A nice example of how sequential and simultaneous choice can fundamentally differ: in the latter, the longer a subject waits to decide, the more variable their decision time.

We show foraging decisions can have independence of decision time and variability, or even an inverted relationship!

End 🧵

In another weird prediction, we show that if the reward in a patch decays linearly when harvested, then the forager should be *more* variable the *earlier* they leave

Also exactly what we see in data: foragers leave earlier in rich environments but are more variable (data, solid; model, dashed)

Perhaps the weirdest prediction is that, under a wide range of conditions, foragers’ stochasticity is independent of when they leave. In other words, their variability is decoupled from their reward information

And that’s exactly what we see in the data (solid lines; model predictions: dashed)

We ask if foragers’ variability can be explained by them making deliberately stochastic leaving choices: basically, whether they flip a biased coin

We show deliberately stochastic choice makes weird predictions for how foragers’ respond to their environment, and test them across tasks and species

Plus, there's now a related conference organized by some wonderful people @unibirmingham.bsky.social in the UK, including @brainapps.bsky.social, @markdhumphries.bsky.social, and others. (Registration for this conference is still open until the 20th October!) uobevents.eventsair.com/the-mechanis...

Thank you, looks useful!

Thanks! The NRN covers a lot, so may not be so accessible for the audience I’ve in mind; will revisit Wang’s follow up in TINs, may well be the answer!