1. Introduction

This article introduces two expressions of restrictive relative clauses in English: non simplified relative clauses (AC) and simplified relative clauses (RAC), and points out that RAC may cause ambiguity in form, leading to the garden path effect. The article emphasizes that Chinese learners need to deal with the dual challenges of ambiguity resolution and sentence structure adjustment in the process of constructing Chinese translations. In addition, although there have been studies exploring the understanding process of garden path sentences, the lack of analysis combined with RAC output data has limited the explanatory power of the results. Therefore, this article compares and analyzes the understanding and target language output of Chinese learners towards AC and RAC through visual translation tasks under the self paced reading paradigm, aiming to deeply examine the relevant factors that affect the ambiguity resolution and target language construction of RAC, and reveal the semantic and syntactic synergy mechanism in the processing of second language relative clauses.

2. Literature review

This article summarizes two main viewpoints on the processing of garden path sentences: one believes that second language speakers find it difficult to construct complete sentence representations, such as the "good enough" model; Another perspective suggests that second language speakers are ultimately able to accurately understand sentences, including garden path models and constraint based models. There are two different views on the starting time of ambiguity resolution: parallel processing view and serial processing view. Research shows that Chinese learners rely on syntactic information when understanding garden path sentences, while low-level learners rely on lexical semantic information. In addition, Chinese learners adopt a progressive approach to sentence comprehension. In order to reveal the underlying factors that affect the processing of garden path sentences, it is necessary to combine the understanding process data with the translated language output results for cross validation.

3. Research Design

This paper elaborates on the research design of Chinese learners' processing patterns in the visual translation of simplified relative clauses in English. The research aims to answer three questions: the differences in Chinese learners' understanding of English AC and RAC, the differences in target language output, and the characteristics of RAC processing. 26 undergraduate students with intermediate English proficiency who are native Chinese speakers participated in the experiment. The experiment adopted a visual translation task under the self paced reading paradigm, and the materials were adapted based on the research of MacDonald et al. The appropriateness of vocabulary was ensured through word frequency analysis and familiarity survey. The experiment divides the target sentence into different regions of interest, with the independent variable being the sentence type and the dependent variables including the key region reading reaction time and target language quality. The experiment collected data in two stages through Latin square matching, using E-prime software, including exercises and formal experimental sessions. The evaluation of translation quality is based on three indicators: accuracy, completeness, and fluency, with silent pauses exceeding 1000 ms being specifically marked.

4. Data analysis

Detailed analysis was conducted on the experimental data, including reaction time during the reading stage and the quality of the translated language. Reading reaction time analysis showed that participants triggered the garden path effect when understanding RAC, and the average reaction time in the disambiguation zone of RAC was significantly longer than that of AC. Target language quality analysis involves three indicators: accuracy, completeness, and fluency, and RAC's target language quality is lower than AC in all three indicators. chi square test showed that there was a significant difference in accuracy and completeness between RAC and AC, but no significant difference in fluency. Further statistics show that the proportion of "average output" in AC translated language is significantly higher than that in RAC translated language. This indicates that the explicit syntactic markers before the AC ambiguity zone help participants identify relative clauses and produce passive structures in the target language, while the lack of this marker in RAC leads to greater challenges for participants in language translation.

5. Discussion

An in-depth analysis was conducted on the differences in processing patterns of simplified relative clauses (RAC) and non simplified relative clauses (AC) in visual translation by Chinese learners. Research has found that the processing difficulty of RAC is generally higher than that of AC, mainly due to the lack of syntactic markers that cause participants to experience garden path effects during the comprehension stage, increasing cognitive load. In the understanding stage, the ambiguity response time of RAC is shorter than that of AC, but the disambiguation response time is significantly longer, which is related to the principle of question element affiliation. Participants in RAC are more likely to encounter syntactic conflicts and need to reanalyze sentence structures.

6. Conclusion

Summarized the processing patterns of Chinese learners in the visual translation of simplified relative clauses in English. Research has found that learners rely on explicit syntactic markers and are susceptible to the garden path effect, leading to longer comprehension time and lower translation quality. In the ambiguity zone, learners adopt a parallel processing mode, but lack overall syntactic guidance and scattered information. The speed of understanding and the quality of visual translation are influenced by various factors, and the overall syntactic construction is crucial for the quality of the translated language. The language proficiency and interpreting experience of learners also have a significant impact on processing, which can be further explored in future research.