Elizabeth J. Lee

Investigating Invariances in Auditory Event Categorization with Model Metamers

This study evaluates how well deep neural networks align with human perception in categorizing natural sounds by testing model metamers across different architectures and training tasks.

A two-dimensional space of linguistic representations shared across individuals

The paper identifies two interpretable dimensions of linguistic processing using fMRI data, showing they generalize across individuals and relate to processing difficulty and meaning abstractness.

Linguistic inputs must be syntactically parsable to fully engage the language network

The paper tests whether syntactic parsing is necessary for language comprehension by analyzing brain responses to syntactically ill-formed but semantically plausible and grammatically correct but semantically implausible sentences.