Exploring the scope and limits of incremental learning in word production - Evidence from picture-word interference and blocked-cyclic naming experiments

近年来，词汇加工的实验研究和计算建模一直是语音产生研究的热点。虽然有令人信服的证据表明，话语产生的语境对词汇选择（即说话者实际使用的词汇选择）有很强的影响，但语境是否也会影响前面的词汇激活过程，我们知之甚少。本研究方案将集中于这个问题。基于参考交际研究的发现，它将重点关注目标话语之前的语言语境的影响，这些语境是由说话者自己或对话伙伴产生的。我们将特别探讨，这些语言语境所施加的限制是否会阻止词汇替代的加工，从而导致局部不恰当的话语。该研究项目将利用在指称交际研究的背景下发展起来的范式，从而弥合心理语言学中两种（在大多数情况下）独立研究传统之间的鸿沟：一方面探索指导指称交际的原则，另一方面对概念-词汇过程进行时序分析。我们的发现将有助于更深入地理解和更全面地建模在语境中说话的认知过程及其灵活性。

In the past years, the experimental investigation and computational modelling of lexical processing was in the focus of speech production research. Although there is there is compelling evidence that the context, in which an utterance is produced, has a strong influence on lexical selection, i.e. the selection of words actually used by a speaker, little is known about whether the context also affects the preceding lexical activation processes. The present research programme will focus on this question. Based on findings from referential communication studies, it will focus on effects from linguistic contexts which precede a target utterance and which are produced by either the speaker herself/himself or a dialogue partner. In particular we will explore, whether constraints imposed by these linguistic contexts can prevent the processing of lexical alternatives which would result in locally inappropriate utterances. The research program will make use of paradigms developed in the context of referential communication studies and will thus bridge the cleft between two - for most parts - independent research traditions in psycholinguistics: the exploration of principles guiding referential communication on the one hand and the chronometric analysis of conceptual-lexical processes on the other. Our finding shall contribute to a deeper understanding and more comprehensive modelling of the cognitive processes underlying speaking in context and their flexibility.