Chapter 1 Introduction
1.1 General Background
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Chapter 1
Introduction
Beware of heard, a dreadful word, That looks like beard and sounds like bird.
And dead, it’s said like bed, not bead;
For goodness sake, don’t call it deed.
Anonymous (Fromkin, 2010)
1.1 General Background
In literate cultures, linguistic abilities encompass not only perception, production, and comprehension of speech, but also elegantly involve the elements of reading and writing of print. The author of the doggerel was trifled with a unique property of the creative human language system which imposes double jeopardy: Spelling patterns can have more than one pronunciation, and sound patterns can be spelled in more than one way (Peereman, Content, & Bonin, 1998; Stone & Van Orden, 1994). Hence, it is considerably essential to cognize that processing of a word can be influenced by knowledge of the pronunciation as well as spelling of other words. These interweaving mappings between spelling and sound consistency give reflective insight
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
into the organization of the mental lexicon which closely interconnect with several essential theories of word identification. Specifically, identifying the circumstances under which sound-to-spelling consistency arises is crucial to obtain profounder consideration concerning how internal lexical architecture reconstructs or adjusts to accommodate word processing from both visual and auditory inputs.
From a traditional information processing view, during the course of reading, it is generally assumed that the activation produced over spelling units feeds forward to sound units. The recognition routes flow from simply downstream, as from orthography to phonology. More than a decade ago, a fascinating hypothesis concerning the visual word recognition was put forward by Stone and Van Orden (1994). The novelty conception of Stone and Van Orden (1994) specified the significance regarding to the bidirectional mappings between the representations of phonology and orthography. In addition, the bidirectional consistency effect was predicted in the context of a recurrent network theory of word perception.
Figure 1. Schematic description of the resonance model (J. C. Ziegler, Petrova, &
Ferrand, 2008)
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
According to basic notion of recurrent network theory, the flow of activation is inherently bidirectional. The construction of visual word recognition is not only influenced by the feedforward mapping from spelling to sound (i.e., whether orthography is pronounced consistently) but also genuinely influenced by the feedback mapping from sound to spelling (i.e., whether phonology is spelled consistently). In reference to this model, orthography-to-phonology (O-P) consistency was called feedforward consistency; on the other hand, the feedback consistency refers to the phonology-to-orthography (P-O) mappings. Both in the visual as well as the auditory modality, P-O consistency is defined as the mappings from phonology to orthography (Ziegler, 2008) . Consistent symmetrical relations between sounds and spellings result in stable and fast activation, whereas inconsistent and asymmetrical relations slow down the system on its course to equilibrium (Tuller, 1994; Ziegler, Van Orden, & Jacobs, 1997).
The assertion in relation to the noticeable presence of a P-O consistency effect was surprising among the earlier theories of visual word recognition, which held the view that phonology should not matter in the visual processing stage. The concept of P-O consistency effect was also originally perceived as a major theoretical challenge to the traditional bottom-up theories of word recognition (Norris, McQueen, & Cutler, 2000). Stone, Vanhoy, and Van Orden (1997) were the first to demonstrate the
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
bidirectional consistency perspective in reading. Their statement of “phonologic coherence hypothesis” suggested that the resonance which emerges between
phonological and visual features was a coherent foundation for building “higher level”
resonances. In addition, phonology mediates word perception through the early local coherence, the primacy of visual-phonologic dynamics is guaranteed.
The aptitude to employ the correspondent relations between phonology and orthography information plays a significant role not only in the construction of reading stages, but also in the mechanism of spoken word recognition. Several researches have reported the influence of P-O consistency in the auditory modality.
Ziegler (2008) pointed out that, in contrast to the visual processing, P-O consistency effect in the auditory modality appears relatively robust. There have been a number of studies demonstrating that orthographic information is activated during auditory word recognition. One of the clearest demonstrations was originally reported by Seidenberg (1979). In the experiment, Seidenberg (1979) had listeners monitor a short list of spoken words for a word that rhymed with a prior cue word. The results demonstrated that the reaction time taken to decide that two spoken words rhyme was shorter when their spellings are similar (e.g., TOAST - ROAST) than when they had different spelling forms (e.g., TOAST - GHOST), and that the opposite effect holds for negative decisions (e.g., which are faster for LEAF - REAF than for LEAF - DEAF).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
These findings established the notion that orthographic presentations to some extent generate influence upon the cognitive processing in auditory modality (Donnenwerth-Nolan, 1981; Taft, 1985).
Plentiful significant empirical evidences from previous researches had conveyed the critical notion that the mapping relation from phonology to orthography is essential during visual and the auditory word recognition in alphabetic writing system.
Nevertheless, it is noteworthy for us to ponder this issue by further investigate whether P-O mapping takes place in the logographic language system like Chinese.
Theoretically, writing systems differ in the manner in which they represent phonological information as well as orthographic information. There are variations between orthographies in the extent to which written words represent their pronunciations among different language systems. According to the statement of Berndt, Reggia, and Mitchum (1987), half of all English words can be spelled accurately on the basis of sound-spelling correspondences alone, meaning that the letters used to spell these words predictably represent their sound patterns. On the other hand, the prominent linguistic trait for Chinese is that it belongs to the logographic writing system with relatively deep orthography, having a more opaque and arbitrary script-speech correspondence. In contrast with English, sound-spelling correspondence in Chinese is not generally solitary one-to-one mapping. Several
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
characters can be phonologically similar but visually dissimilar; on the other hand, some orthographically similar characters may have different pronunciations.
Nevertheless, there are some valid clues for pronunciation in the constituents of Chinese characters. DeFrancis (1989) pointed out that Chinese orthography is a speech-based script due to the fact that more than 85% of Chinese characters are phonograms, in which the clues to its pronunciation were embedded in a part of the character. Basically, phonograms are complex characters which typically composed of a semantic radical and phonetic radical. For example, in the character “媽”(/ma1/), the core formation of this phonogram is constructed by a semantic radical “女”and a phonetic radical “ 馬 ”(/ma3/). The composed phonetic radicals are often characterized as being an essential clue to specify the pronunciation of the entire character; the semantic radicals on the other hand offer a hint to the character’s meaning (Lee, Tsai, Su, Tzeng, & Hung, 2005). Increasing researches have demonstrated that reading a complex character might necessarily involves the processing of its radical (Yeh & Li, 2004). Therefore, the associations between radicals as well as the sound-spelling correspondences in phonograms are gradually regarding as a critical and discussible issue which relates to not only the visual word but also spoken word recognition.
The interlaced mappings between representations of Chinese phonology and
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
orthography had been investigated in several studies. Lee et al. (2005) suggested that the concept of “consistency” is a better index for describing the intricate relations
between sounds and spellings in Chinese language system. The various degrees of connections between phonograms and their radical parts can be addressed through the theoretical concept of consistency, which refers to whether the pronunciation of a character agrees with its orthographic neighbors. There has been a great deal of research investigating the processing relations between word recognition and the orthography-to-phonology (O-P) consistency in different languages and with different manipulations of orthographic information (Pattamadilok, Morais, Ventura, &
Kolinsky, 2007; Ventura, 2004; Ziegler, Petrova, & Ferrand, 2008). Furthermore, the flow of O-P consistency is essentially focusing on the mapping direction from spellings to sounds.
Different from the view of O-P consistency, theoretically, the definition of
“phonology-to-orthography consistency” (P-O consistency) is a statistical indication
of the mappings from phonetic sound to orthographic spelling in a Chinese character.
P-O inconsistent words possess the characteristics that their pronunciation is capable of being orthographically represented in multiple ways with different phonetic radicals (e.g., /jan2/ sounds can be orthographically represented by “研”, “顏” and
“延”). In contrary, the P-O consistent characters have a greater possibility that the
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
sounds could be mapped to specific phonetic radicals (e.g., /nɑw3/ sounds can be orthographically represented by “腦”, “瑙” and “惱” with the same phonetic radical).
The investigation concerning the mappings from phonologic codes to orthographic codes in non-alphabetic writing system such as Chinese is rather limited and lacking of a wide range of supplementary researches from other various perspectives and empirical evidences.