ABSTRACT
An important question in psycholinguistics is whether or not the semantic transparency of words plays any role during the processing of compound words. Studies of different languages have produced mixed results suggesting that semantic properties of compounding do not have a universal effect. This paper uses masked priming at three different SOAs to study the possible effects of semantic transparency and grammatical class on processing of Persian and Cantonese compounds. We found priming effects in all types of compounds across both languages regardless of semantic transparency and grammatical class. Our findings revealed processing patterns which were unique to each language. We contend that compounding is a complex process influenced by the interactions among the unique properties of each language such as productivity, headedness, and the writing system.
Acknowledgements
We would like to thank William Marslen-Wilson for hosting Shuk K. Cham at University of Cambridge and giving feedback on the design of the Cantonese study. We would like to thank Eva Wittenberg and Marco Marelli for giving us precious comments on the study and manuscript. We would like to thank Stephen Politzer-Ahles for his great comments on the analysis of the data.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
The norms, data for both Experiments and R codes are available online at the following DOI (10.17605/OSF.IO/68AGB). None of the Experiments was preregistered.
Notes
1 Cantonese follows a left to right writing system, while Persian is right to left. We use VNC for consistency purposes throughout the manuscript to denote order of constituents for both Cantonese and Persian, although the order in Persian is NV.