Nayoung Kwon

The processing of raising and nominal control: an eye-tracking study

According to some views of sentence processing, the memory retrieval processes involved in dependency formation may differ as a function of the type of dependency involved. For example, using closely matched materials in a single experiment, Dillon et al. (2013) found evidence for retrieval interference in subject-verb agreement, but not in reflexive-antecedent agreement. We report four eye-tracking experiments that examine examine reflexive-antecedent dependencies, combined with raising (e.g., “John seemed to Tom to be kind to himself…”), or nominal control (e.g., “Johns agreement with Tom to be kind to himself…”). We hypothesized that dependencies involving raising would (a) be processed more quickly, and (b) be less subject to retrieval interference, relative to those involving nominal control. This is due to the fact that the interpretation of raising is structurally constrained, while the interpretation of nominal control depends crucially on lexical properties of the control nominal. The results showed evidence of interference when the reflexive-antecedent dependency was mediated by raising or nominal control, but very little evidence that could be interpreted in terms of interference for direct reflexive-antecedent dependencies that did not involve raising or control. However, there was no evidence either for greater interference, or for quicker dependency formation, for raising than for nominal control.

Predicting semantic features in Chinese: Evidence from ERPs

This article reports two ERP studies that exploited the classifier system of Mandarin Chinese to investigate semantic prediction. In Mandarin, in certain contexts, a noun has to be preceded by a classifier, which has to match the noun in semantically-defined features. In both experiments, an N400 effect was elicited in response to a classifier that mismatched an up-coming predictable noun, relative to a matching classifier. Among the mismatching classifiers, the N400 effect was graded, being smaller for classifiers that were semantically related to the predicted word, relative to classifiers that were semantically unrelated to the predicted word. Given that the classifier occurred before the predicted word, this result shows that fine-grained semantic features of nouns can be pre-activated in advance of bottom-up input. The studies thus extend previous findings based on a more restricted range of highly grammaticalized features such as gender or animacy in Indo-European languages (Szewczyk & Schriefers, 2013; Van Berkum, Brown, Zwitserlood, Kooijman, & Hagoort, 2005; Wicha, Bates, Moreno, & Kutas, 2003).

Processing control information in a nominal control construction: An eye-tracking study

In an eye-tracking experiment, we examined the processing of the nominal control construction. Participants’ eye-movements were monitored while they read sentences that included either giver control nominals (e.g. promise in Luke’s promise to Sophia to photograph himself) or recipient control nominals (e.g. plea in Luke’s plea to Sophia to photograph herself). In order to examine both the initial access of control information, and its later use in on-line processing, we combined a manipulation of nominal control with a gender match/mismatch paradigm. Results showed that there was evidence of processing difficulty for giver control sentences (relative to recipient control sentences) at the point where the control dependency was initially created, suggesting that control information was accessed during the early parsing stages. This effect is attributed to a recency preference in the formation of control dependencies; the parser prefers to assign a recent antecedent to PRO. In addition, readers slowed down after reading a reflexive pronoun that mismatched with the gender of the antecedent indicated by the control nominal (e.g. Luke’s promise to Sophia to photograph herself). The mismatch cost suggests that control information of the nominal control construction was used to constrain dependency formation involving a controller, PRO and a reflexive, confirming the use of control information in on-line interpretation.

Attraction Effects in Honorific Agreement in Korean

Previous studies have suggested that sentence processing is mediated by content-addressable direct retrieval processes (McElree, 2000; McElree et al., 2003). However, the memory retrieval processes may differ as a function of the type of dependency. For example, while many studies have reported facilitatory intrusion effects associated with a structurally illicit antecedent during the processing of subject-verb number or person agreement and negative polarity items (Pearlmutter et al., 1999; Xiang et al., 2009; Dillon et al., 2013), studies investigating reflexives have not found consistent evidence of intrusion effects (Parker et al., 2015; Sturt and Kwon, 2015; cf. Nicol and Swinney, 1989; Sturt, 2003). Similarly, the memory retrieval processes could be also sensitive to cross-linguistic differences (cf. Lago et al., 2015). We report one self-paced reading experiment and one eye-tracking experiment that examine the processing of subject-verb honorific agreement, a dependency that is different from those that have been studied to date, in Korean, a typologically different language from those previously studied. The overall results suggest that the retrieval processes underlying the processing of subject-verb honorific agreement in Korean are susceptible to facilitatory intrusion effects from a structurally illicit but feature-matching subject, with a pattern that is similar to subject-verb agreement in English. In addition, the attraction effect was not limited to the ungrammatical sentences but was also found in grammatical sentences. The clear attraction effect in the grammatical sentences suggest that the attraction effect does not solely arise as the result of an error-driven process (cf. Wagers et al., 2009), but is likely also to result from general mechanisms of retrieval processes of activating of potential items in memory (Vasishth et al., 2008).

Processing Information During Regressions: An Application of the Reverse Boundary-Change Paradigm

Although 10–15% of eye-movements during reading are regressions, we still know little about the information that is processed during regressive episodes. Here, we report an eye-movement study that uses what we call the reverse boundary change technique to examine the processing of lexical-semantic information during regressions, and to establish the role of this information during recovery from processing difficulty. In the critical condition of the experiment, an initially implausible sentence (e.g., There was an old house that John had ridden when he was a boy) was rendered plausible by changing a context word (house) to a lexical neighbor (horse) using a gaze-contingent display change, at the point where the readers gaze crossed an invisible boundary further on in the sentence. Due to the initial implausibility of the sentence, readers often launched regressions from the later part of the sentence. However, despite this initial processing difficulty, reading was facilitated, relative to a condition where the display change did not occur (i.e., the word house remained on screen throughout the trial). This result implies that the relevant lexical semantic information was processed during the regression, and was used to aid recovery from the initial processing difficulty.

Effects of Inference Demands and Structural Complexity on Transitive Inference Processing: Comment on Lee and Kwon (2013)

In a recent study that investigated the processing of transitive inferences, Lee and Kwon (2013) concluded that only informational complexity leads to processing difficulty, that structural complexity affects earlier stages of processing than informational complexity, and that working memory resources are shared between sentence processing and other cognitive processes. Here, the evidence for these findings is scrutinized, and it is argued that there exist significant difficulties in drawing such conclusions based on the methodology and presented data.