An independent consensus was slightly more persuasive than a dependent consensus on average (see red vs. blue above), but this was underscored by substantial individual variability (see bottom panel below).

A consensus (relative to disagreement) was very persuasive, but the strength of this effect depended on how “knowable” the claim was: how likely it was to have a ground truth.

Ps rated their belief in 60 real world claims before and after seeing 4 social media posts either 1) reaching a consensus using independent sources, 2) using dependent sources, or 3) disagreeing (contested).

We know that a consensus of opinions is persuasive, but how reliable is this effect across people and types of consensus, and are there any kinds of claims where people care less about what other people think? This is what we tested in our new(ish) paper in @psychscience.bsky.social

We also saw much more overall engagement in both confederate conditions, likely because politically charged content tended to draw more activity, and the confederates only posted political content. 5/7

Interestingly, participants (especially those on the left) disliked the Right-wing confederate condition more than the others. Alternatively, in the Left-wing confederate condition, all participants enjoyed the experience similarly. 4/7

We found small but statistically significant effects suggesting that, compared to the control, the presence of Right-wing confederates led to more Right-wing belief and a greater sense of political division in the US. Left-wing confederates didn’t seem to have an effect. 3/7

Excited to share another new #CogSci paper with
@andyperfors.bsky.social
(and others not on Bluesky). We built an isolated social media platform using a Mastodon server (that we're calling “MAGPIE Social”), then recruited participants to discuss 4 political topics over several days. 1/7

We also saw much more overall engagement in both troll conditions, likely because politically charged content tended to draw more activity, and the confederates only posted political content. 5/7

Interestingly, participants (especially those on the left) disliked the Right-wing confederate condition more than the others. Alternatively, right-wing participants in the Left-wing confederate condition enjoyed the experience similarly to Left-wing participants. 4/7

We found small but statistically significant effects suggesting that, compared to the control, the presence of Right-wing confederates led to more Right-wing belief and a greater sense of political division in the US. Left-wing confederates didn’t seem to have an effect. 3/7

Building off a popular theoretical perspective that describes knowledge acquisition as a stochastic search through a space of ideas, we demonstrate how features of social reasoning can be mapped onto features of stochastic search algorithms to figure out who is more likely to know the truth. 2/3

Happy to share some new theoretical work with @andyperfors.bsky.social
that will appear at CogSci this year! We argue (and demonstrate through simulations) that people can determine how much to trust other agents by thinking about how those agents have acquired their knowledge. 1/3

Reflecting on when I built an app to analyse social media activity pertaining to the Russia/Ukraine war and it uncovered some extremely key intel #academicsky #PhDSky

Hi new followers! I’m a final yeah PhD candidate researching how social reasoning facilitates efficient cognition.

Also, I found this amazing hat at an op shop in county Victoria. @andyperfors.bsky.social, can we rename the Complex Human Data Hub to the “Data Divas Hub”?

First 🟦 sky post!

Ever struggled with allocating your time across different tasks? In this project, we looked at how cognitive constraints, like time pressure & cognitive load, affect how people prioritise different goals.

#AcademicSky #PsychSciSky #PhDSky

www.sciencedirect.com/science/arti...