Coreference processing of Control constructions and their pronoun-containing counterparts can be studied experimentally using priming or interference paradigms. We replicate findings in a priming study on non-finite Control constructions in Norwegian (Larsen & Johansson, 2020) and contrast them with their finite counterparts using interference effects in a grammatical maze (G-maze) design. We asked participants to read sentences word-by-word and to select the grammatically correct continuation from two options. When the ungrammatical option was a potential antecedent from within the sentence, we predicted interference, i.e., longer reaction times compared to an unrelated baseline. We observed a trend towards significant interference effects when a participant was presented with either of the potential noun phrase (NP) antecedents of PRO in competition with the infinitive marker (test position zero) during the processing of a Control sentence. This indicates reactivation of potential antecedents at the infinitive marker, and a reactivation position (PRO) near or at the infinitive marker. We also observed significant differences between Control constructions and their pronoun counterparts. A significant interference effect was recorded for Subject Pronoun constructions when either potential NP antecedent of the pronoun was presented in competition with the pronoun itself. A similar trend was recorded for Object Pronoun sentences.

可以使用启动或干扰范式通过实验研究控制结构及其包含代词的对应物的共指处理。我们在挪威语的非有限控制结构（Larsen & Johansson，2020）中复制了一项启动研究的结果，并使用语法迷宫（G-maze）设计中的干扰效应将它们与有限对应物进行对比。我们要求参与者逐字阅读句子，并从两个选项中选择语法正确的延续。当不合语法的选项是句子中的潜在先行词时，我们预测干扰，即与不相关的基线相比，反应时间更长。我们观察到当参与者在处理控制句期间被呈现与不定式标记（测试位置零）竞争的 PRO 的任一潜在名词短语 (NP) 前身时，出现显着干扰效应的趋势。这表明在不定式标记处潜在先行词的重新激活，以及在不定式标记附近或处的重新激活位置 (PRO)。我们还观察到控制结构与其代词对应物之间的显着差异。当代词的潜在 NP 先行词与代词本身竞争时，对主语代词结构记录了显着的干扰效应。宾语代词句子也有类似的趋势。