From: LAURA SLOCUM 1. The concept of "multiple representations" makes sense, but is also nebulous (e.g., how many different "representations" does a person have?). I hope that we can consider during class discussion the question that Jan raises on p. 244: "Do children with phonological disorders have smaller lexicons because they have phonological disorders? Or, do these children have phonological disorders because they have small lexicons?" 2. Also valid is her point that "we need to consider general learning principles in the study of phonological acquisition" (p. 246). How do adults learn new words (or L2 learners) and how might this help to explain L1 acquisition in children? From: nagar@ling.ohio-state.edu The author presents a commentary on ?Lexical representation in acquisition?. Two themes are talked about. The first one has to do with the representation and their levels; lexical, phonological or phonetic, that the child is learning. In connection with this the author says that contrast between levels is not present at all levels and can only be present at any one level. The second theme pertains to the importance of lexicon in acquiring the sounds of a language. The author then talks about papers and works that support the themes that were laid down. Experiments show that 14 month olds have acquired some contrasts but not others. As the child acquires more vocabulary she learns to make contrasts between sounds. In the beginning of acquisition, contrast is minimal. The author raises the question of the role of lexical items and their volume in acquisition for children with phonological disorders. The role of lexicon in acquiring morphology and phonology is emphasized. The author ends the commentary with questions that would help future research like studying phonological acquisition in other languages, phonological acquisition in languages with different prosodic systems, influence of syntax in phonological acquisition etc. From: Fangfang Li this commentary outlines two major themes in phonological acquisition. the first one has something to do with a multidimentional system which encodes contrasts in different levels of representation. The second theme is related to the role of lexicon in the acquisition of phonological system. comments: 1. how many layers are there in the phonological representation? all the evidence are sporadic, and it seems a unified account is needed. 2. compared with other languages, mandarin chinese seems to be able to tease most of the factors out, since it has limited morphology and very simple CV syllabe structure. From: Eun Jong Kong - Children build the contrast at multi-level representations while they acquire a language. - The adjustment of contrast among different levels of representation would be achieved both by infants' cognitive efforts of meaning association with phonetic forms and by their articulatory efforts of accurate word production based on their awareness of phonetic details of articuation. - It suggests that the frequent experiences of linguistic forms affect infants'awareness of sublexical units and the accurate/independent use of those units in lexical learning. Kathleen Currie Hall Notes on: Edwards, Jan. (2000). Commentary: Lexical representations in acquisition. In Michael B. Broe & Janet B. Pierrehumbert (Eds.), Papers in Laboratory Phonology V: Acquisition and the lexicon (pp. 240-249). Cambridge: Cambridge University Press. Summary: In this article, the author reviews three papers presented at LabPhon V (one by Scobbie et al. on covert contrast; one by Stager and Werker on attentional effects of a lexical vs. discrimination task; and one by Beckman and Edwards on the role of lexical frequencies on the acquisition of phones vs. sequences of phones. She discusses two main themes of these papers and points out avenues of further research. Details: 1. The first theme is the nature of mental representations: phonology clearly consists of different levels of representation, and children can acquire contrasts at one level but not at another. 2. The second theme is the role of the lexicon: the lexicon is clearly important in developing grammar, especially in a model where grammar is simply made up of generalizations over the lexicon. Vocabulary size then plays a role in a child's ability to use language. 3. Future directions include looking at languages other than English (particularly ones with more complex morphological systems or different prosodic systems), looking at the influence of syntax on phonological acquisition, and looking at word frequency and familiarity. From: HELENA RIHA p. 240 ?Children build up a multidimensional system of phonological and phonetic representations. ? The word ?go?, for example, will be represented as a lexical item having to do with movement and taking a single argument at one level of re presentation.? I would argue that words are also represented as a part of idioms that children learn as units. In particular, a sample of speech in a psycholinguistics textbook (Carroll 2004) comes to mind in which a child says, ?Mommy, you're cramping my style.? That idiom is probably represented as one unit in the child?s lexicon. p. 241 ?A child can acquire a contrast at one level of representation, but not at another.? This is an important insight, and it would fit with the way children are exposed to language in the environment. Linguistic information is presented and perceived in different ways in naturalistic contexts, and it would make sense that the child gleans different bits of information about various units he or she is acquiring. Those pieces of information are eventually linked together once the child has developed an adult grammar and lexicon. p. 242 ?The 14-month old child, like the adult, hears fine phonetic detail in one task ? but completely ignores it in another task.? Isn?t this just a normal use of appropriate attentional sets for different linguistic tasks? In some hearing tasks, we listen for meaning (e.g., when hearing ?Happy New Year!?, we know the content of the phrase and do not need to listen for phonetic detail in the word or phrase), while in other linguistic tasks, we listen for phonetic detail (e.g., when hearing an international colleague?s name for the first time). Children seem to learn to adjust their attention at an early age as needed for different communicative tasks. This is what Edwards seems to allude to when answering the question of why young children don?t pay attention to fine phonetic detail in early word learning (p. 243). From: ASIMINA SYRIKA -Do bilingual children have different representations for their languages? At what age and how can these be observed? Does it have to do with dominance factors? -What about mapping? Do they have similar many-to-many mappings for their languages or is there overlap? -When you learn a second language later than birth-early childhood do you acquire different represenations for contrasts in the new lexicon even if these are 'similar' to the ones in your native language? How can you tell which are similar and which are not? Is it mainly phonetic similarity that matters? -How reliable are the findings that report that a given child has a contrast at one level of representation but not at another? -Ferguson & Farwell and Lindblom argue that a word is more likely to be learnt when it is similar to already existing words in the lexicon (p.241). What exactly is meant by 'similar'? Does similarity include aspects like prosody, number of syllables, stress, sound in the beginning of the word etc.? Could it also be the case that the more similar two words are the hardest it is not to confuse them? This is often a problem second language learners have. -What's the relationship between word familiarity and sound preference? How can you tell when a child is more familiar with a certain sound and when he/she prefers this sound? Is sound preference correlated with sound familiarity? -How certain can one be of a child's phonological knowledge at the different levels of representation? How do you measure this knowledge? -What's the role of the phonetic environment? If the child has a contrast in one place (eg. word-initially) but not in another what does that tell us about the child's representation of this contrast? -How do you measure the child's vocabulary size? -p.249 'Children's early two and three word utterances are usually composed of very familiar/frequent words (e.g. Bloom 1991). Given the diversity of input that children receive, what counts as a familiar/frequent word? From: Junko Davis - The author refers to the claim of Lindblom and summarizes the point as "the more similar a word is to already existing words in the lexicon, the more likely it is that the word will be learned." (p.241) What is this "similarity" referring to? Is it the overall similarity that consists of multiple aspects (which does not seem to have a tangible method to measure...)? - Is there any order of acquisition in the different levels of representation? Does each different level of representation tend to be independent or inter-dependent? - There is a very interesting question posed by the author -- Do children with phonological disorders have smaller lexicons because they have phonological disorders? Or do these children have phonological disorders because they have small lexicons? (p.244) Could there be any experimental method to resolve this question? - If a word is produced by a phonologically disordered child in an intelligible way but not accurate, is the word considered a part of his/her lexicon? From: Kirk Baker Edwards. Two themes covered: 1) children build phon[etic|ological] representations relating articulation<->acoustics and within-system contrasts, 2) lexicon as source of generalizations. Points out phonetic discrimination vs. word discrimination, covert contrast, generalizations from the lexicon. Last one seems like has to be well-constrained to account for the fact that people pretty much come up with the same kind of generalizations regardless of input base.