Universality in eye movements and reading: A replication with increased power.

Area of Science:

• Cognitive Psychology
• Psycholinguistics
• Reading Science

Background:

• Previous eye movement research on Chinese, Finnish, and English reading showed language-specific differences in fine-grained eye movement characteristics but no overall difference in sentence reading times.
• The original study was criticized for statistical underpowering and methodological limitations, including a small sample size and suboptimal statistical analyses.
• These limitations prompted a re-evaluation using a larger sample and advanced statistical modeling to address concerns about universality in reading processing.

Purpose of the Study:

• To re-investigate reading processing across Chinese, Finnish, and English using a larger sample size and robust statistical methods.
• To address criticisms regarding statistical power and analytical approaches in the original 2016 study.
• To examine the influence of visual complexity and potential cultural changes on reading speed across different orthographic systems.

Main Methods:

• Recruited at least 80 new participants for each language (Chinese, Finnish, English).
• Employed formal statistical modeling to analyze eye movement data across all three languages simultaneously.
• Incorporated an index of visual complexity to account for language-specific visual processing demands.

Main Results:

• Contrary to the original findings, new analyses indicated significantly shorter total sentence reading times for Chinese readers compared to Finnish and English readers.
• Fine-grained eye movement characteristics showed consistent differences across languages, aligning with the original study's observations.
• The study found evidence supporting universality in certain aspects of reading processing despite observed differences in reading times.

Conclusions:

• The observed faster reading times in Chinese may be attributed to cultural and societal changes over the past decade.
• The findings support the hypothesis of universal mechanisms underlying reading processing, even across languages with distinct orthographic features.
• The re-analysis provides a more statistically robust foundation for understanding cross-linguistic reading, addressing prior concerns about sample size and analytical rigor.