This is happening soon! (3:15 PM Greenwich time)

New publication PNASNews: Frequent winners explain apparent skewness preferences in experience-based decisions pnas.org/doi/epdf/10.10… With Mikhail Spektor and Gael Le Mens Thread(1/7)>>>

Large language models show promise as cognitive models. The behaviors they produce often mirror human behaviors, suggesting we might gain insight into human cognition by studying LLMs. But why do LLMs behave like humans at all?🤔🧵arxiv.org/abs/2405.19313

Update to our paper on presentation format and context effects in choice: We added a second experiment showing that the *commensurability* of attributes is critical: attraction and compromise effects disappear when attribute values are on the same scale. osf.io/preprints/psya…

In prepping to teach graduate-level stats this fall, I have finally discovered the beauty of LaTeX for presentations. I may never open PowerPoint again.

This summer, I had the pleasure of mentoring 3 computer science majors and 1 psychology major on research at the intersection of artificial intelligence and cognitive science. Thanks to the Binghamton University⁩ BUPNUR program for making this possible! binghamton.edu/news/story/505…

New preprint out! Context induces distortions in value representations: a test across multiple elicitation methods and learning modalities, where we significantly extend our understanding of context effect in learning and DM by Magdalena Soukupova w/ basilegrc osf.io/preprints/psya…

Now is a great time to join the Psychology Department at Binghamton University! We’re aiming to hire an Associate or Full Professor specializing in computational cognition and mental health. Please share! binghamton.interviewexchange.com/jobofferdetail…

Nicolas Yax will present "Prompt Sensitivity in Language Models: Variations and Stability in Bandit Tasks" (machine behavior 🤖) joint with William Hayes & Georgia Turner Saturday, Oct 12th, 2024 from 2:30pm-4:00pm

In a new paper out from our lab with first-author Mianzhi (Rudolf) Hu and Hilary Don, we present a model where uncertainty, modulates control of decision-responses between a Gaussian system that learns exact reward values, and a Dirichlet system that learns reward frequency.

In our latest preprint, we show how constraining a joint RL/sequential sampling model with eye-tracking data allows it to predict gaze biases in decisions from experience: biorxiv.org/content/10.110…