FLEAT VI

The introduction of multimedia allows for a more effective use of images in vocabulary learning. When presenting vocabulary to L2 learners through a visual medium, one must be aware not only of how the word is presented, but also of the grouping with other vocabulary of similar meaning. This study investigates the mechanism of L2 lexical access and lexical network by looking at Japanese Learners’ process of L2 (English) orthographic representation (written word) and conceptual representation (imagery) during their L2 speech production. Forty-seven highly proficient Japanese learners of English (JLE) were tested on the Picture Naming task (PN) and the Word Naming task (WN) in controlled (picture/word only), matched (e.g. picture of a cat showed with a word “cat”), and unmatched conditions (e.g. picture of a cat showed with a word “dog”). The data gained from the experiments were analyzed to measure the speed of lexical access.

The results revealed that (1) conducting PN takes significantly longer than WN. (2) Among three conditions, the reaction time for the matched condition was the fastest by a statistically significant degree in both PN and WN. On the other hand, the reaction time for unmatched conditions was the slowest by a statistically significant degree in both PN and WN. Further analysis was conducted to investigate the picture-word interference paradigm. Unmatched pairs were divided by categorical relation and strength of relation. (3) Both categorically related pairs and strongly related pairs did not show stronger interference for JLE. It was concluded that (1) naming a picture forces learners to access its meaning, while naming a word (oral reading) could be done without accessing its meaning. (2) Having both verbal (written word) and non-verbal (picture) information accelerates the lexical access. (3) Even proficient JLE depend on their L1 lexical network when producing L2 words.