AI-Driven Metalinguistic Engagement and EFL Reading Performance: The Roles of Morphological, Phonological, Syntactic, and Orthographic Awareness
This study demonstrates that for Chinese EFL undergraduates, AI-driven metalinguistic engagement and four key linguistic awareness domains (morphological, phonological, syntactic, and orthographic) significantly enhance reading performance through the partial mediation of word recognition efficiency, with these effects further moderated by baseline L2 proficiency.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Reading a second language is a complex mental feat, especially for university students in China who must navigate academic texts written in English. To understand a sentence, the brain does not simply look at letters and guess the meaning; it relies on a set of hidden skills that work together to decode the text. These skills include recognizing how words are built from smaller parts, hearing the sound patterns within those words, understanding how sentences are structured, and spotting the correct spelling patterns. When these skills are sharp, the brain can recognize words quickly and accurately, freeing up mental energy to grasp the deeper meaning of the text. In recent years, artificial intelligence has entered the classroom as a tool that can offer immediate feedback on these specific skills, but until now, it was unclear exactly how this technology interacts with the brain's reading machinery or whether it helps students of all skill levels equally.
A team of researchers from Guizhou Education University set out to map these connections in a large-scale study involving 488 undergraduate students from four different cities in China: Shenzhen, Wuhan, Xi'an, and Hangzhou. These students represented a wide range of backgrounds, from engineering to the humanities, and varied in their years of study and initial English proficiency. The researchers wanted to see if four specific types of language awareness—understanding word parts, sound patterns, sentence structure, and spelling rules—could predict how well a student reads. They also wanted to test if a student's engagement with AI tools that provided specific feedback on these four areas could boost their reading performance. Crucially, they hypothesized that the speed and accuracy with which a student recognizes words acts as the bridge between these skills and the final ability to understand a text.
To find the answers, the researchers gathered data through a mix of timed tasks and questionnaires. The students completed exercises that tested their ability to break down complex words, identify sound patterns, correct sentence errors, and judge spelling validity. They also took a quick test where they had to decide as fast as possible if a string of letters was a real English word, a task designed to measure how efficiently their brains could access word meanings. Finally, the students reported on how often and how deeply they used AI tools to get help with these specific language features while reading. The researchers then analyzed this data to see how the different pieces fit together, looking for patterns that explained why some students improved their reading more than others.
The study found that all four language skills contributed to better reading, but they did so in distinct ways. Students who were good at breaking down word parts and understanding sentence structures showed the strongest gains in reading comprehension. However, the ability to hear sounds and recognize correct spelling patterns remained important even for adult learners, proving that these foundational skills never fully disappear. The research also confirmed that the speed and accuracy of word recognition is the key mechanism at work. When students had strong language skills, they recognized words faster and more accurately, and this efficiency was what directly led to better understanding of the text. It is as if the brain's ability to instantly identify words clears the path for the rest of the reading process to happen smoothly.
Perhaps the most significant finding was the role of artificial intelligence. The study showed that when students used AI tools to get specific feedback on word parts, sounds, sentence structures, or spelling, their reading performance improved. This was not just because the AI made learning more interesting; the data suggested that these targeted prompts helped students notice and correct their mistakes in real time, which strengthened their mental representations of words and sentences. This feedback loop accelerated the process of word recognition, allowing students to read more fluently. The researchers noted that this effect was particularly strong for students who already had a solid foundation in English. For these learners, the AI feedback acted as a powerful amplifier, helping them refine their skills and reach higher levels of comprehension.
The study also revealed that a student's starting level of English mattered. Those who began with a stronger grasp of the language were better able to take advantage of both their natural language skills and the AI feedback. This suggests that while AI tools are beneficial for everyone, they work best when they are paired with a learner who has enough basic knowledge to process the feedback effectively. The researchers did not find that the AI replaced the need for foundational skills; rather, it worked best when those skills were already being developed. The results indicate that the most effective approach for universities is a curriculum that systematically builds all four areas of language awareness while integrating AI tools that offer targeted, specific prompts during reading tasks.
This research provides a clear picture of how modern technology can support traditional language learning. It moves beyond the idea that AI is just a general helper and shows that it is most effective when it is used to sharpen specific mental skills related to reading. By focusing on the efficiency of word recognition, educators can better understand how to structure their lessons and select the right digital tools. The study suggests that for Chinese university students, the path to better reading comprehension lies in a combination of strong foundational training in word structure, sound, grammar, and spelling, supported by AI tools that provide immediate, specific guidance. This approach ensures that students do not just read more, but read with greater speed, accuracy, and understanding.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.