Parafoveal processing of Arabic diacritical marks

Area of Science:

• Psycholinguistics
• Cognitive Science
• Computational Linguistics

Background:

• Diacritics in Arabic convey vowel information, aiding pronunciation and disambiguating homographs.
• Skilled readers often omit diacritics, leading to common heterophonic homographs with dominant and subordinate meanings.
• Understanding parafoveal processing of diacritics is crucial for Arabic reading models.

Purpose of the Study:

• To investigate the parafoveal processing of Arabic diacritics during reading.
• To examine how diacritization accuracy and expected pronunciation patterns influence reading behavior.
• To determine the impact of diacritic presence/absence on parafoveal and foveal processing.

Main Methods:

• Utilized the boundary paradigm with native Arabic readers.
• Presented heterophonic homographs in sentences with varying diacritization previews (identical, inaccurate, none).
• Measured fixation durations as an indicator of processing difficulty.

Main Results:

• Inaccurate diacritic previews led to significantly longer fixation durations, especially at close launch sites.
• Parafoveal processing of diacritic identity was confirmed.
• Readers' expectations for dominant/subordinate pronunciations influenced processing based on perceived diacritic presence or absence.

Conclusions:

• Arabic readers actively process diacritics in the parafovea.
• Reading expectations, shaped by diacritic cues, modulate parafoveal and foveal processing.
• Absence of diacritics primes dominant pronunciations, while presence primes subordinate ones.