Shape Guides Visual Pretense

Peng Qian, Tomer D Ullman · 2025

Read Shape Guides Visual Pretense by Peng Qian, Tomer D Ullman (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/pmc-12768553.txt.

Description

Peng Qian and Tomer Ullman study what guides pretend play with objects, in a twenty twenty-five paper. People often treat one thing as another: a bottle becomes a car, a teapot a swan. The pairing feels flexible but not random. Across four studies with 716 people, the team compared human choices with those of vision AI models. They weighed shape likeness against surface traits such as color.

People favored matches in shape and in how parts line up between the real and pretend objects. Surface likeness mattered far less. Current vision models missed the human pattern, leaning on surface traits instead. The authors say pretense rests on spatial and physical fit.

Catalog record

CREATORPeng Qian, Tomer D Ullman
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE62 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC12768553

Browse the full library · About the Tower of Babel