多文本閱讀是現代讀者學習新知的有效路徑之一（Britt & Rouet,2012）。探討兒童多文本閱讀認知歷程有助於開發證據本位的多文本閱讀教學模式，以提升學童多文本閱讀理解之表現（Beker et al.,2019）。本研究以眼動技術探究詞彙能力對四年級學生多文本閱讀理解歷程之影響。結果發現，隨著學生詞彙能力越佳，多文本閱讀後所形成的文本表徵的概念數，較單文本閱讀時來得少；但是會形成更多的跨文本整合概念數。此外，由單文本閱讀與閱讀文本組合中的第一篇文章眼動資料發現，詞彙能力愈佳的學生，閱讀文本組合中的第一篇文本時，詞彙處理時間（FFD與GD）較長，但重新回視的時間較短（RRT）。反之，詞彙能力較弱的學生，詞彙處理時間與重新回視的時間，都比單文本閱讀時來得短。進一步比較多文本閱讀的第一篇與第二篇文本眼動型態，則發現詞彙能力愈佳的學生，閱讀文本組合中的第二篇文本時，詞彙重新回視的時間比第一篇文本來得長，且發生重新回視的詞彙比率（Reread）也較高。反之，詞彙能力較弱的學生則在第二篇文本發生重新回視的詞彙比率較第一篇文本低。綜合上述研究發現，本研究進一步討論詞彙能力在學生多文本閱讀歷程的角色及其在教育應用上的方向。

Studies have demonstrated that young readers can integrate information across texts. However, the literature on the online processes of young readers during multiple-text reading is limited. The present study uses children's vocabulary ability as a covariable to investigate how they process information across texts and to determine how multiple-text reading differs from single-text reading in terms of eye-movement control, lexical access, subsequent integration, and macro-integration. For this study, we ensured that the texts used in the single-text and multiple-text reading scenarios were equivalent; for the multiple-text reading scenario, participants read a first text followed by a second text. Five sets of Traditional Chinese texts were used in this study, with each set comprising two expository texts centered on a common topic. The two texts in each set did not present conflicting information. Instead, the information in the second text expanded on the information in the first text. To create texts suitable for elementary school students, the text structures of all 10 experimental texts in addition to their word-level characteristics and article length were standardized. The results of this study indicate that the tested fourth-grade children could adjust their reading strategies spontaneously, depending on whether they were engaged in single-text reading or multiple-text reading. Regarding the role of vocabulary during multiple-text reading, we discovered that the fourth graders with higher vocabulary proficiency exhibited increased engagement in lexical processing and macro-integration when they were reading the first and second texts, respectively. By contrast, the fourth graders with lower vocabulary proficiency exhibited less engagement in lexical processing while reading the first text and were less likely to reread the second text. The fourth graders with higher vocabulary proficiency were more inclined to adjust their reading strategies depending on the text they were reading, whereas those with lower vocabulary proficiency were more inclined to adopt a minimalist strategy. We discussed these findings in relation to current models of multiple-text comprehension and explored their educational implications.