𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( )

Postdoc at ETHz - Ass. Editor @HSScomms
Exploring Opinion dynamics and Collective intelligence with social simulations and belief networks.

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

This is a the network of political attitudes in the US in 2021. (Thus, nodes - i.e. the circles - are not people🧑 but attitudes🗣️).

In the top figure you see the colors based on response level to a survey (i.e. red=the most Republican response, blue=the most Democrat, grey=the neutral response, etc).

Interestingly, we find two very weird clusters. Indeed, while one cluster is mainly Democrat and the other mainly Republican, we see that the neutral attitudes are completely absorbed in the Republican cluster.

Even more surprising: three attitudes which should be mildly democrat (e.g. Abortion should be illegal: somewhat disagree) are still in the Republican cluster!

This is confirmed by looking at the self-identity associated to every attitude. Indeed, all but one of the attitudes in the Republican cluster were chosen by self-identified Republicans.

Why is that?

What these graphs are telling us is that Democrats had a very clear positioning on these items - i.e. if you are a Democrat you will strongly agree/disagree.

On the contrary, Republicans seemed to be way more scattered across different positions, ranging very widely from "Strongly Republican" to even include some "Mildly Democrat" positions.

Interestingly, this also tells us that expressions of neutral positions, will probably not be perceived as neutral, but as Republican attitudes...

Of course, this is a very simplified extract, and you can find all the detailed and rigorous information in the publication:📝
https://t.co/GP66d3MCfx

An even deeper study on the dataset can be found here:
https://t.co/KLYFhekkmG

Thank you to @Quayle and @Adrian_Lueders for their wonderful work on these articles!

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Today at 15:15 CET I'll be speaking at the Social Simulation Fest about Science Communication and Narratisation! 🎤

It's an online event, so you can still join by registering here: 👇
https://t.co/IhwWDwbBbx

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Opinion dynamics models do _not_ need to be an abstraction of reality. You can also build them to reproduce measured human behaviour in experiments.

https://t.co/T3LgJwtlO1

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

It took me a while, but we finally have the second episode of the Computational Social Science podcast!

This time, we talk about collective intelligence 🧐

You can find it on YouTube...
https://t.co/C3JlVxVMsZ

...or on your favourite platform:
https://t.co/xvLsa65DJF

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Just found out that the paper where Galton introduced correlation has fewer than 1,000 citations. It makes sense that people don’t cite it every time, but that still feels surprisingly low!

https://t.co/l8FuPMfZao

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

@juleeyoh @OzzyMag @I_never_do_this @Laiize1337 @meishato Also, we never measured open-mindedness, and it's not possible to extract it from our data.
(I'm one of the authors of the study btw)

https://t.co/vlZ0VNaW0I

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Year 3 of asking ChatGPT to write a joke about ABMs.

Main finding: Instead of producing better jokes, it’s just becoming aware that they’re simply not funny.

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Note: since this is very innovative, we still don't have "the whole system". Here we show that a completely "flat" system could work even with millions of people collaborating together. But future works should explore potential exploits, limitations, improvements, etc.

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

I mean, at a certain point they even wrote that we claimed something when we actually claimed the exact opposite... 😅

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

So, somebody wrote a comment on one of my articles. Their core message is rather sound (indeed it's the same as what we wrote in our article).

The problem is that they make so many mistakes I don't really know what to make of this...

https://t.co/cYCnDiZrA8

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Today we have the first public event of the Special Interest Group on Strongly Empirical Modelling, with a talk from Bruce Edmonds titled: Towards inferring clusters of “cognitive models” from survey data. See you at 2pm CET!
https://t.co/HpQyNFZ3X3

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Why do people prefer using statistical tests that nobody ever heard of instead of just bootstrapping? (I'm really asking, it's not a complaint or sarcastic or anything)
Wouldn't bootstrapping also make results more understandable?

[Comput. Soc. Science @COSS_eth]

A talk between @JustaNormalDino and @dangerdailisan about remixing and collective intelligence!

https://t.co/pnrxE3gjzY

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Another figure from a (now retracted) scientific publication. I'm not an expert, but I feel proportions may be a little off...

https://t.co/McE5l0tR8w

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Yes, this is a real image from a scientific publication 😅

https://t.co/8XiYHHVXrU

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Some years ago I applied ResIN to a Harry Potter dataset. The more two people are similar (in terms of how they relate to other characters), the tighter their connection. As you can see, the HP world is heavily polarized.
But now my question is: why did I do something like that??

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Apparently ignoring all those predatory journals made me very popular with girls!
(btw, all these emails are from a single day)

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

To cite this meme:
Photoshop et al. (2025)

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

How can we develop a better understanding of Complex Attitude-Identity Systems (CAIS)? Here👇we introduce a theory of Belief Networks to explain how to accurately measure them. This will help us understand how beliefs connect, evolve and shape identity.

https://t.co/PIZioPIBcB

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

Today, both Samuel Moor-Smith and I will give a presentation at the Social Simulation Conference. If you are interested in the validation of ABMs, their classes of equivalence and how psychometrics plays an important role, join us!

[𝐷𝑖𝑛𝑜.𝐶𝑎𝑟𝑝𝑒𝑛𝑡𝑟𝑎𝑠( ) @JustaNormalDino]

I am finally back from holid... what happened? Did I sleep for 50 years?