LLM linguistic competence drives left-right brain activity prediction asymmetry

Researchers have identified a left-right asymmetry in how large language models (LLMs) predict human brain activity, which emerges as the models develop formal linguistic competence. This asymmetry, observed using fMRI data and models like OLMo-2-7B and Pythia, correlates with the LLMs' ability to distinguish acceptable from unacceptable sentences and generate well-formed text. The study found that this predictive accuracy difference between brain hemispheres does not align with performance on arithmetic, Dyck language tasks, or reasoning-based text tasks, suggesting a specific link to linguistic processing. AI

IMPACT Suggests a specific neural correlate for formal linguistic competence in LLMs, potentially guiding future model development and brain-computer interface research.

RANK_REASON Academic paper detailing research findings on LLM capabilities. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• English
• French
• functional magnetic resonance imaging
• Laurent Bonnasse-Gahot
• OLMo-2-7B
• Pythia
• Standard Chinese

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →