ORISE Intelligence Community Postdoc @Harvard || agency, affect, & uncertainty in learning, decision-making, & computational psychiatry || www.hayleydorfman.com
Hayley Dorfman
Loading...
Our computational model shows that ambiguous evidence & prior beliefs are combined to infer what probably happened.
This produces confirmation-like effects: if you believe things are better or worse, you'll continue to update beliefs about ambiguous outcomes in a valence-consistent way.
Most reinforcement learning studies use clear outcomes - reward or punishment. But real life isn't always so black & white.
We developed a task where people saw the magnitude of an outcome, but not always whether it was positive or negative.
We often interpret ambiguous feedback in ways that confirm what we already believe.
A laugh can feel supportive or mocking.
A neutral email can feel encouraging or disappointing.
A new paper studies how this happens during reward learning.
Hayley Dorfman
Hayley Dorfman
The LUMeN Lab at Emory is hiring ✨
We’re seeking a full-time RA/lab manager and a postdoc to work on new NIH-funded studies! Formal job ads/apps forthcoming. We’re eager to hit the ground running so please share these flyers and/or reach out if interested!
Cognition
We also show that optimists interpreted ambiguous outcomes more positively, but didn't earn more reward.
We think that these individual differences in how we resolve ambiguity during learning could be relevant to understanding aspects of mood and anxiety disorders.
Also want to add a big thank you to our reviewers, who took the time to provide thoughtful and helpful suggestions and truly helped make this paper better. Doesn't always happen that way!
Postdoc position available at postlab.psych.wisc.edu. Pls send cv to [email protected]
🌵🏜️🌵 Out now in @cognitionjournal.bsky.social with @rbhui.bsky.social!
If your advisor sends you an unclear email, do you interpret it as good or bad? 😏😱
In a new paper, we show how people make inferences about this type of ambiguous feedback during learning.
www.sciencedirect.com/science/arti...
Post 6 “Ambiguity and confirmatory reward learning”
New paper from Hayley Dorfman (@hayleydorfman.bsky.social ) and Rahul Bhui ( @rbhui.bsky.social )
Read more here: www.sciencedirect.com/science/arti...
Hayley Dorfman
Ali Cohen
Hayley Dorfman
Our interests in human memory and cognition encompass the cognitive and neural basis of working memory, attention, control, and consciousness.News Madison Symposium on Memory & Control. On 30 May 202...
We tend to interpret feedback in ways that confirm our pre-existing beliefs. Such confirmatory tendencies are often viewed as cognitive flaws, but mig…
