Word reading and picture naming: Phonological encoding in English language production

In Roelofs' (2004) form preparation study examining processes involved in both word reading and picture naming, he concluded that phonological encoding mechanisms might be shared for the two tasks. Importantly, in his earlier form preparation research Roelofs (1999) argued that phonemic features are not involved during phonological encoding and indeed, most current models of general language production such as Word-form Encoding by Activation and VERification (WEAVER, e.g., Roelofs, 1997a) account for the role of phonemic features once the phonological encoding process has been completed. However, whilst Kinoshita's (2000) re-interpretation of the locus of the masked onset priming effect (MOPE) implies an encoding process for word reading that is similar to that incorporated into WEAVER (e.g., Roelofs, 1997a) and by extension to picture naming, Lukatela, Eaton and Turvey‟s (2001) results suggest that features may well be involved in the word reading processes. The main purpose of the research undertaken within this thesis was to evaluate phonological encoding for both word reading and picture naming to assess the validity of Roelofs' (2004) claims. This was conducted with the employment of the masked priming paradigm as well as the masked sandwich priming paradigm and by the manipulation of phonemic feature overlap in both the initial and end/coda segment position of primes and monosyllabic targets. From the cumulative results of this research, the notion that encoding mechanisms might be shared between these two tasks could not be ruled out. Importantly, phonemic feature effects were consistently observed across both word reading (with lexical primes) and picture naming. Controversially, these particular findings suggest that conventional thinking is misguided to ignore the role of phonemic features during the phonological encoding process.