Artificial Aphasias in Lesioned Language Models

• URL: http://arxiv.org/abs/2605.16222v1
• Date: Fri, 15 May 2026 17:33:07 GMT
• Title: Artificial Aphasias in Lesioned Language Models
• Abstract summary: Aphasias reveal the functional organization of human language by providing causal links between affected brain regions and specific symptom profiles.<n>We introduce an aphasia-inspired technique to characterize the emergent functional organization of language models.
• Score: 1.5749416770494704
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Aphasias, selective language impairments which can arise from brain damage, reveal the functional organization of human language by providing causal links between affected brain regions and specific symptom profiles. Drawing on this literature, we introduce an aphasia-inspired technique to characterize the emergent functional organization of language models (LMs). We ``lesion'' (zero-out) model parameters and measure the effects of this intervention against clinical aphasia symptoms, as diagnosed by the Text Aphasia Battery (TAB). When applied to 112,426 outputs from five 1B-scale LMs, the full range of evaluated symptoms surface, but in distributions largely distinct from those of humans. Our method uncovers broad symptom-profile differences between attention components (query, key, value, output) and feed-forward components (up, gate, down), with weaker evidence for differences among components within the same mechanism. We also find an effect of depth, where lesions in early layers disproportionately cause syntactic and semantic symptoms while late-middle layers yield higher rates of phonological and fluency deficits. Although some LM lesions induce quantitatively more similar profiles to some human aphasia types than others, qualitative differences in symptom patterns between LMs and humans suggest that aphasia syndromes are heavily influenced by the details of learning and processing rather than being a domain-invariant consequence of disrupted language processing.

Related papers

• Activation-Guided Neuron Intervention to Induce Alzheimer's-Related Computational Language Phenotypes in a Large Language Model [11.44806025415537]
  Changes in spontaneous speech provide an early signal of cognitive dysfunction in Alzheimer's disease.<n>We introduce an activation-guided intervention framework using Qwen3-8B.<n>Amplifying AD-associated neurons produced graded impairments in story recall, verbal fluency, working memory, procedural discourse, scene construction, and coreference resolution.
  arXiv  Detail & Related papers  (2026-08-04T03:28:37Z)
• Adapting Self-Supervised Speech Representations for Cross-lingual Dysarthria Detection in Parkinson's Disease [72.0406069194794]
  Speech representations often encode language-dependent structure that can confound dysarthria detection.<n>We propose a representation-level language shift that aligns source-language self-supervised speech representations with the target-language distribution.<n>We evaluate the approach on oral DDK recordings from Parkinson's disease speech datasets in Czech, German, and Spanish.
  arXiv  Detail & Related papers  (2026-03-23T17:23:39Z)
• Tokenization and Morphological Fidelity in Uralic NLP: A Cross-Lingual Evaluation [9.23725598061561]
  This study systematically compares three subword paradigms -- Byte Pair.<n>(BPE), Overlap BPE (OBPE), and Unigram Language Model -- across six Uralic languages.<n>We show OBPE consistently achieves stronger morphological alignment and higher tagging accuracy than conventional methods.
  arXiv  Detail & Related papers  (2026-02-04T05:59:25Z)
• Component-Level Lesioning of Language Models Reveals Clinically Aligned Aphasia Phenotypes [40.41503864764337]
  We introduce a component-level framework that simulates aphasia by selectively perturbing functional components in large language models.<n>Our pipeline identifies subtype-linked components for Broca's and Wernicke's aphasia, and induces graded impairments by progressively perturbing the top-k subtype-linked components.<n>Across architectures and lesioning strategies, subtype-targeted perturbations yield more systematic, aphasia-like regressions than size-matched random perturbations.
  arXiv  Detail & Related papers  (2026-01-27T15:47:22Z)
• Graph Modelling Analysis of Speech-Gesture Interaction for Aphasia Severity Estimation [0.0]
  Aphasia is an acquired language disorder caused by injury to the regions of the brain that are responsible for language.<n>Recent advancements in speech analysis focus on automated estimation of aphasia severity from spontaneous speech.<n>In this work, we propose a graph neural network-based framework for estimating aphasia severity.
  arXiv  Detail & Related papers  (2026-01-27T14:11:36Z)
• Far from the Shallow: Brain-Predictive Reasoning Embedding through Residual Disentanglement [43.96899536703126]
  Modern large language models (LLMs) are increasingly used to model neural responses to language.<n>Their internal representations are highly "entangled," mixing information about lexicon, syntax, meaning, and reasoning.<n>This entanglement biases conventional brain encoding analyses toward linguistically shallow features.
  arXiv  Detail & Related papers  (2025-10-26T22:46:26Z)
• Devising a Set of Compact and Explainable Spoken Language Feature for Screening Alzheimer's Disease [52.46922921214341]
  Alzheimer's disease (AD) has become one of the most significant health challenges in an aging society.<n>We devised an explainable and effective feature set that leverages the visual capabilities of a large language model (LLM) and the Term Frequency-Inverse Document Frequency (TF-IDF) model.<n>Our new features can be well explained and interpreted step by step which enhance the interpretability of automatic AD screening.
  arXiv  Detail & Related papers  (2024-11-28T05:23:22Z)
• A Methodology for Explainable Large Language Models with Integrated Gradients and Linguistic Analysis in Text Classification [2.556395214262035]
  Neurological disorders that affect speech production, such as Alzheimer's Disease (AD), significantly impact the lives of both patients and caregivers. Recent advancements in Large Language Model (LLM) architectures have developed many tools to identify representative features of neurological disorders through spontaneous speech. This paper presents an explainable LLM method, named SLIME, capable of identifying lexical components representative of AD.
  arXiv  Detail & Related papers  (2024-09-30T21:45:02Z)
• Investigating Critical Period Effects in Language Acquisition through Neural Language Models [70.6367059367609]
  Second language (L2) acquisition becomes harder after early childhood. ceasing exposure to a first language (L1) after this period (but not before) typically does not lead to substantial loss of L1 proficiency. It is unknown whether these CP effects result from innately determined brain maturation or as a stabilization of neural connections naturally induced by experience.
  arXiv  Detail & Related papers  (2024-07-27T19:17:10Z)
• Divergences between Language Models and Human Brains [59.100552839650774]
  We systematically explore the divergences between human and machine language processing.<n>We identify two domains that LMs do not capture well: social/emotional intelligence and physical commonsense.<n>Our results show that fine-tuning LMs on these domains can improve their alignment with human brain responses.
  arXiv  Detail & Related papers  (2023-11-15T19:02:40Z)
• Sensitivity, Performance, Robustness: Deconstructing the Effect of Sociodemographic Prompting [64.80538055623842]
  sociodemographic prompting is a technique that steers the output of prompt-based models towards answers that humans with specific sociodemographic profiles would give. We show that sociodemographic information affects model predictions and can be beneficial for improving zero-shot learning in subjective NLP tasks.
  arXiv  Detail & Related papers  (2023-09-13T15:42:06Z)
• Information-Restricted Neural Language Models Reveal Different Brain Regions' Sensitivity to Semantics, Syntax and Context [87.31930367845125]
  We trained a lexical language model, Glove, and a supra-lexical language model, GPT-2, on a text corpus. We then assessed to what extent these information-restricted models were able to predict the time-courses of fMRI signal of humans listening to naturalistic text. Our analyses show that, while most brain regions involved in language are sensitive to both syntactic and semantic variables, the relative magnitudes of these effects vary a lot across these regions.
  arXiv  Detail & Related papers  (2023-02-28T08:16:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.