems. | Statistical Modeling, C...

statmodeling.stat.columbia.edu

4 19

Reposted by Ben Kawam

Dieter Lukas @dieterlukas.fediscience.org.ap.brid.gy · 19d

'The Desperation of Causal Inference in Ecology'

"What’s the solution? If you ask me, be less gaga over any statistical method and teach everyone basic biological models, simulate data from it and then fit their statistical model to it." […]

Original post on fediscience.org

fediscience.org

2 5

Reposted by Ben Kawam

Afra Salazar @salazarafra.bsky.social · 21d

Shot from La Jetée (1962) by Chris Marker.

1 4

Ben Kawam @benkawam.bsky.social · 28d

Thanks to everyone involved—@danielredhead.bsky.social, @jgyou.bsky.social, Dan Franks, Connor Philson, Marijtje van Duijn, @jordanhart96.bsky.social, MB McElreath, @rmcelreath.bsky.social, @eapower.bsky.social, Sebastian Sosa, Cody Ross, @steglich.bsky.social, Michael Weiss & @ljnbrent.bsky.social!

3

Ben Kawam @benkawam.bsky.social · Sep 4

If you’re interested in the use of causal diagrams for animal social network analysis (I promise it quickly starts to make sense!), we also have this paper out, bringing causal graphs together with Bayesian multilevel models.

www.biorxiv.org/content/10.1...

A causal framework for the drivers of animal social network structure

A major goal of behavioural ecology is to explain how phenotypic and ecological factors shape the networks of social relationships that animals form with one another. This inferential task is notoriou...

www.biorxiv.org

1

Ben Kawam @benkawam.bsky.social · Sep 4

For the full paper, see ecoevorxiv.org/repository/v....

Please do not hesitate to reach out with questions or comments!

Five misunderstandings in animal social network analysis

ecoevorxiv.org

2 2

Ben Kawam @benkawam.bsky.social · Sep 4

The key is to make causal assumptions explicit, and to clearly define what (usually causal) quantity the analysis is trying to estimate.

(This is not unique to animal social network analysis!)

1 1

Ben Kawam @benkawam.bsky.social · Sep 4

Here, for instance, the graphical rules tell us that that controlling for Sab is necessary to correctly capture the effect of X on y. That is, we control for the “spurious” part of the dependency structure, and interpret the remaining association pattern causally.

1

Ben Kawam @benkawam.bsky.social · Sep 4

By using graphical assumptions about how the network data were generated, we can figure out which part of the dependency structure is our inferential target, and which part threatens our inferences.

1

Ben Kawam @benkawam.bsky.social · Sep 4

Consider now individual-level features (age, personality, genotype) affecting how they interact with others. E.g., the young age X of individual “1” causes it to socialise more across partners 2, 3, 4—resulting in high values for y12, y13, y14 (i.e. in their inter-dependency).

1 1

Ben Kawam @benkawam.bsky.social · Sep 4

For example, here, variation in sampling effort (Sab) results in yab and yba to be associated, i.e. non-independent.

Note that the graphs do *not* represent the social networks themselves, but the process that generated the social network edges y.

1 1

Ben Kawam @benkawam.bsky.social · Sep 4

To illustrate this point, let’s use graphs representing causal relationships (arrows) between variables (nodes). General graphical rules tell us which variables are associated given a certain causal structures (e.g., fork, chain), and which variables to control for to block this association.

1

Ben Kawam @benkawam.bsky.social · Sep 4

After a few years of possibly unhealthy exposure to causal & Bayesian inference, things did get better. Today I believe it is (sometimes) possible to make sense of such dependencies by viewing them, not as threatening or interesting in themselves, but as symptoms of *underlying causal mechanisms*.

1

Ben Kawam @benkawam.bsky.social · Sep 4

It felt as if it would have been good to rely on some sort of principles.

1

Ben Kawam @benkawam.bsky.social · Sep 4

As a student trying to make sense of primate societies, I was very confused. It also felt wrong to use statistical methods that no one seemed to agree upon in the literature, and which gave different answers depending on seemingly arbitrary modelling choices.

1 1

Ben Kawam @benkawam.bsky.social · Sep 4

Are you studying animal sociality? Confused about the "non-independence" of social network data: what on earth are they? are they threatening your results? should you attempt to get rid of them?

A thread summarising a part of our new paper (ecoevorxiv.org/repository/v...).

Five misunderstandings in animal social network analysis

ecoevorxiv.org

1 11 19

Ben Kawam @benkawam.bsky.social · Sep 4

After a few years of possibly unhealthy exposure to causal & Bayesian inference, things did get better. Today I believe it is (sometimes) possible to make sense of such dependencies by viewing them, not as threatening or interesting in themselves, but as symptoms of *underlying causal mechanisms*.

Ben Kawam @benkawam.bsky.social · Sep 4

It felt as if it would have been good to rely on some sort of principles.

1