Poster #106 - Acquisition of letter-sound correspondence in Japanese-speaking 2-year-olds: An eye-tracking study

• In Event: 3-206 - Poster Session 13
  In Poster Session: PS 13 Section - Language, Communication

Integrative Statement

Understanding the underlying mechanisms of Japanese hiragana single-letter-reading acquisition is critical because reading hiragana is the fundamental stage of Japanese-reading acquisition. Japanese has a unique language system, which consists of phonographic hiragana and logographic kanji. Most hiragana letters are mapped to one sound, although many kanji characters have multiple pronunciations. Based on these features, hiragana reading tends to be acquired earlier than kanji, and most children can read almost all of the 71 hiragana letters before they start formal reading instruction (< 6-year olds) (Muraishi & Amano, 1972). However, it remains unclear at what age children start to learn hiragana letter-sound correspondence. When we consider vocabulary development, children first understand the meaning of words before they begin to produce them. Hiragana-letter reading might be acquired in a similar way, such that younger children who cannot read yet might understand letter-sound correspondence. In this study, we addressed young children’s implicit knowledge of hiragana using an eye-tracking technique.

Since Japanese toddlers generally start to read hiragana from about the age of 3, our participants were comprised of 19 younger (M = 27.59 mo., R = 25.50-29.10 mo., females: 8) and 18 older 2-year-old toddlers (M = 33.77 mo., R = 31.33-35.84 mo., females: 8). The children’s implicit letter-sound correspondence was evaluated using the intermodal preferential looking method (e.g., Golinkoff et al., 1987). Two hiragana letters were presented side by side on the screen, one of which was sounded as a target (“X: Which is X?”) Six pairs of stimuli to which two hiragana letters were yoked were presented twice to evaluate the children’s letter-sound knowledge of 12 letters.

To reduce the effect of visual preference, we evaluated the increase in looking at the target, which was calculated by subtracting the target looking ratio by looking at its ratio as a distracter (Bergelson & Swingley, 2012). A positive value indicates that a toddler was looking for sound matched letters. For older 2-year-old toddlers, the increase in looking at the target significantly exceeded zero (M = .068, p = .036, Wilcoxon test; Fig. 1), suggesting that they tended to fixate longer on the target stimuli. However, younger 2-year-old toddlers did not show a positive increase in looking at a target (M = -.001, p = .716, Wilcoxon test; Fig. 2), implying they did not distinguish target letters from distracters. After our eye-tracking experiment, we confirmed whether the toddlers were able to read the presented letters. For younger 2-year-old toddlers, 18 of 19 could not read all hiragana letters (p < .001, binominal test); 19 of 20 older 2-year-old toddlers could not read all letters (p < .001, binominal test).

Our present study showed that older 2-year-old toddlers who cannot read yet did significantly fixate on the target stimuli. Although previous studies argued that hiragana-reading acquisition starts at 3 years of age, our study suggests that toddlers over 2 years of age implicitly recognize the sounds of hiragana.

Authors

• Hiroki Higuchi, NTT Communication Science Laboratories
  Presenting Author

• Yuko Okumura, NTT Communication Science Laboratories
  Non-Presenting Author

• Tessei Kobayashi, NTT Communication Science Laboratories, Nagoya University
  Non-Presenting Author