Processing of Orthographic Representations in Transparent Orthography: Evidence from Masked Priming

The present study investigated visual word recognition in a transparent orthography by examining the effects of lexicality, word frequency, word length, prime duration, and task conditions in a masked priming paradigm. Using Bosnian as a model of transparent orthography, 100 adults completed inclusion and exclusion stem-completion tasks with prime durations of 0 ms, 50 ms, and 250 ms. A robust lexicality effect was observed, with real words producing higher accuracy than pseudowords and nonwords. Longer stimuli yielded lower accuracy and slower responses, whereas frequency effects emerged primarily in accuracy and varied across prime durations. High-frequency words showed an advantage at the 50-ms SOA that was reduced at the 250-ms SOA. Inclusion and exclusion conditions produced different patterns of accuracy and reaction times, highlighting the contribution of both automatic and controlled processes to visual word recognition. Furthermore, lexicality effects, and frequency effects on accuracy, varied as a function of the time available for prime processing. The findings support models proposing parallel contributions of lexical and sublexical processing mechanisms in reading. Consistent with dual-route models and the psycholinguistic grain size theory, familiar lexical items benefited from lexical activation, whereas longer and less familiar stimuli placed greater demands on sublexical decoding processes. More broadly, the results suggest that reading in transparent orthographies reflects dynamic interactions between lexical activation, orthographic structure, and processing time rather than relying solely on grapheme-phoneme decoding. These findings contribute to theoretical accounts of word recognition and have implications for literacy instruction and reading development in transparent writing systems.