The Effects of L2 Lexical Frequency, Language Context, and Script differences on L2 → L1 Noncognate Masked Translation Priming:A Behavioural and Electrophysiological Investigation

Previous research on bilingual lexical processing has established asymmetric translation priming patterns in unbalanced bilinguals, with stronger priming from the first language (L1) to the second language (L2) than vice versa. Although this phenomenon has been replicated several times, notably under masked priming conditions, the mechanisms underlying these asymmetric priming patterns remain unclear. Experiments 1A and 1B of this thesis examined the hypothesis that such masked translation priming asymmetries occur as a result of a temporal delay of L2 words relative to L1 words in the bilingual lexicon. To this end, unbalanced Spanish-English bilinguals performed lexical decision tasks (LDT) to L1 targets preceded by L2 masked translation primes or unrelated primes that were either as frequent as their L1 targets (Experiment 1A) or more frequent than L1 targets (experiment 1B). The results showed significant masked translation priming only when L2 primes were more frequent than L1 targets. To investigate whether the general activation level of the weaker language (i.e., L2) can be increased as a function of the language context of the task, unbalanced Spanish-English bilinguals performed mixed-language LDTs (experiment 2A) and single-language LDTs (experiment 2B) combined with the masked translation priming paradigm. The results showed that in the mixed-language context, remarkably similar translation priming effects were obtained in both directions, whereas in the single-language context, the traditional translation priming asymmetry emerged. To examine whether disparity of script between L1 and L2 affects L2→L1 translation priming, two further experiments using Arabic-English bilinguals were conducted in both directions. Experiment 3 followed experiment 1B in terms of frequency manipulation in which the critical materials consisted of English (L2) words having higher frequencies than their Arabic (L1) translations and Experiment 4 followed experiment 2A in terms of the language context of the task in which frequency-matched translation pairs were presented in mixed-language LDTs. The results from both experiments showed that despite the attempts to increase the chances of obtaining L2→L1 masked translation priming, a clear direction-dependent translation priming asymmetry was found as significant priming emerged only in the L1→L2 direction. These findings taken together indicate that the masked translation priming asymmetry is a multifactorial phenomenon where the elusiveness of L2→L1 translation priming depends on various factors including subjective word frequencies, the language context of the task, and the graphic distance between L1 and L2