Research

Ohio State Dissertations in Linguistics (OSDL)

Kim Ainsworth-Darnell (1998)

The Effects of Priming on Recognition Latencies to Familiar and Unfamiliar Orthographic Formsof Japanese Words

Advisor: Julie E. Boland

Download .pdf file [PDF]

Abstract:

One of the longest standing debates in the visual word recognition literature concerns whether or not the transparency of the symbol-sound relationship for an orthographic system influences how words in that system are processed. The current study investigates this issue with Japanese, a language with multiple orthographies, including the phonologically transparent hiragana and the phonologically opaque kanji. Utilizing a visual lexical decision task, I presented native speakers of Japanese with the kanji and hiragana forms of nouns that were either familiar in both orthographies (orthographically neutral) or only familiar in kanji (kanji dominant).

In Experiment 1, targets were preceded by one of three visual primes: a semantically related real word, asemantically unrelated real word, or a string of three asterisks. In all three priming conditions, orthographically familiar targets were recognized more quickly and accurately than unfamiliar ones. Importantly, the expected priming effect was found, but it did not interact with the orthographic depth of the target word.

In Experiment 2, targets were preceded by one of three auditory primes: the spoken form of the target, a phonologically legal nonword, or a 100 ms tone. The pattern of the results was the same as in Experiment 1, with the additional finding of a larger priming effect for unfamiliar orthographic forms compared to their familiar counterparts.

In Experiment 3, targets were preceded by one of five visual primes: the target word in the same orthography, the target word in the opposite orthography, a nonword in the same orthography, a nonword in the opposite orthography, or a string of asterisks. The patterns of response times and error rates were similar to the previous two experiments, and the larger priming effect for unfamiliar orthographic forms compared to familiar orthographic forms was again observed. Collectively, these findings are consistent with previous work demonstrating that orthographic familiarity has a greater influence over the processing of written words in Japanese than orthographic depth (e.g., Besner & Hildebrandt, 1987; Hirose, 1984, 1985, 1992; Sasanuma, Sakuma & Tatsumi, 1988). They extend the literature by demonstrating that orthographic familiarity effects are reliable within-word and that phonological and orthographic identity priming affect the recognition speed for Japanese words according to orthographic familiarity, not orthographic depth. These results support a weighted network model in which the strength of the associations between semantic, phonological, and orthographic information reflects the reader's experience.