Do domain-general executive resources play a role in linguistic prediction? Re-evaluation of the evidence and a path forward.
Do domain-general executive resources play a role in linguistic prediction? Re-evaluation of the evidence and a path forward.
复制标题
领域通用执行资源在语言预测中发挥作用吗？
DOI：
10.1016/j.neuropsychologia.2019.107258
复制
发表时间：
2020
期刊：
影响因子：
2.6
通讯作者：
Fedorenko,Evelina
中科院分区：
文献类型：
--
作者：
Ryskin,Rachel;Levy,RogerP;Fedorenko,Evelina
大多数目前的语言理解帐户同意预测的作用，但他们不同意的领域一般执行资源在预测行为的重要性。在这篇评论文章中，我们简要回顾了语言预测的证据，以及用于论证预测利用领域一般执行资源的研究结果。最令人信服的证据是，在执行资源较低的人群（如儿童、老年人和第二语言（L2）学习者）中，语言理解过程中的预测行为明显减少。我们建议，这些人口之间的差异可以解释，而不调用执行资源。相反，这些人群所具有的语言经验的数量和类型的差异可能会影响参与预测行为的可能性，或者只是使预测效果在为年轻成年母语人士设计的范式中更难检测。因此，特定领域的预测机制仍然是一个可行的可能性。我们讨论了如何进一步测试语言预测的帐户，做与不需要域一般的执行资源，使用行为，计算和脑成像方法。
Most current accounts of language comprehension agree on a role for prediction, but they disagree on the importance of domain-general executive resources in predictive behavior. In this opinion piece, we briefly review the evidence for linguistic prediction, and the findings that have been used to argue that prediction draws on domain-general executive resources. The most compelling evidence is an apparent reduction in predictive behavior during language comprehension in populations with lower executive resources, such as children, older adults, and second language (L2) learners. We propose that these between-population differences can be explained without invoking executive resources. Instead, differences in the quantity and kind of language experience that these populations bring to bear may affect the probability of engaging in predictive behavior, or simply make prediction effects more difficult to detect in paradigms designed for young adult native speakers. Thus, domain-specific prediction mechanisms remain a viable possibility. We discuss ways to further test accounts of linguistic prediction that do vs. do not require domain-general executive resources, using behavioral, computational, and brain imaging approaches.