Generative causal testing to bridge data-driven models and scientific theories in language neuroscience
arXiv:2410.00812v3 Announce Type: replace Abstract: Representations from large language models are highly effective at predicting BOLD fMRI responses to language stimuli. However, these representations are largely opaque: it is unclear what features of the language stimulus drive the response in each brain area. We present generative causal testing (GCT), a framework for generating concise explanations of language selectivity in the brain from predictive models and then testing those explanations in follow-up experiments using LLM-generated stimuli.This approach is successful at explaining sel
The increasing sophistication of large language models (LLMs) and neuroimaging techniques allows for a deeper, more interpretable understanding of language processing in the brain.
This research provides a framework to bridge opaque AI models and neuroscience, potentially unlocking new insights into brain function and enabling more interpretable AI systems.
The ability to generate clear, testable explanations for how the brain processes language using LLMs moves beyond simple prediction to causal understanding and hypothesis generation.
- · Neuroscience researchers
- · AI interpretability researchers
- · Cognitive science
- · LLM developers
- · Black-box AI models in neuroscience
Improved understanding and causal models of human language processing.
Development of more biologically plausible and interpretable AI models.
Potential for new therapeutic interventions for language-related neurological disorders based on causal insights.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL