When humans and large language models (LLMs) process the same text, activations in the LLMs correlate with brain activity measured, e.g., with functional magnetic resonance imaging (fMRI). Moreover, it has been shown that, as the training of an LLM progresses, the performance in predicting brain activity from its internal activations improves more in the left hemisphere than in the right one. The aim of the present work is to understand which kind of competence acquired by the LLMs underlies the emergence of this left-right asymmetry. Using the OLMo-2 7B language model at various training checkpoints and fMRI data from English participants, we compare the evolution of the left-right asymmetry in brain scores alongside performance on several benchmarks. We observe that the asymmetry co-emerges with the formal linguistic abilities of the LLM. These abilities are demonstrated in two ways: by the model's capacity to assign a higher probability to an acceptable sentence than to a grammatically unacceptable one within a minimal contrasting pair, or its ability to produce well-formed text. On the opposite, the left-right asymmetry does not correlate with the performance on arithmetic or Dyck language tasks; nor with text-based tasks involving world knowledge and reasoning. We generalize these results to another family of LLMs (Pythia) and another language, namely French. Our observations indicate that the left-right asymmetry in brain predictivity matches the progress in formal linguistic competence (knowledge of linguistic patterns).

翻译：当人类与大型语言模型（LLMs）处理相同文本时，LLMs的激活状态与通过功能磁共振成像（fMRI）测量的大脑活动存在相关性。研究表明，随着LLM训练进程推进，从其内部激活状态预测大脑活动的性能在左半球的提升幅度显著高于右半球。本研究旨在探究LLMs获得的何种能力导致了这种左右不对称性的出现。通过使用不同训练阶段的OLMo-2 7B语言模型及英语参与者的fMRI数据，我们比较了大脑预测分数左右不对称性的演变与多项基准测试性能的关系。研究发现，这种不对称性与LLM形式语言能力的形成同步显现。该能力通过两种方式体现：模型在最小对比对中为可接受语句分配高于语法不可接受语句的概率，或生成格式规范文本的能力。相反，左右不对称性与算术运算、Dyck语言任务的表现无关，亦不涉及需要世界知识和推理的文本任务。我们将这些结论推广至另一LLM系列（Pythia）及法语场景。研究结果表明，大脑可预测性的左右不对称性与形式语言能力（语言模式知识）的进展相匹配。