Chinese Reading | MCUs in Reading

Reading in Chinese

Over the last two decades, we have developed a strong collaborative international relationship with a group of colleagues working in Tianjin Normal University and through our joint efforts, we have carried out a significant amount of research to investigate the nature of Chinese reading.

Unlike English or other alphabetic languages, the Chinese writing system is character based. Chinese characters are closely packed box like symbols that are presented horizontally adjacent to form sentences. Characters are comprised of strokes with some features like dots, lines and curves. Different characters may differ in the number of strokes, and when they are written, their constituent strokes are written in a fixed order. Here are two examples of writing Chinese characters：

Words in Chinese are comprised of one or more characters. However, Chinese is an unspaced language - word units are not demarcated by spaces at their beginning and end. For this reason, Chinese readers must engage in a cognitive process termed "word segmentation", that is, they must make decisions about where word boundaries lie as they read a sentence. This process is very important because it is necessary for the identification of each of the words of a sentence, something that itself is a critical aspect of natural reading. The situation is further complicated by the fact that there is often ambiguity as to the location of word boundaries in sentences, and it is also quite common for Chinese readers to disagree about where the word boundaries in a sentence actually lie. In fact, the Chinese did not have a term for “word” until the concept was imported from the West at the beginning of the 20th century (Packard, 1998). It is, perhaps, not so surprising, therefore, that Chinese readers are sometimes unclear about exactly what a word is, and they are often unclear as to how to distinguish a word from other linguistic units like phrases (Hoosain, 1992; Li, Zang, Liversedge, & Pollatsek, 2015; Liu, Li, Lin, & Li, 2013; Zang, Liversedge, Bai, & Yan, 2011).

Here is a simple Chinese sentence:

武汉市长江大桥欢迎您。

It can be segmented in different ways and the different segmentations lead to different interpretations, for example:

武汉市|长江大桥|欢迎|您。

Wuhan city|Yangtze river bridge|welcome|you (literal translation)

武汉|市长|江大桥|欢迎|您。

Wuhan city|major|Jiangdaqiao|welcome|you (literal translation)

These important characteristics of written Chinese allow us to investigate theoretical questions that it is simply impossible to address in English. Indeed, how Chinese readers segment text into units for word identification as they read normally is a puzzle. The fact that written Chinese is dense, character based, unspaced and ambiguous in relation to word boundaries means that it is an ideal written language in which to explore whether readers identify Multi-Constituent Units (MCUs), as well as words, as they read. This is why we chose to conduct experiments to investigate reading in Chinese.

Professor Simon Liversedge and Dr Chuanli Zang at the University of Central Lancashire, together with their colleagues Professor Xuejun Bai and Professor Guoli Yan at Tianjin Normal University, undertake eye movement experiments to analyse the way people read Chinese text and how they segment sentences into words and other multi-constituent units (MCUs). This work is conducted as part of an ESRC funded project that is held at the University of Central Lancashire.

In this project, our objective is to understand how real time word identification and eye movement control are operationalised over stimuli with indefinite lexical status during natural reading. Specifically, we aim to investigate two important theoretical questions (1) whether there is evidence to support our MCU Hypothesis (a theoretical account that we have formally proposed in recently published work, see Zang, 2019). And here we must acknowledge that our ideas derive from important earlier work (e.g., see Bybee, 1985; Pinker, 1998). Relatedly, we wish to know (2) how word (or MCU) segmentation occurs on-line during natural Chinese reading?

If you would like to know more about the project please feel free to email Professor Liversedge (SPLiversedge@uclan.ac.uk) or Dr Zang (CZang@uclan.ac.uk) and we would be happy to provide more details. You may also find some of the references below to be interesting.

Zang, C. (2019). New perspectives on serialism and parallelism in oculomotor control during reading: The Multi-Constituent Unit Hypothesis. Vision, 3, 50, 1-13.
Li, X., Zang, C., Liversedge, S.P., & Pollatsek, A. (2015). The role of words in Chinese reading. In Pollatsek, A., & Treiman, R. (Eds). The Oxford Handbook of Reading (pp. 232-244). New York, NY: Oxford University Press.
Zang, C., Liversedge, S.P., Bai, X., & Yan, G. (2011). Eye movements during Chinese reading. In S. P. Liversedge, I. D. Gilchrist, & S. Everling. (Eds). The Oxford Handbook of Eye Movements (pp. 961–978). New York, NY: Oxford University Press.

For a fuller list of the Chinese papers we have published, please see our publication section.