LouAnn Gerken - US grants

ABSTRACT The research examines children's representations of two potential cues to the location and grammatical category of words and phrases: linguistic stress and function morphemes. The overwhelming majority of English words begin with a strongly stressed syllable. If children were aware of this regularity, they could use it to identify the beginnings of words in the speech stream. Function morphemes, such as articles and verb inflections, occur at the beginnings and ends of phrases, and could help children to isolate these units. Function morphemes also provide cues about the type of unit that has been isolated, such as Noun Phrase or Verb Phrase. However, we know little about children's representations of stress and function morphemes, and therefore about whether they are able to use such cues in language acquisition. One set of experiments explores the phonological and syntactic nature of function morpheme representations with the use of imitation picture pointing tasks. Another set of experiments is designed to extend previous work on the effect of stress patterns on children's morphemic and non- morphemic weak syllable omissions, and to provide new information about their representation of the canonical stress patterns of their language.

9411185 Gerken ABSTRACT This research examines infants' and children's sensitivity to two potential cues for early language acquisition: prosody and function morphemes. Prosodic changes such as pausing, pitch resetting and syllable lengthening potentially cue the locations of linguistically relevant units. Function morphemes such as articles and verb inflections occur in grammatically specified contexts (e.g.,"the" occurs only in NPs) and potentially allow learners to distinguish among phrase types. One set of proposed experiments examines the effects of prosody and function morphemes on 2-year-olds' sentence comprehension. Another set uses sentence imitation with 2-year-olds and preferential listening with 9-month-olds to examine syntactic and discourse factors that might influence learners' sensitivity to phonological phrases-prosodic units that appear to govern some aspects of prosody in adult speakers. ***

The most frequent source of data used to understand the nature of both normal and disordered language development is children s productions. The properties of child utterances that deviate from adult forms are of particular interest, since they appear to reflect parts of the adult system that the child has not yet mastered. One of the most common types of deviation from adult forms seen in normally developing children between the ages of 24 and 30 months, as well as in older children with language disorders, is omissions, including omissions of mono-syllabic grammatical morphemes (e.g., the ) and unstressed mono-morphemic syllables (e.g., the first syllable of giraffe ). Data from numerous researchers suggest that such omitted syllables are not simply missing from children s representations of language. The proposed research continues in this framework, but begins to move beyond the question of whether normally developing children have some representation of syllables they omit to a question of why they omit these syllables. In keeping with this expanded focus, the proposal begins to draw more heavily on studies of adult language production as a framework for understanding child production. The proposed experiments are divided into three sets. Exps. 1-4 are based on observations of the role of prosody (i.e., rhythm and intonation) in children s omissions, along with an observed bias for children to produce final syllables. The goal of these experiments is to compare two prosodic accounts of children s omissions, one proposed previously and a new account derived from current linguistic theories of prosody as well as studies of adult language production. If the new account is supported, it would provide a unified account of English-speaking children s weak syllable omissions and a framework for viewing the relation of immature and mature language systems. Exps. 5-8 are based on the observation that children are more likely to omit weak sylla bles from sentences containing an unfamiliar word. These experiments promise to shed light on how utterance complexity affects child productions and how prosody fits within the larger language production system. Two of the experiments in this section also use work on adult language production to make predictions about child production. Exps. 9-10 are based on a pilot study that suggests children maintain a slot for weak syllables that they appear to omit. These experiments have the potential to confirm the notion that children have some representation of elements that they do not produce in an adult-like way. They also have the potential to relate child omissions to studies of adult production. Taken together, the experiments will reveal considerable information about normally developing children s growing knowledge of language and how this knowledge is implemented by their changing language production system.

One way to study the properties of normal adult language is to focus on these properties in other populations in whom language is just developing or in whom it has broken down. This approach to the study of language depends on finding a common pattern across different language users that can give us insight into how language works as a whole. This dissertation research project will use three populations (young children with normally developing language, slightly older children with specific language impairment, and adults with nonfluent aphasia) to address what role prosody (e.g., the rhythm structure of words and sentences) plays in their speech and how mental representations of language affect the production of speech. Specifically, the research will examine a particular phenomenon common to the speech of these populations: the omission of initial unstressed syllables within multisyllabic words. Acoustic data will be used to investigate the question of whether, when children or language disordered populations omit these syllables, their mental representation of the prosody of the words remains intact, and how prosodic structure is preserved in their productions.

[unreadable] DESCRIPTION (provided by applicant): The last 50 years of research on language development have revealed in young language learners two, potential conflicting, abilities. On the one hand, infants are exquisitely sensitive to specific surface properties of their input and the statistics thereof. On the other hand, infants are able to rapidly generalize beyond their input based on highly abstract formal properties. The relative importance of sensitivity to surface versus abstract linguistic properties is at the heart of theoretical debates about the nature of human language. [unreadable] [unreadable] The research proposed here reflects our ongoing attempt to determine how language learners combine their sensitivity to surface properties of language and their ability to abstract beyond these properties. From the research performed during the current period of funding, we have formed the following three hypotheses: (1) Memory demands, task demands and previous experience all influence the level of abstraction achieved for formally identical input. (2) Language learners form only the most abstract representation necessary to account for the set of data they have encountered. (3) Under some circumstances, adults with subtle developmental language disabilities are less likely than adults with normal language to generalize based on abstract properties of language. [unreadable] [unreadable] To test these hypotheses, participants are familiarized with stimuli from an artificial language-like system. They are then tested on new stimuli that are either consistent or inconsistent with the particular properties of the familiarization stimuli under study. Significant discrimination of consistent versus inconsistent test items is taken as evidence of learning. Using this paradigm, the research compares learning across different input conditions (two formally identical language systems with different surface instantiations, formal systems that conform versus do not conform to properties of natural languages) and across populations (6- to -12-month-old infants, adults with normal language, adults with impaired language abilities). The findings from this research are likely to reveal important information about how infants and adults are influenced by the specifics of their language input as they generalize beyond it. The findings also have the potential to provide new insights into the role of learning in language disorders. [unreadable] [unreadable]

Nationally and at The University of Arizona, far too few women seek--and succeed in faculty careers in science and engineering fields. The destructive role played by subtle discrimination is becoming well-understood. Without intervention, an overly homogeneous culture tends to reproduce itself through the operation of unsuspected biases. Even well-intentioned interventions may be thwarted by subtle discrimination. The University of Arizona seeks not just to diversify its science and engineering faculty but to permanently eradicate subtle discrimination, building programs to counteract its effects while cultivating new practices and new ways of thinking. The Provost, academic Vice Presidents, and Deans will join in developing and implementing strategies on three fronts. First, new programs will be launched to increase the number of stars among women scientists by fostering the development of social capital through networking, collaboration, and mentoring with local, national, and international scientific communities. Second, new expectations will be set for the stewardship of faculty careers through programs aimed at leaders, administrators, and departmental groups, including inquiry-based learning. Third, inequitable practices will be eradicated through development of technology with a dual function of gradually changing attitudes while directly changing the way business is done at the university: software implementations of personnel processes will have built-in mechanisms for evaluating the impact of every
decision on overall equity. Eradicating subtle discrimination against women faculty will produce healthier science and engineering disciplines, and it will inevitably elevate awareness of all other forms of subtle discrimination.

Infants and children are remarkably adept at learning language, much more adept in fact than their parents. Until recently, two theories competed to explain the language learning skill of young children. In one view, infants and children are like sponges, soaking up every detail of their environment, including the linguistic environment. In a competing view, infants come into the world with the range of human languages (or grammars) already "built-in." Their linguistic environment serves only to allow them to select the particular grammar of their native language, such as what speech sounds will be used, whether verbs come in the middle or the end of sentences, etc. Recently, a third view has begun gaining support. This view shares the position of the first view that infants and children are highly sensitive to every detail of their environment. It shares the position of the second view that infants and children have as their goal determining the particular linguistic system or grammar of their community. One way to discriminate among these three theories is the amount and kind of input required for optimal language learning. The first "sponge" view (or Associative theories) suggests that a great deal of input is needed and that it does not matter much whether that input consists of many repetitions (tokens) of the same example (type; e.g., hearing the word dogs as the plural of dog multiple times) or different examples (e.g., hearing dogs, pigs, and mugs as the plurals of their respective singulars). The second "built-in grammars" view (or Innate Domain theories) suggests that extremely little input is needed and that even a single repetition of a single example is sufficient for the infant or child to learn a particular component of the native language. The third "grammar-seeking sponge" view (or Hypothesis Selection theories) suggests that a very small number (e.g., three) of different examples are needed for learning and that different examples are much more helpful than repetitions of an example. This project will compare these three theories by asking about the amount of input needed for language learning. It also attempts to embed the third theory in a more realistic model of human learners by asking if learning always improves as the number of examples increases, or whether there is a trade-off between the number of examples and the number of repetitions of each example.

This research will contribute to our scientific knowledge by asking whether humans are reflections of their environment, their biology, or whether and how the mind is a product of a biological brain with a primary goal of "making sense" of the environment. Moreover, because the research draws on existing studies of infant vision and infant social interaction, it places the study of language development in a broader developmental context. Finally, the research explores the specific kinds of input that maximize learning. Therefore, it has the potential to ultimately inform how we present linguistic and other information to infants, school children, and even adults in order to result in the best possible learning in the shortest possible time.

Recent work in language acquisition and cognitive development shows remarkable learning abilities in infancy. Much of the theoretical development in these fields is based on effects measured immediately after a learning experience, however, sleep is instrumental in transforming specific details of what is learned to a more abstract memory (Gomez, Bootzin, & Nadel, 2006; Hupbach, Gomez, Bootzin, & Nadel, 2009). The ability to abstract away from the specific details of a learning experience is crucial for infants who must be able to summarize and apply key aspects of a learning experience to novel scenarios, much like being able to abstract the block letter "A" to cursive. If memories are too specific infants will not be able to connect prior learning to new scenarios with slightly different information. A more abstract memory can more easily be applied to a wider range of information. A first project will investigate the means by which sleep leads to abstraction. A second project investigates how sleep-dependent memories are connected across time in an attempt to understand how knowledge is amassed over multiple learning experiences. Polysomnographic recording will provide information about how sleep-dependent memories are consolidated in the developing infant brain.

The proposed work is unique in bridging three areas of research: language acquisition, memory, and sleep. It has potential to be transformative to the degree that it 1) impacts language learning theories (to date based on results obtained immediately after a learning experience, not taking the changes associated with intrinsic sleep and memory processes into account); 2) the way empirical learning research is conducted (to scale up to the constraints of real-world learning researchers will need to begin to measure time-dependent effects); and 3) informs us about the relationship between phases of sleep and memory formation in the developing infant brain, dynamics that could have a profound effect on theories of language and memory change, on understanding when normal change goes awry, and for learning in educational practice. In addition to the practical benefits for society, the proposed work has benefits closer to home with training of undergraduate students a significant part of the grant. These students will gain extensive one-on-one experience in conducting scientific research that will prepare them to be highly competitive candidates for graduate programs, and ultimately, for careers in teaching and science.

When we listen to someone speaking sentences in our language, each word seems as clearly identifiable as a familiar face in a crowd. But in fact, each time we hear a particular word spoken, it is acoustically different from the last time we heard it spoken. Different talkers, different rates of speech, and different emotional states all affect the specific acoustic form that a word takes. Experienced language users have no trouble overcoming this variability in the acoustic manifestation of words, but people who are still learning the language - young children and second language learners - struggle. Thus, uniquely identifying particular words from a highly variable acoustic signal is a skill that must be mastered through experience for each language learned. The main goal of this research is to explore one factor that might contribute to that mastery. A secondary question is how words are represented in the mind so as to allow both rapid identification from the acoustic signal and rapid production.

Understanding when input from multiple talkers is helpful for learning new words will address the broader question of how people learn to perceive and produce words, a crucial task in learning a first or second language. Additionally, the way that speech perception and production are related has been a long-standing issue in phonetics as well as in language development. There is usually agreement that the knowledge about how to perceive a word influences knowledge about how to produce that word. However, there is much disagreement as to whether there is a single representation that is used for both perceiving a word and producing a word, or whether there are two different, but related representations. If there is a difference between perception and production in terms of benefit from multiple talkers, we will have evidence for different representations for perception and production, at least when a word is newly learned. A more nuanced understanding of the benefit of multiple talkers may inform methods for second language learning. More broadly, it may also shed light on how language is represented in the human mind.

This research addresses two puzzles about language: Only humans have it, and learning a language as an infant results in better language skill than learning as an adult. Learning a language entails discovering rules, such as the English rule that the plural ending on nouns either sounds like "z" as in "kids" or like "s" as in "kits". In simplified terms, the rule is: add the "z" sound if the word ends in a voiced consonant ("d" is made with vocal cords vibrating) OR add the "s" sound if the word ends in a voiceless consonant (no vocal cord vibration for "t"). 11-month-olds can learn an "OR" rule from just 4 words of an invented language, but adults cannot learn the rule even from 76 words. Non-human animals, like human adults, also fail to learn OR rules in the visual domain (e.g., press the button for either small black squares OR large white triangles). Thus, infants are adept at learning a type of rule that is frequent in human language, but that adults and animals do not easily learn. Understanding why infants succeed so easily promises to reveal differences in how adults and infants learn language, and will perhaps allow superior, infant-like, language learning by computers.

Infants are exposed to words from an invented language that reflects one of three types of rules (the OR rule or 2 other types that are more easily learned by adults and animals). Their rule learning is tested by measuring their attention to new words that reflect the same or a different rule. Adults are exposed to the same words under several conditions that are designed to improve learning of the OR rule. Their rule learning is tested by asking them to rate new words on their similarity to the words that they just heard.

To learn a language, children must go beyond simply imitating speech and learn the rules of the language from their surrounding linguistic environment. One way to tell that children learn rules is that they apply rules to an overly broad set of words. For example, English-learning children produce forms like "breaked" or "catched" at around 3 years of age. These forms, called past tense overregularizations, show that children have implicitly discovered the past tense formation rule: namely, "add -ed to a verb to create the past tense." But does it take children until age 3 to discover such rules? A better understanding of the timeline of rule discovery will help us to develop more realistic milestones to measure children's progress in first language learning. It will also help to us understand why the past tense form is so difficult for second language learners of English and allow more informed pedagogical intervention for these learners.

This work has two goals: the first is to seek evidence that the past tense rule can be found in 16-month-olds, which would be consistent with many observations that children display significant receptive abilities long before their productive abilities reach the same level. The second goal is to determine if the production of past tense overregularizations can be explained by factors other than the discovery of a grammatical rule. There is growing evidence that overregularizations in production reflect children's need to simplify their utterances as they attempt to say more and more complex sentences. Factors that may influence a child's need to simplify utterances are frequency of the lexical item and the child's age. The proposed project will combine a behavioral study with 16-month-olds and a corpus analysis of longitudinal, spontaneous child speech to achieve these goals.

Abstract A hallmark of English-learning children with developmental language disorder (DLD, AKA specific language impairment) is the inconsistent production of grammatical morphology. However, recent work also implicates the phonological domain, as indicated by deficits in producing word and nonword forms. The hypothesis driving the current proposal is that morphological and phonological deficits are causally linked by a broader deficit in sequential pattern learning. The unique approach taken here combines what we already know about morpho-syntactic deficits in DLD with recent developments in the fields of linguistics and language acquisition. First, it is possible to divide phonological and morphological patterns into three pattern types (Single Feature, OR/Disjunction, Family Resemblance/Prototype), with these types having a long history of study in visual pattern learning. Importantly, children with DLD appear to have maximum difficulty with morpho- syntactic patterns of the OR type (e.g., regular past tense). In contrast, studies using artificial grammars show that infants who are typically developing are highly adept at learning Single Feature and OR pattern types; Family Resemblance patterns may be weaker. Typical adults are adept at Single Feature and Family Resemblance patterns, but appear to be, at least superficially, more like children with DLD in their performance on the OR pattern. With these intriguing findings as a starting point, the proposed research links phonological and morphological sequence learning in children with DLD. In Aim 1, we ask if 4- to 6-year-olds (typically developing (TD), DLD alone, DLD + speech sound disorder (SSD), and SSD with no morpho-syntactic deficit) and adults are sensitive to input examples that fit one of these three patterns. We predict that, consistent with their long-documented morpho-syntactic deficit, children with DLD will have particular difficulty with the phonological OR pattern. Consistent with their intact morpho-syntactic skills, children with SSD should show no deficits in the OR pattern, revealing a link between OR pattern learning, phonology, and morpho-syntax. Aim 2 explores whether the inclusion of a semantic subcategory cue facilitates learning the OR pattern. Aim 3 asks if dependence on the associatively organized lexicon can account for the infant-to-adult developmental changes observed for the OR pattern (which is not associatively organized) and thereby explores the possibility that children with DLD rely on their lexicons to compensate for their sequential pattern learning deficit. The results of the proposed studies promise to help identify the underlying mechanism(s) of DLD and to suggest possible intervention strategies, such as employing semantic cues to the OR pattern and strengthening lexical organization.