@carolinepioger.bsky.social @juliegrezes.bsky.social @tarrynbalsdon.bsky.social @charlottejacquemot.bsky.social

@romanececchi.bsky.social @maevalhotellier.bsky.social @mconstancecorsi.bsky.social @constancedestais.bsky.social @sophiebavard.bsky.social et moi

It probably depends on the licence your papers are published under. In any case, you can argue that it is not a fair use (because commercial and you are not cited at all). You can check your rights here openpolicyfinder.jisc.ac.uk

We have many more cool results in the paper (such as transitions from volatile to stable environments and vice-versa) check it out! And huge thanks to co-authors Leo Sperber and @stepalminteri.bsky.social

TL;DR: positivity bias is more robust, choice history effects are environment-dependent. Both asymmetric update and choice history effects co-exist!

Okay but is there a versatile strategy that works across tasks? We simulated two mega tasks, one with all stable and another with all volatile environments. We observed that positivity bias emerged in both cases, but in the first environment we observed perseveration and in the second alternation.

The script flips in volatile environments: negativity bias progressively emerged as volatility increased. This time, we also observed a strong tendency for alternation, intensified as the reversal frequency increased. These results were consistent across all reversal probability distributions.

We found that in stable (no reversal) environments, positivity bias emerges every time but in rich environments, replicating the results by Cazé and Van der Meer. On the flip side, perseveration only emerged when long learning periods were involved

We used an evolutionnary algorithm to find the optimal set of parameters in each of these environments, evolving 1000 agents through 200 generations.

We ran simulations of 2-armed bandit tasks using a Q-learning model with both an asymmetric update rule and choice history bias. We tested out several difficulty levels, environment richness, learning periods, reversal frequency and probability distribution.

We knew empirically that people tended to repeat past choices, but the mechanism uderlying repetition remained debated: is it perseveration, a choice repetition effect, or positivity bias, an asymmetric learning of +/- rewards? Is it always the case, regardless of contingencies and task structure?

Sounds nice! How do I join?

I don't know if I have the technical (coding) expertise needed for that, but I definitely have some front-end/graphics design experience if that could help!

..Is one where the user does not have to spend hours figuring out things before they can create something okay-ish. Also, the automated slide layout in PowerPoint makes it so straightforward to make impeccable presentations. LibreOffice in that regard is just not it...

I'd be happy to contribute! On top of my mind (but I last used LibreOffice a year ago), the GUI on either Writer or Impress were not quite user-friendly (Windows 95 vibes). Not so easy to align things on slides either. Probably it all goes in the "laziness" category, but I believe that a good tool..

Hi, this is an interesting idea! RL is indeed quite used to explain social learning, and I'd be happy to discuss that with you :)