Component-Level Lesioning of Language Models Reveals Clinically Aligned Aphasia Phenotypes

TL;DR

This study demonstrates that perturbing specific components in large language models can simulate aphasia-like language impairments, providing a scalable and interpretable platform for studying language deficits and recovery.

Contribution

Introduces a novel component-level perturbation framework for LLMs that reproduces clinical aphasia phenotypes, linking model components to linguistic functions.

Findings

Targeted perturbations produce more systematic aphasia-like regressions than random ones.

Modular models show more localized and interpretable phenotype-to-component mappings.

Component perturbations correlate with clinical aphasia severity measures.

Abstract

Large language models (LLMs) increasingly exhibit human-like linguistic behaviors and internal representations that they could serve as computational simulators of language cognition. We ask whether LLMs can be systematically manipulated to reproduce language-production impairments characteristic of aphasia following focal brain lesions. Such models could provide scalable proxies for testing rehabilitation hypotheses, and offer a controlled framework for probing the functional organization of language. We introduce a clinically grounded, component-level framework that simulates aphasia by selectively perturbing functional components in LLMs, and apply it to both modular Mixture-of-Experts models and dense Transformers using a unified intervention interface. Our pipeline (i) identifies subtype-linked components for Broca's and Wernicke's aphasia, (ii) interprets these components with linguistic probing tasks, and (iii) induces graded impairments by progressively perturbing the top-k subtype-linked components, evaluating outcomes with Western Aphasia Battery (WAB) subtests summarized by Aphasia Quotient (AQ). Across architectures and lesioning strategies, subtype-targeted perturbations yield more systematic, aphasia-like regressions than size-matched random perturbations, and MoE modularity supports more localized and interpretable phenotype-to-component mappings. These findings suggest that modular LLMs, combined with clinically informed component perturbations, provide a promising platform for simulating aphasic language production and studying how distinct language functions degrade under targeted disruptions.