EXPLORING VISUAL WORD RECOGNITION IN LATE SECOND LANGUAGE LEARNERS: A FOCUS ON LETTER TRANSPOSITION

Abstract

Previous studies in the first language (L1) revealed that letter-transposed (TL) primes (e.g., baisc for basic) have a significant advantage over letter-substituted (SB) primes, although this advantage accompanies a significant cost when compared to identity (ID) primes. These phenomena are often cited as evidence for position-flexible yet precise lexical processing in L1. The present study investigated whether Korean L2 learners of English employ the same processing abilities as native speakers. A lexical decision task using the masked priming paradigm was conducted with 144 English words and 144 nonwords. Targets were preceded by four prime types: ID, TL, SB, and unrelated (UN) primes. Both native speakers and L2 learners exhibited significant TL advantage and cost. However, the two groups differed significantly in the repetition priming effect as well as in the differences between TL and UN conditions, and between the SB and UN conditions. These findings in the L2 learners were not entirely compatible with either the Lexical Tuning Hypothesis (Castles, Davis, Cavalot, & Forster, 2007) or the Multiple Route Model (Coltheart, Rastle, Perry, Langdon, & Ziegler, 2001; Grainger, Lété, Bertand, Dufau, & Ziegler, 2012; Grainger & Ziegler, 2011). Instead, the L2 learners’ performance was primarily influenced by the formal similarity in terms of letter identity and letter position. Thus, high formal similarity between prime and target pairs resulted in faster and more accurate responses to word targets, regardless of prime lexicality. Given that the TL advantage and cost were comparable to native speakers, the results highlight the need for diverse comparisons to appropriately examine L2 visual word recognition.