arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2109.01226cs.CLcs.AIcs.ITcs.LGmath.IT
收藏

完形填空到目前为止:分布信息比人类可预测性判断更好地预测N400幅度

So Cloze yet so Far: N400 Amplitude is Better Predicted by Distributional Information than Human Predictability Judgements

University of California San Diego(加州大学圣迭戈分校)

James A. Michaelov, Seana Coulson, Benjamin K. Bergen

摘要:

更容易预测的单词更容易处理--它们读得更快,引发的与处理困难相关的神经信号更小,最明显的是与事件相关的脑电位的N400成分。因此,有人认为,预测即将到来的单词是语言理解的一个关键组成部分,研究N400的幅度是研究我们所做预测的一种有价值的方法。在这项研究中,我们调查是否计算语言模型或人类的语言预测更好地反映了自然语言刺激调制的N400的幅度的方式。人类语言预测与计算语言模型的一个重要区别是,虽然语言模型的预测完全基于前面的语言上下文,但人类可能依赖于其他因素。我们发现,三个顶级当代语言模型- GPT-3、RoBERTa和ALBERT --的预测比人类预测更接近N400。这表明N400的预测过程可能比以前认为的对语言的表面统计更敏感。

英文摘要:

More predictable words are easier to process - they are read faster and elicit smaller neural signals associated with processing difficulty, most notably, the N400 component of the event-related brain potential. Thus, it has been argued that prediction of upcoming words is a key component of language comprehension, and that studying the amplitude of the N400 is a valuable way to investigate the predictions we make. In this study, we investigate whether the linguistic predictions of computational language models or humans better reflect the way in which natural language stimuli modulate the amplitude of the N400. One important difference in the linguistic predictions of humans versus computational language models is that while language models base their predictions exclusively on the preceding linguistic context, humans may rely on other factors. We find that the predictions of three top-of-the-line contemporary language models - GPT-3, RoBERTa, and ALBERT - match the N400 more closely than human predictions. This suggests that the predictive processes underlying the N400 may be more sensitive to the surface-level statistics of language than previously thought.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑