A number of studies have shown that syllables play an important role in visual word recognition in Spanish. We report three lexical decision experiments with a masked priming technique that examined whether syllabic effects are phonological or orthographic in nature. In all cases, primes were nonwords. In Experiment 1, latencies to CV words were faster when primes and targets shared the first syllable ( ju.nas-JU.NIO) than when they shared the initial letters but not the first syllable ( jun.tu-JU.NIO). In Experiment 2, this syllabic overlap could be phonological þ orthographical (vi.rel-VI.RUS) or just phonological (bi.rel-VI.RUS). A syllable priming effect was found for CV words in both the phonological þ orthographical and the phonological condition. In Experiment 3 we compared a ‘‘phonologicalsyllable’’ condition (bi.rel-VI.RUS) with two control conditions (fi.rel-VI.RUS and vir.ga-VI.RUS). We found faster latencies for the phonological-syllabic condition than for the control conditions. These results suggest that syllabic effects are phonological in nature.

this article was supported by National Science Foundation Grant SBR-9616519. We are grateful to Max Coltheart, Laurie Feldman, Rumjahn Hoosain, Yuriko Kayamoto, and Sandy Pollatsek for their thoughtful comments. We thank Li Long for her assistance in recruiting participants in Beijing. Correspondence concerning this article should be addressed to Li Hal Tan or Charles A. Perfetti, Learning Research and Development Center, 3939 O'Hara Street, University of Pittsburgh, Pittsburgh, Pennsylvania 15260. Electronic mall may be sent to tanlh@hkucc.hku.hk. Recent research with Chinese single-character words, however, has suggested a stronger role for phonology in Chinese reading. In a primed perceptual-identification paradigm, Perfetti and Zhang (1991, Experiment 3) observed a synchrony of phonologic and semantic priming effects when a single-character prime was exposed for 50 ms, followed by a character target of 35 ms. Using a backward-masking procedure, Tan, Hoosain, and Peng (1995) exposed a target character for 60 ms, followed by a mask character that was presented for 40 ms. As in Perfetti and Zhang's experiment, they found no evidence for semantic effects in the absence of phonological effects. Equally interesting, when the target character had vague meaning, they found phonological effects in the absence of semantic effects. Such results suggest a very rapid activation of phonology and are not consistent with a meaning-first hypothesis (see Tan & Perfetti, 1998, for a detailed review of recent discoveries). The generality of these demonstrations of phonology is limited, however, because the research has used exclusively single-character words. Estimates of modem Chinese show that multiple-character words are actually more common than single-character words. According to...

f discussion for more than 20 years. Several sublexical structures have been proposed as functional units of encoding: morphemes (Gibson & Guinet, 1971; Murrell & Morton, 1974; Rapp, 1992; Taft & Forster, 1975), Basic Orthographic Syllabic Structure-BOSS- (e.g. Taft, 1979), word body (Kay & Bishop, 1987; Patterson & Morton, 1985) and, recently, the body of the BOSS (Forster & Taft, 1994; Taft, 1992). Among Requests for reprint should be addressed to Carlos J. lvarez, Departamento de Psicologa Cognitiva, Universidad de La Laguna, 38201 Tenerife, Spain. Electronic mail may be sent to calvarez@ull.es. these proposals, the syllable has been one of the sublexical units which has received a greater attention (e.g., Ferrand, Segu & Grainger, 1996; Lima & Pollatsek, 1983; Millis, 1986; Printzmetal, Treiman & Rho, 1986; Rapp, 1992; Rapp, Alway & Caramazza, 1993; Spoehr & Smith, 1973; Taft & Forster, 1976; Tousman & Inhoff, 1992). In addition, several syllable-relat

issages et de Dveloppement, Centre National de la Recherche Scientifique-URA D1838, Universit de Bourgogne, 6 boulevard Gabriele, F-21000 Dijon, France. The present research was supported by grants from the Fund for Joint Basic Research, from the Ministry of Scientific Policy, and ARC Contract 91/96-148 to the Laboratory of Experimental Psychology of the Free University of Brussels. I am grateful to Lucia Colombo, Jonathan Grainger and Daniel Holender for their constructive comments on earlier drafts of the manuscript. INTRODUCTION In alphabetic writing systems, words that are orthographically similar generally sound alike. Mere observation indicates that the reader takes advantage of this principle when reading aloud. For example, when new words or pseudowords are encountered, a phonological description can be generated thanks to the statistical regularity of the mapping between orthographic and phonological units. Most current models of reading capture the reader's ability to tran

Few linguistic concepts are at the same time so intuitively clear and so hopelessly elusive as that of the syllable. The density of research on this notion has grown relentlessly in the last two decades, but apparently this seems to have increased, rather than reduced, the fog surrounding it. Given this state of affairs, presuming to solve once and forever the syllable's puzzle would definitely appear to be too daring a goal. My purpose here is simply to attempt to assemble a few pieces of this growing puzzle, in the hope of putting some (provisional) order into one part of this complex picture. 1.1. Starting the game. The pieces with which my game starts concern the behaviour of Italian subjects in psycholinguistic experiments. As summarized in Bertinetto (to appear/a), in segment(s) substitution tasks Italian subjects do not show any strong syllabic effect. 1 This contrasts with the behaviour of English subjects, who consistently provided evidence of a robust structural hierarchy within the syllable. 2 Indeed, as repeatedly demonstrated by Rebecca Treiman and Bruce