ABSTRACT
Is written word production affected by phonological properties of target words? We report three experiments using masked priming to investigate this issue. Chinese was chosen as the target script because sound and spelling can be largely dissociated. Participants wrote down names of objects, and latencies were measured on a graphic tablet. Objects were preceded by masked prime words which were either phonologically and orthographically related (PO) to the picture name, phonologically related but orthographically unrelated (P), or unrelated. Priming effects were found for both types of related primes with prime exposure durations of 58 ms (Experiment 1) and 33 ms (Experiment 2), with PO priming larger than P priming. Priming disappeared in Experiment 3 when a manual semantic judgment was required instead of written naming, suggesting that facilitation in the earlier experiments originated at the orthographic output level. These findings strengthen the existing evidence for the involvement of phonology in written word production.
Acknowledgement
We acknowledge Wenyu Li for help in data collection.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. Note, however, that Tainturier and Rapp (Citation2001) made the prediction that for the writing of multiple words, the role of phonology might be more pronounced than in single word written production. Due to the slow execution speed of handwritten output, buffering and rehearsal of constituents in phonological short-term memory is required for production of longer written utterances. Indeed, patients with preserved writing skills but difficulties in accessing phonology tend to produce “agrammatic” written utterances (e.g. Bub & Kertesz, Citation1982), which is in line with the idea that phonological buffering plays an important role in multi-word written generation.