The Orthographic Neighbor Effect In Chinese Character Recognition And Its Influencing Factors

Orthographic processing is the first phase of Chinese character recognition. A major role for Chinese character recognition has been attributed to the orthographic neighbors of a character, i.e., characters that are orthographically similar to a target character. Orthographic neighborhoods were defined as the characters with the same structure and the same phonetic radicals as targets in the present study. By using lexical decision task(LDT) with the masked priming paradigm, three experiments tested the effect of orthographic neighborhood and the impacts of the phonetic radical family size, consistency level and the relative frequency of the neighor and the target.In Experiment 1, the prime was the neighbor of the target with the same phonetic radical,a significant facilitatory priming effect was observed in response time for target characters that different from the previous results of alphabetic languages. This effect was not modulated by the relative frequency of the prime and target characters. Moreover, significant interaction between orthographic neighborhood size(N), character consistency and the frequency of target character was found in experiment 1 regardless of neighbor priming or unrelated priming. For the large N condition, high frequency characters of low consistency level were responsed faster than high consistency level, but the result was the opposite of low frequency characters. For the high consistency level condition, high frequency characters of small N were responsed faster than large N, and the low frequency characters were exact opposite. In experiment 2, we used pseudo-character as the prime with the same structure and the same phonetic radicals as the target but unpronounceable and no semantic. Then in experiment 3, the prime was replaced by false-character which came from the character in experiment 2 by exchanging the radical positons. Both pseudo-character and false-character showed that high consistency level were responsed faster than low consistency level, especially for large N of low frequency characters. Comprehensive analysis of the results of three experiments, we found that the facilitatory effect of neighbor priming condition was the strongest in identification of high frequency characters. The target character was responsed slower when the prime was pseudo-character. And false-character priming condition was the slowest. Howerver, there were no significant differences between the three priming conditions when identified the low frenquency targets.These results demonstrated that:1) Orthographic neighbor primes exerted facilitatory effects on target lexical decision. Moreover, this effect was not modulated by the relative frequency of the prime and target characters.2) Three factors of orthographic neighborhood size(N), consistency level and frequency of target character manipulated in three experiments were found to affected the orthographic processing interactively. In the large N, the recognition of high frequency targets was inhibited with higher consistency level, while the low frequency targets recognition was facilitated. In the condition of high consistency level, the larger the character family size, the slower identification of high frequency targets and the faster identification of low frequency targets.3) Orthographic position information did facilitate the recognition of Chinese characters. The neighbor with the same position of phonetic radical accelerated the identification of the target. And this facilitation arising from the position was more obvious in high frequency characters.4) The neighbor’s orthography and phonology were all activated when the target was presented. However, the phonology of higher frequency neighbors were activated earlier. This activation of phonological information allowed feedback to orthographic level, thus influenced orthographic processing.