Examining radical position and function in Chinese character recognition using the repetition blindness paradigm

ABSTRACT

Repetition blindness (RB) is the failure to report the second occurrence of repeated items in a rapid serial visual presentation stream. The two-stage model of RB by Bavelier (1994) states that more properties shared between the repeated items lead to a larger RB effect. We used RB paradigm to examine the position (left or right) and the function (semantic or phonetic) of radicals in Chinese character recognition. Compared to the repeated radicals with the same position and function, RB was reduced when they were in different positions (Experiment 1A), but not when they had different functions (Experiment 1B). Similar RB-effect was observed when only one, or both, of the repeated radicals provided valid semantic or phonetic cues to characters (Experiments 2A and 2B). These results suggest that radicals are encoded with position but not function information. The radical function is likely implemented in lateral connections between semantic and phonological representations of characters.

KEYWORDS:

• Reading
• orthography
• semantic
• phonetic
• lexical access

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The other four dictionaries are the Mandarin Daily News Dictionary (He, 1981), the Chinese Character Dictionary (Huang, 1995), the Da-Xue Dictionary (Zhang, 1973), and the Chinese Dictionary (Zhou, 1996).

2. In R, the model was glmer (Accuracy ∼ 1 + Position * Function * Radical Repetition + (1 + Position * Function * Radical Repetition|Subject) + (1 + Position * Function * Radical Repetition|Trial), family = binomial, glmerControl (optimizer="bobyqa", optCtrl = list(maxfun = 100000))).

3. The model was glmer (Accuracy ∼ 1 + Function * Position * Radical Repetition + (1 + Function * Position * Radical Repetition|Subject) + (1 + Function * Position * Radical Repetition|Trial), family = binomial, glmerControl (optimizer="bobyqa", optCtrl = list (maxfun = 100000))).

4. The model was glmer (Accuracy ∼ 1 + Semantic Function * Radical Repetition + (1 + Semantic Function * Radical Repetition|Subject) + (1 + Semantic Function * Radical Repetition|Trial), family = binomial, glmerControl(optimizer="bobyqa", optCtrl = list(maxfun = 100000))).

5. The model was glmer(Accuracy ∼ 1 + Phonetic Function * Radical Repetition + (1 + Phonetic Function * Radical Repetition |Subject) + (1 + Phonetic Function * Radical Repetition |Trial), family = binomial, glmerControl(optimizer="bobyqa", optCtrl = list(maxfun = 100000))).

6. To date, nevertheless, there seems to be a lack of evidence that suggests that semantic radicals have the activation of their associated sounds.

Additional information

Funding

This research was supported by Taiwan’s National Science Council [grant no. NSC94-2752-H-002-008-PAE]; and Ministry of Science and Technology [grant no. MOST 101-2410-H-002-083-MY3] to Su-Ling Yeh.