Graphemes are perceptual reading units

Grapheme coding in L2: How do L2 learners process new graphemes?

“…More precisely, complex graphemes were responded to slower than simple graphemes, a finding in line with observations on monolingual adults (Rey et al, 2000) and children (Marinus & De Jong, 2011). This implies that graphemes were functional units during visual word recognition in L2.…”

“…Experiment 2 was designed to further test graphemic coding in L2 by using nonword stimuli in a letter detection task. As previously observed in monolinguals, graphemic effects do not depend on the lexical frequency of the words (Rey et al, 2000), leading to a theoretical interpretation of this effect at a pre-lexical level. So, if the findings of Experiment 1 are robust and unbiased by the small number of items in one of the conditions, then a similar pattern of grapheme coding should emerge in Experiment 2 when using nonwords and a larger number of items in all three conditions.…”

“…Aside from these specific cases, matching between the three conditions was conducted on a 1-to-1 basis as far as possible. Nevertheless, given the sublexical locus of the grapheme effect (Rey et al, 2000), this should not bias our data. Table 1 represents mean RTs and percentage error (and standard deviations) for the letter-present targets.…”

“…A similar effect was found using a perceptual identification task for high and low frequency words (Rey, Jabobs, Schmidt-Weigand & Ziegler, 1998;Rey & Schiller, 2005) and nonwords (Bolger, Borgwaldt & Jakab, 2009) containing either complex or simple graphemes. Interestingly, Rey, Ziegler & Jacobs (2000) showed that detecting a letter embedded in a complex grapheme (e.g., detecting the letter "A" in beach) was slower than detecting this letter in a simple grapheme (e.g., such as the "A" in black).…”

“…Three experiments using the letter detection task (Rey et al, 2000) were presented to L2 learners with varying lengths of L2 exposure to assess the extent of grapheme coding in a L2.…”