Examining the effects of character and word predictability in Chinese reading

The present study applied Luke and Christianson's (2016) analytical approach to examine predictability effects in Chinese sentence reading. Because Chinese words consist of a variable number of meaning-bearing characters but lack explicit word boundaries, our study examined both character- and word-based cloze task metrics, and their relative capacity to explain the facilitative effects of predictability on eye movements during reading. Our results indicate that both types of metrics can be equally effective in accounting for eye-movement behavior and further suggest that both graded pre-activation of semantic and syntactic features and prediction of specific lexical forms operate in tandem during Chinese reading. These findings highlight the importance of understanding how language-specific factors influence predictive processing during reading.