Orthographic Forms Research Articles

Improved effects of word‐retrieval treatments subsequent to addition of the orthographic form

Background: The participant, an individual with moderate–severe Wernicke's aphasia, had not benefited from two word‐retrieval cueing treatments in a previous investigation. The participant insisted that her performance would have been improved if the written word had been provided as part of the cueing process. This research was supported by Department of Veterans Affairs, Rehabilitation Research and Development. Aims: The purpose of this investigation was to examine the effects of adding the orthographic form of targeted action names to a semantic cueing treatment (SCT) and a phonologic cueing treatment (PCT). Methods & Procedures: The participant received SCT and PCT applied to the retrieval of action names. The treatments both provided the written word form paired with the pictured action in conjunction with cueing hierarchies. Two, sequential multiple baseline designs across behaviours were employed to examine the acquisition and response generalisation effects of treatment. Outcomes & Results: Improved accuracy of action naming was found for both treatments. Gains were limited to trained items; no changes were observed in naming of untrained actions. Conclusions: It appeared that the participant utilised the orthographic word form to develop associations between the visual object recognition system and the orthographic input lexicon, thus facilitating access to the phonological output lexicon.

Reading words in Spanish and English: Mapping orthography to phonology in two languages

English-Spanish bilinguals named visually presented words aloud in each language. The words included cognates (e.g., fruit-fruta) and non-cognate translations (e.g., pencil-ládpiz). The cognates were selected so that the orthographic and phonological similarity of their lexical form in each language varied orthogonally. Cognate naming latencies were influenced by the cross-language match of the orthographic and phonological codes. When the orthographic forms were similar in the two languages, naming latencies were slowed by dissimilar phonology, providing evidence for feed-forward activation from orthography to phonology across languages. When the orthographic forms were dissimilar, the effects of the corresponding phonological match were not statistically reliable. The results suggest that lexical access is non-selective across bilinguals' two languages, and that the degree of consistency between orthographic and phonological codes influences the manner in which cross-language competition is manifest. Findings are discussed in terms of feed-forward and feed-backward activation dynamics across languages.

Effects of Phonotactic and Orthotactic Probabilities During Fast Mapping on 5-Year-Olds' Learning to Spell

The purpose of this study was to investigate the orthographic-processing skills of typically developing 5-year-old preschool children. Of interest was whether phonotactic probabilities and/or orthotactic probabilities affected their ability to quickly learn the orthographic forms of 12 novel words. Orthographic processing was measured by the children's ability to spell and identify spellings of the novel words. Specifically, we were interested in whether (a) children quickly stored or "fast mapped" orthographic information after minimal exposure to novel words during storybook readings, (b) phonotactic and orthotactic probabilities affected orthographic fast-mapping skills, and (c) orthographic processing explained unique variance on a measure of the children's early spelling abilities. The results of this study indicated that young children quickly fast mapped orthographic information after minimal exposure to novel words, and their spelling (generation or reproduction but not recognition) was influenced by phonotactic and orthotactic probabilities. The significance of this work is that it demonstrates that preschoolers can fast map orthographic words they see onto spoken words they hear while listening to storybooks read to them and that the spelling of preschoolers is influenced uniquely by both phonological and orthographic information (probability of frequent letter and sound sequences in English words).

Can syllabification improve pronunciation by analogy of English?

In spite of difficulty in defining the syllable unequivocally, and controversy over its role in theories of spoken and written language processing, the syllable is a potentially useful unit in several practical tasks which arise in computational linguistics and speech technology. For instance, syllable structure might embody valuable information for building word models in automatic speech recognition, and concatenative speech synthesis might use syllables or demisyllables as basic units. In this paper, we first present an algorithm for determining syllable boundaries in the orthographic form of unknown words that works by analogical reasoning from a database or corpus of known syllabifications. We call this syllabification by analogy (SbA). It is similarly motivated to our existing pronunciation by analogy (PbA) which predicts pronunciations for unknown words (specified by their spellings) by inference from a dictionary of known word spellings and corresponding pronunciations. We show that including perfect (according to the corpus) syllable boundary information in the orthographic input can dramatically improve the performance of pronunciation by analogy of English words, but such information would not be available to a practical system. So we next investigate combining automatically-inferred syllabification and pronunciation in two different ways: the series model in which syllabification is followed sequentially by pronunciation generation; and the parallel model in which syllabification and pronunciation are simultaneously inferred. Unfortunately, neither improves performance over PbA without syllabification. Possible reasons for this failure are explored via an analysis of syllabification and pronunciation errors.

Pronunciation Keys in American Dictionaries

Working Knowledge: Hard Problems ofLexicography Pronunciation Keys in American Dictionaries T X 3 William A. Kretzschmar, Jr. University of Georgia 1he principal American desktop dictionaries all use pronunciation keys that are designed to make pronunciation transcriptions as immediately recognizable and intuitive as possible to dieir readers. The dictionaries' editors believe that their pronunciation keys either capitalize on English spelling or take advantage of traditional American diacritical marking practices. There is, however, considerable variation among the prominent dictionaries, especially in their treatments of low and back vowels and diphthongs. Some editors prefer heavy use of diacritical marks, including several marks more than the traditional use of macron (and sometimes breve) to mark "long" and "short" vowels. Especially among the diphthongs, there is also variation in the choice ofW or VC pairs, presumably because editors have chosen different common spellings as basic to the pronunciation. The result is the situation in American lexicography where no reader, whether naive or expert, can know what the transcriptions in the popular dictionaries represent without referring to the pronunciation key. Even when readers consult the key, they find sample words provided to illustrate die symbols, not internationally-agreed-upon IPA symbols or other means to represent specific phonetic values for the symbols of the key. The benefit of this practice is that American readers of dictionaries can pronounce words using their own regional variant pronunciations without being contradicted by the dictionary: for instance, Southerners can Dictionaries:Journal ofthe Dictionary Society ofNorth America 27 (2006) , 127-132 128William Kretzschmar pronounce the vowel of ride as a monophthong while Northerners use a diphthong, and vice versa for Upper Midwestern monophdiongal pronunciations of go and stay compared to Southern diphthongs. The same symbol from the dictionary key comes to represent more than one sound, which surely results in a poor representation of American pronunciation . American dictionary editors thus indulge in a mild subterfuge regarding American pronunciation, apparendy not wanting to annoy any part of their market with the facts. The situation is different in England, where IPA-based pronunciation keys are the norm. British dictionaries sell in die world market as well as the domestic market so actual phonetic pronunciations have value for English learners, and domestic buyers are thought to be able to cope with die facts even though British regional differences in accent are more marked dian American differences. The Oxford Dictionary of Pronunciation for Current English (Upton, Kretzschmar, and Konopka 2001), prepared for the world market, does offer IPA transcriptions of American English alongside British English. The model adopted there presents pronunciations that, if used, would not be recognized as regionally or socially marked. The subterfuge is gone, at the cost of the editors having to make decisions about regionally and socially marked variants . It is no benefit to dictionary users to avoid giving them specific information about pronunciation. American editors might just as well leave out the pronunciations altogether — except that their market does care how words are pronounced and uses dieir dictionaries as their authority . British editors have frankly accepted the responsibility to act as an audiority, and in so doing have accepted die need to create and to defend specific pronunciation models for practical use by their users. To address this problem, American dictionaries could use a pronunciation key designed with four goals: 1.To be as immediately recognizable and intuitive as possible, witii reference to Americans' predilection for spelling pronunciations . 2.To be as consistent internally as possible, in order to promote a realistic understanding of American pronunciation among readers. 3.To avoid unnecessary introduction of confusing multiple typographical resources like diacritics, caps, boldface, and italics. 4.To maintain a symbol-by-symbol correspondence with the IPA, so that each symbol represents only one sound. Pronunciation Keys in American Dictionaires129 Lee Pederson's ABC code (1987) already goes a long way in that direction , although not developed specifically for dictionaries. He notes that we lack an adequate transition between normal orthography and phonemic representation, so that "editors overlook sets ofphonological words ordered under a common orthographic form, tacitly enter phonological units without comment, or resort to the unsystematic phonographies of cartoonists and literary comedians" (1987, 49). The terms longand short for vowels...

Putting back the horse before the cart

A common assumption in studies of English world-wide is that speakers of an L2 frequently rely on the orthographic form of words in settling upon their pronunciation. I call this the “spelling form hypothesis”. This paper critically examines the assumptions behind this hypothesis and shows them to be implausible, and in fact fallacious for advanced interlanguages and fluent sub-varieties of a second language. Using data from Black South African English involving the equivalents of schwa in “standard” varieties (e.g. RP, General American and “cultivated” South African English), I argue that phonetic and orthographic forms often do not coincide. Furthermore, even where they coincide, other phonological factors than the use of spelling are usually at play.

Imaging phonology without print: Assessing the neural correlates of phonemic awareness using fMRI

Acquisition of phonological processing skills, such as the ability to segment words into corresponding speech sounds, is critical to the development of efficient reading. Prior neuroimaging studies of phonological processing have often relied on auditory stimuli or print-mediated tasks that may be problematic for various theoretical and empirical reasons. For the current study, we developed a task to evaluate phonological processing that used visual stimuli but did not require interpretation of orthographic forms. This task requires the subject to retrieve the names of objects and to compare their first sounds; then, the subject must indicate if the initial sounds of the names of the pictures are the same. The phonological analysis task was compared to both a baseline matching task and a more complex control condition in which the participants evaluated two different pictures and indicated whether they represented the same object. The complex picture-matching condition controls for the visual complexity of the stimuli but does not require phonological analysis of the names of the objects. While both frontal and ventral posterior areas were activated in response to phonological analysis of the names of pictures, only inferior and superior frontal gyrus exhibited differential sensitivity to the phonological comparison task as compared to the complex picture-matching control task. These findings suggest that phonological processing that is not mediated by print relies primarily on frontal language processing areas among skilled readers.

Doublet challenge: form comes before function in children's understanding of their orthography

Several current spelling models suggest that children cannot have any knowledge of orthographic form before they have acquired knowledge about orthographic function. We evaluated this proposition by using an orthographic choice task to inspect Finnish schoolchildren's knowledge of two aspects of consonant doublet use: the allowed doublet position (an aspect of orthographic form) and the type of phonemic information they represent (an aspect of orthographic function). The results challenged the view of the existing spelling models, since they showed that already at the beginning of the first school year children possessed formal knowledge of doublet use and knew that word-initial doublets are not allowed. However, these children were ignorant of the function of doublets, i.e. that they stand for long consonants.

Units of representation in visual word recognition.

As you read these words, a complex sequence of processes are at work in your brain, identifying visual patterns (letters) that are mapped onto familiar units (words), the meanings of which are combined to allow comprehension. In this description, a mental dictionary or lexicon linking word forms (orthography) to word meanings (semantics) plays a central and critical role in the reading process. However, some fundamental questions concerning the functional and neural organization of the mental lexicon remain unanswered. How are words composed of more than one unit (such as darkness ) stored in the lexicon? Are they stored as whole forms, or are complex words broken down into their constituent parts or morphemes ( dark and ness )? In this issue of PNAS, Devlin et al. (1) describe a functional MRI (fMRI) study that makes significant progress in understanding the functional and neural architecture of the systems involved in accessing the meaning of written words, and provide evidence in support of a controversial approach to the processing of complex words. Traditional linguistic accounts of how complex words are stored propose that the mental lexicon is organized morphemically. Morphemic organization ensures that there is no redundancy in the representation of related words created by using either derivational (e.g., trusty, distrust, untrustworthy ) or inflectional ( jumps, jumped, jumping ) morphemes. These theoretical arguments have been supported by the results of psycholinguistic experiments. For instance, responses to a simple word ( hunt ) are speeded or primed by a prior presentation of a related word ( hunter ), suggesting that these words have shared entries in the mental lexicon (2). However, priming could also arise from shared meaning or shared orthographic form, and research has also demonstrated that priming effects for morphologically related word pairs ( hunter – hunt ) are distinct from priming effects for items that have an equivalent amount of overlap …

Independent Access to Phonological and Orthographic Lexical Representations: A Replication Study

Traditionally, it has been assumed that written abilities are completely dependent on phonology (the so-called phonological mediation theory). In this view, the semantic system activates the phonological form of a word, which then activates the corresponding orthographic representation. This hypothesis predicts that in double naming tasks (two temporally close responses to the same picture in the two output modalities), the word selected for a written response will always correspond to the one selected for a spoken output. On the contrary, the orthographic autonomy theory hypothesises the independence of the written output lexicon, which, in the presence of severe damage to conversion procedures, could be directly activated from the semantic system. In this case, the theory predicts that double naming tasks give rise to lexically inconsistent responses. We describe a non-fluent aphasic patient, RA, with agrammatic speech and severe word- finding difficulties. He also presented with severe damage to the phonology-to-orthography and orthography-to-phonology conversion procedures. His performance on picture naming tasks was significantly worse in the spoken than in the written output modality. Errors were predominantly omissions. Few semantic paraphasias were present. Semantic errors did not arise from a deficit in the semantic system since the patient performed flawlessly on a semantic questionnaire and a word picture matching task and gave correct definitions of the items. Although not predicted by the phonological mediation theory, on double naming tasks the patient gave inconsistent responses (i.e. he correctly wrote the stimulus “z-e-b-r-a” but he orally produced “giraffe”). Taken together, these results further confirm the hypothesis that access to the orthographic form of words can be accomplished independently, without the mediation of phonology (the orthographic autonomy hypothesis).

Determining the role of phonology in silent reading using event-related brain potentials

The goal of the present study was to delineate phonology's role in silent reading using event-related brain potential (ERP) techniques. Terminal endings of high cloze sentences were manipulated in four conditions in which the terminal word was: (1) the high cloze ending and thus orthographically, phonologically and semantically congruent (e.g., The gambler had a streak of bad luck.); (2) a pseudohomophone that was orthographically incongruent, but was phonologically congruent to the anticipated ending (e.g., The ship disappeared into the thick phog [fog].); (3) a word that was orthographically, phonologically and semantically incongruent to expectations (e.g., The dog chased the cat up the Queen [tree].); or (4) a nonword and consequently orthographically, phonologically and semantically incongruent to expectations (e.g., The gas station is about two miles down the bole [road].). A N270 was elicited by orthographically incongruent words and nonwords (conditions 2, 3 and 4), likely reflecting violations of orthographic form expectations, while the presence of the N400 to semantically incongruent words and nonwords (conditions 3 and 4) reflected violations of semantic expectations. The relative absence of the N400 response to pseudohomophones (condition 3) indicates that integrating word meaning with sentential context is influenced by the phonological representation of the presented letter string. The implication of these results for theories of word recognition is discussed.

Effects of audio-visual presentation of target words in word translation training

Komaki and Akahane-Yamada (Proc. ICA2004) used 2AFC translation task in vocabulary training, in which the target word is presented visually in orthographic form of one language, and the appropriate meaning in another language has to be chosen between two choices. Present paper examined the effect of audio-visual presentation of target word when native speakers of Japanese learn to translate English words into Japanese. Pairs of English words contrasted in several phonemic distinctions (e.g., /r/–/l/, /b/–/v/, etc.) were used as word materials, and presented in three conditions; visual-only (V), audio-only (A), and audio-visual (AV) presentations. Identification accuracy of those words produced by two talkers was also assessed. During pretest, the accuracy for A stimuli was lowest, implying that insufficient translation ability and listening ability interact with each other when aurally presented word has to be translated. However, there was no difference in accuracy between V and AV stimuli, suggesting that participants translate the words depending on visual information only. The effect of translation training using AV stimuli did not transfer to identification ability, showing that additional audio information during translation does not help improve speech perception. Further examination is necessary to determine the effective L2 training method. [Work supported by TAO, Japan.]

Orthographic neighborhood and concreteness effects in the lexical decision task

The experiment reported here investigated the sensitivity of concreteness effects to orthographic neighborhood density and frequency in the visual lexical decision task. The concreteness effect was replicated with a sample of concrete and abstract words that were not matched for orthographic neighborhood features and in which concrete words turned out to have a higher neighborhood density than abstract words. No consistent effect of concreteness was found with a sample of concrete and abstract words matched for orthographic neighborhood density and frequency and having fewer neighbors and higher-frequency neighbors than the words of the first sample. Post hoc analyses of the results showed that orthographic neighborhood density was not a nuisance variable producing a spurious effect of concreteness but, instead, that the existence of higher-frequency neighbors constitutes a necessary condition for concreteness effects to appear in the lexical decision task. This finding is consistent with the hypothesis that semantic information is accessed and used to generate the responses in lexical decision when inhibition from orthographic forms delays the target word recognition.

Evidence for an automatic orthographic code in the processing of visually novel word forms

Some psycholinguistic theorists contend that lexical representations are intrinsically phonological, and that visual word recognition requires the matching of orthographic information to core phonological representations, while others postulate that in skilled readers orthographic and phonological representations attain a more equivalent status. We provide evidence for the latter position, showing that an automatically accessible orthographic representation for a phonologically familiar word can be established prior to any visual encounter with that word. We trained subjects with the phonology of unfamiliar words and then included them in orthographic form among the targets in a masked-priming lexical decision task. We showed a pattern of priming for these words at the first visual encounter consistent with under-specified orthographic representations. Repeated presentation in the lexical decision context did not change this pattern of priming, suggesting that a more detailed or semantic analysis might be necessary for these representations to become as specific as those for familiar words.

A test of the orthographic recoding hypothesis

The Orthographic Recoding Hypothesis [D. E. Gaygen and P. A. Luce, Percept. Psychophys. 60, 465–483 (1998)] was tested. According to this hypothesis, listeners recognize spoken words heard for the first time by mapping them onto stored representations of the orthographic forms of the words. Listeners have a stable orthographic representation of words, but no phonological representation, when those words have been read frequently but never heard or spoken. Such may be the case for low frequency words such as jargon. Three experiments using visually and auditorily presented nonword stimuli tested this hypothesis. The first two experiments were explicit tests of memory (old–new tests) for words presented visually. In the first experiment, the recognition of auditorily presented nonwords was facilitated when they previously appeared on a visually presented list. The second experiment was similar, but included a concurrent articulation task during a visual word list presentation, thus preventing covert rehearsal of the nonwords. The results were similar to the first experiment. The third experiment was an indirect test of memory (auditory lexical decision task) for visually presented nonwords. Auditorily presented nonwords were identified as nonwords significantly more slowly if they had previously appeared on the visually presented list accompanied by a concurrent articulation task.

Acquired dysgraphia in Chinese: Further evidence on the links between phonology and orthography

Background: How similar is the "architecture" of language function in different languages? In the domain of reading and writing, the relationship between orthographic and phonological units of the language differs dramatically between alphabetic and logographic writing systems. It is therefore appropriate to ask whether an architecture assumption, such as direct links between word pronunciations and their written forms, developed in the course of research on English and other alphabetic writing systems, applies to logographic Chinese. Aims: We describe a native Cantonese speaker, TUA, with an acquired dysgraphia following a right hemisphere stroke. The study was designed to identify the conditions under which TUA succeeded or failed to produce the correct orthographic form of words in his native language, and thereby to draw inferences about his impaired and intact writing procedures with further implications for normal writing. Methods & Procedures: Nine different single-word tests were administered to TUA, mainly of writing to dictation but also of written picture naming and other relevant skills. Almost all monosyllabic words in Cantonese are homophones and, in the majority of our tests, the target words were lower-frequency homophones, with the intended alternative disambiguated by a semantic cue. Outcomes & Results: TUA made few errors in writing high-frequency words to dictation, but had error rates as high as 60-70% on lower-frequency homophonic target words. The majority of his errors were homophones of, or at least closely phonologically related to, the target item. This deficit in writing to dictation co-existed with adequate ability to define the correct lower-frequency homophonic alternative, and with the unusual pattern of more correct responses in written picture naming than in writing to dictation of the same target items. Although TUA's poor performance in writing was accompanied by several nonlinguistic impairments (e.g., in visual memory, sequencing, and visuo-constructional ability), his dysgraphia had a predominantly linguistic nature and did not invite an account in terms of these additional cognitive impairments. Conclusions: We propose that TUA had a disrupted ability to activate lower-frequency orthographic representations, especially under conditions of response competition from a higher-frequency written word with the same pronunciation. This pattern is compatible with a proposal of direct activation of orthography by phonology, at least for higher-frequency words, in Chinese.

Verschriftungsversuche des Ägäis-Makedonischen im 20. Jahrhundert

The article presents a survey of all the attempts that have been made in the 20th century to write down the Aegean Macedonian dialects who are without any Slavic roofing since 1912-13. In the years between 1943 and 1956 these attempts followed the ethno-political zig-zag course of the Greek, Yugoslavian and Soviet CPs, vacillating between the use of dialectal varieties and the Macedonian standard language. As a result of the Macedonian ethnic revival in the 1990s, the language policy of the RAINBOW-party can be assessed as a repeated failure to establish a regiolectal variety which would be in conformity with regional identity. Today, however, the use of Greek characters in written Aegean Macedonian texts has a relatively stable orthographic form. Future reversing language shift measures can build on it.

The automatic cognate form assumption: Evidence for the parasitic model of vocabulary development

The Parasitic Hypothesis, formulated to account for early stages of vocabulary/L/development in second language learners, claims that on initial exposure/L/to a word, learners automatically exploit existing lexical material in the L1/L/or L2 in order to establish an initial memory representation. At the level of/L/phonological and orthographic form, it is claimed that significant overlaps/L/with existing forms, i.e. cognates, are automatically detected and new forms/L/are subordinately connected to them in the mental lexicon. In the study reported here, English nonwords overlapping with real words in Spanish (pseudocognates), together with noncognate nonwords, were presented to Spanish-speaking learners of English in a word familiarity task. Participants reported significantly higher levels of familiarity with the pseudocognates and showed greater consistency in providing translations for them. These results, together with measures of the degree of overlap between nonword stimuli and translations, were interpreted as evidence for the automatic use of cognates in early word learning.

The Semantic Effect on Retrieval of Radicals in Logographic Characters.

This study examined the semantic effect on retrieval of radicals ofJapanese kanji. In the retrieval task, a stimulus word written inhiragana (Japanese syllabary) was presented one by one on a display, andparticipants quickly named the left radical of the target kanjicharacter that corresponded to the stimulus hiragana word. It was foundthat the mean naming latency was shorter and fewer errors were made whenthe left radicals were semantically related to the target kanji thanwhen they were not. Also remarkable was a momentary retrieval failure(i.e., no response) characterized as a `tip-of-the-pen' state which evena high-frequency word precipitated more often in the semanticallyunrelated condition. These results highlight a critical role thatmeaning plays in the activation of orthographic forms of kanji. Somecharacteristic features of writing in kanji are discussed.

Deviance Analysis of English Language Test Data to Explicate World English Varieties

As yet, there has been little evidence of major testing organizations pursuing research that takes account of the World Englishes research paradigm as propounded by Braj B. Kachru (1985 & ff.). Drawing on the synchronic, etic approach to the study of diverse corpora developed in Hassall (1998) and Hassall and Ganesh (1996, 1999), this paper proposes that a deviance analysis should be undertaken of data from international English language tests. It is proposed that the statistical technique of correspondence analysis (CA) should be applied to the frequency of orthographic forms across discrete groups of individuals from diverse geographical and linguistic regions, whilst undertaking a language task identified as meaningful by the major testing organizations. Once correlated with readily available data regarding the participants’ regional background and first language, it is suggested that valuable profiles may be assembled and comparisons be made of the selection of formal features of the candidates’ English language performance. It is hoped that the present empirical approach may encourage further studies that support pedagogical innovation developing a world Englishes perspective, as depicted in Hassall, 1996a (1996b, 1996c and forthcoming, 2000).