Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2410.00812 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915178001465344 |
|---|---|
| author | Antonello, Richard Singh, Chandan Jain, Shailee Hsu, Aliyah Guo, Sihang Gao, Jianfeng Yu, Bin Huth, Alexander |
| author_facet | Antonello, Richard Singh, Chandan Jain, Shailee Hsu, Aliyah Guo, Sihang Gao, Jianfeng Yu, Bin Huth, Alexander |
| contents | 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 selectivity both in individual voxels and cortical regions of interest (ROIs), including newly identified microROIs in prefrontal cortex. We show that explanatory accuracy is closely related to the predictive power and stability of the underlying predictive models. Finally, we show that GCT can dissect fine-grained differences between brain areas with similar functional selectivity. These results demonstrate that LLMs can be used to bridge the widening gap between data-driven models and formal scientific theories. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_00812 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Generative causal testing to bridge data-driven models and scientific theories in language neuroscience Antonello, Richard Singh, Chandan Jain, Shailee Hsu, Aliyah Guo, Sihang Gao, Jianfeng Yu, Bin Huth, Alexander Computation and Language Neurons and Cognition 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 selectivity both in individual voxels and cortical regions of interest (ROIs), including newly identified microROIs in prefrontal cortex. We show that explanatory accuracy is closely related to the predictive power and stability of the underlying predictive models. Finally, we show that GCT can dissect fine-grained differences between brain areas with similar functional selectivity. These results demonstrate that LLMs can be used to bridge the widening gap between data-driven models and formal scientific theories. |
| title | Generative causal testing to bridge data-driven models and scientific theories in language neuroscience |
| topic | Computation and Language Neurons and Cognition |
| url | https://arxiv.org/abs/2410.00812 |