The Influence of Contextual Predictability on Word Segmentation in Chinese Reading: An Eye-Tracking Study

Area of Science:

• Cognitive Psychology
• Linguistics
• Neuroscience

Background:

• Word segmentation is crucial for reading, especially in languages like Chinese without explicit word spacing.
• Previous research indicates contextual predictability aids lexical identification but its effect on word segmentation is unknown.

Purpose of the Study:

• To investigate how contextual predictability influences word segmentation preferences in Chinese reading.
• To examine the interplay between segmentation type and predictability using eye-tracking.

Main Methods:

• An eye-tracking study with 76 native Chinese speakers.
• Used ambiguous three-character strings as regions of interest.
• Employed a 2x2 within-subjects design varying segmentation type (AB-C vs. A-BC) and contextual predictability (high vs. low).

Main Results:

• Highly predictable words were skipped more often, but predictability did not affect other eye-movement measures.
• A consistent preference for AB-C segmentation was observed, with shorter reading times compared to A-BC.
• No interaction was found between contextual predictability and segmentation type.

Conclusions:

• Chinese reading demonstrates a robust preference for AB-C segmentation.
• Contextual predictability and word segmentation appear to be independent processes.
• Findings support the Chinese Reading Model (CRM), suggesting predictability has minimal impact on segmentation during reading.