Modeling Attention during Dimensional Shifts with Counterfactual and Delayed Feedback

Tyler Malloy, Roderick Seow, Cleotilde Gonzalez · 2025

Read Modeling Attention during Dimensional Shifts with Counterfactual and Delayed Feedback by Tyler Malloy, Roderick Seow, Cleotilde Gonzalez (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-11161.txt.

Description

Malloy, Seow and Gonzalez wrote a 2025 paper on how people aim their focus when a task changes. In choice tasks, people can use focus even on features they have not seen before. That matters when new feature values appear and links to rewards stay fixed or shift. Most past work used fast feedback, though.

The team compared two models of focus. One builds a score from a memory of past events, the other updates focus from how wrong reward guesses were. Tests used tasks with shifts inside or across feature types, and with fast, slow or what-if feedback. The memory-based score fit human-like behavior best across these cases.

Catalog record

CREATORTyler Malloy, Roderick Seow, Cleotilde Gonzalez
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE19 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-11161

Browse the full library · About the Tower of Babel