Infant word segmentation revisited: edge alignment facilitates target extraction

In a landmark study, Jusczyk and Aslin (1995) demonstrated that English-learning infants are able to segment words from continuous speech at 7.5 months of age. In the current study, we explored the possibility that infants segment words from the edges of utterances more readily than the middle of utterances. The same procedure was used as in Jusczyk and Aslin (1995); however, our stimuli were controlled for target word location and infants were given a shorter familiarization time to avoid ceiling effects. Infants were familiarized to one word that always occurred at the edge of an utterance (sentence-initial position for half of the infants and sentence-final position for the other half) and one word that always occurred in sentence-medial position. Our results demonstrate that infants segment words from the edges of an utterance more readily than from the middle of an utterance. In addition, infants segment words from utterance-final position just as readily as they segment words from utterance-initial position. Possible explanations for these results, as well as their implications for current models of the development of word segmentation, are discussed.     

Rapid recognition at 10 months as a predictor of language development

Infants' ability to recognize words in continuous speech is vital for building a vocabulary. We here examined the amount and type of exposure needed for 10-month-olds to recognize words. Infants first heard a word, either embedded within an utterance or in isolation, then recognition was assessed by comparing event-related potentials to this word versus a word that they had not heard directly before. Although all 10-month-olds showed recognition responses to words first heard in isolation, not all infants showed such responses to words they had first heard within an utterance. Those that did succeed in the latter, harder, task, however, understood more words and utterances when re-tested at 12 months, and understood more words and produced more words at 24 months, compared with those who had shown no such recognition response at 10 months. The ability to rapidly recognize the words in continuous utterances is clearly linked to future language development.     

Isolated words enhance statistical language learning in infancy

Infants are adept at tracking statistical regularities to identify word boundaries in pause-free speech. However, researchers have questioned the relevance of statistical learning mechanisms to language acquisition, since previous studies have used simplified artificial languages that ignore the variability of real language input. The experiments reported here embraced a key dimension of variability in infant-directed speech. English-learning infants (8-10 months) listened briefly to natural Italian speech that contained either fluent speech only or a combination of fluent speech and single-word utterances. Listening times revealed successful learning of the statistical properties of target words only when words appeared both in fluent speech and in isolation; brief exposure to fluent speech alone was not sufficient to facilitate detection of the words' statistical properties. This investigation suggests that statistical learning mechanisms actually benefit from variability in utterance length, and provides the first evidence that isolated words and longer utterances act in concert to support infant word segmentation.     

Reenactment: an embodied cognition approach to meaning and linguistic content

A central finding in experimental research identified with embodied cognition (EC) is that understanding actions involves their embodied simulation, i.e., executing some processes involved in performing these actions. Extending these findings, I argue that reenactment??the overt embodied simulation of actions and practices, including especially communicative actions and practices, within utterances??makes it possible to forge an integrated EC-based account of linguistic meaning. In particular, I argue: (a) that remote entities can be referred to by reenacting actions performed with them; (b) that the use of grammatical constructions can be conceived of as the reenactment of linguistic action routines; (c) that complex enunciational structures (reported speech, irony, etc.) involve a separate level of reenactment, on which characters are presented as interacting with one another within the utterance; (d) that the segmentation of long utterances into shorter units involves the reenactment of brief audience interventions between units; and (e) that the overall meaning of an utterance can be stated in reenactment terms. The notion of reenactment provides a conceptual framework for accounting for aspects of language that are usually thought to be outside the reach of EC in an EC framework, thus supporting a view of meaning and linguistic content as thoroughly grounded in action and interaction.     

An auditory cue-depreciation effect.

An experiment is reported in which subjects first heard a list of words and then tried to identify these same words from degraded utterances. Paralleling previous findings in the visual modality, the probability of identifying a given utterance was reduced when the utterance was immediately preceded by other, more degraded, utterances of the same word. A second experiment replicated this "cue-depreciation effect" and in addition found the effect to be weakened, if not eliminated, when the target word was not included in the initial list or when the test was delayed by two days.     

Lexical competition in young children's word learning

In two experiments, 1.5-year-olds were taught novel words whose sound patterns were phonologically similar to familiar words (novel neighbors) or were not (novel nonneighbors). Learning was tested using a picture-fixation task. In both experiments, children learned the novel nonneighbors but not the novel neighbors. In addition, exposure to the novel neighbors impaired recognition performance on familiar neighbors. Finally, children did not spontaneously use phonological differences to infer that a novel word referred to a novel object. Thus, lexical competition--inhibitory interaction among words in speech comprehension--can prevent children from using their full phonological sensitivity in judging words as novel. These results suggest that word learning in young children, as in adults, relies not only on the discrimination and identification of phonetic categories, but also on evaluating the likelihood that an utterance conveys a new word.     

Repairing inappropriately specified utterances: Revision or restart?

In two experiments, we investigated repair of inappropriately specified utterances (e.g.,blue square instead ofdark blue square) to determine whether speakers can simply revise a speech plan or must begin anew. In Experiment 1, speakers produced a prime utterance followed by a target utterance that differed from the prime in only one word, but in Experiment 2, they only planned the prime but produced the target. If speakers must repair by restarting, both situations would involve planning from scratch, with a possible benefit of residual activation. If they can repair by revising, though, only the first situation would involve planning from scratch, and the second situation would involve revision of a plan. The targets had either one word more (addition) or fewer (deletion) than the prime sentences. The restart hypothesis predicted that any cost of addition over deletion should be similar in the two experiments. In contrast, the revision hypothesis predicted an extra cost of addition in Experiment 2, because addition involves retrieval of an extra word. In support of the revision hypothesis, there was no difference between additions and deletions in Experiment 1, but in Experiment 2 additions took longer than deletions. We conclude that speakers can repair utterances by revising a speech plan.     

The identification of metaphor

A number of philosophers, linguists and psychologists have made the dual claim that metaphor is cognitively significant and that metaphorical utterances have a meaning not reducible to literal paraphrase. Such a position requires support from an account of metaphorical meaning that can render metaphors cognitively meaningful without the reduction to literal statement. It therefore requires a theory of meaning that can integrate metaphor within its sematics, yet specify why it is not reducible to literal paraphrase. I introduce the idea of a “second-order meaning”, of which metaphor is but one instance, that is a function on literal-conventional, i.e., first-order meaning, and outline a linguistic framework designed to provide a representation of linguistic meaning for both. This framework is designed to represent linguistic units ranging from a single word to an entire text since I argue that the by-now familiar position that the sentence is the appropriate unit for metaphor has mislead us into asking the wrong questions about metaphorical meaning. With this apparatus, we can specify the conditions under which an utterance may transcend the constraints on first-order meaning (transgressions not always apparent on the sentential level), without thereby being “meaningless”. Conversely, we can specify the conditions that may render apparently odd utterances first-order meaningful rather than metaphorical. In this way we see how metaphorical language differs both from deviant language and from specialized language such as technical language, fanciful and fantastical language (in fairy tales, science fiction, etc.).     

Comprehending illocutionary force

According to speech act theory (Searle, 1969), utterances have both a propositional content and an illocutionary force (the speech act performed with the utterance). Four experiments were conducted to examine whether utterance comprehension involves speech act recognition. Participants in all experiments first read remarks that could be characterized by a particular speech act (e.g., beg). A recognition probe reaction time procedure was used in Experiments 1 and 2; participants indicated whether a probe word had literally appeared in the last remark that they had read. Participants were significantly slower at making this judgment (and made significantly more errors) when the probe represented the speech act performed with the prior remark than when it did not. A lexical decision task was used in Experiments 3 and 4, and participants were significantly faster at verifying target words representing the speech act performed with a remark, relative to control words. Overall, the results suggest that speech act recognition may be an important component of the comprehension of conversational remarks.     

The identification of names in maternal speech to infants

Mothers' speech to 14-month-old infants was recorded during a free-play session. The findings suggest that this speech contained a number of characteristics which could facilitate the identification of names from the other words in an utterance; names were frequently the loudest word of an utterance and were highly likely to be positioned at the end of utterances. Thus, infants appear to receive a verbal input which may help them to identify semantically important words.This research was supported by an S.S.R.C. grant to H. R. Schaffer, Dept. of Psychology, University of Strathclyde, Glasgow, Scotland.     

Adult-child discourse: Developmental interaction between information processing and linguistic knowledge

In the present study, the discourse interaction between adult and child was examined in terms of the content of their utterances, and the linguistic and contextual relations between their messages, in order to investigate how children use the information from adults' input sentences to form contingent responses. The analyses described were based on longitudinal data from four children from approximately 21 to 36 months of age. Categories of child discourse, their development and their interactions with aspects of prior adult utterances form the major results of the study. Child utterances were identified as adjacent (immediately preceded by an adult utterance), or as nonadjacent (not immediately preceded by an adult utterance). Adjacent utterances were either contingent (shared the same topic and added new information relative to the topic of the prior utterance), imitative (shared the same topic but did not add new information), or noncontingent (did not share the same topic). From the beginning, the adjacent speech was greater than nonadjacent speech. Contingent speech increased over time; in particular, linguistically contingent speech (speech that expanded the verb relation of the prior adult utterance with added or replaced constituents within a clause) showed the greatest developmental increase. Linguistically contingent speech occurred more often after questions than nonquestions. The results are discussed in terms of how the differential requirements for processing information in antecedent messages is related to language learning.     

Auditory event related potentials during lexical categorization in the oddball paradigm.

Event related potentials (ERPs) and reaction times (RTs) were recorded from 18 subjects, performing lexical categorization of words and nonwords. Three sets of monosyllable utterances, differentiated in semantic and rhyming attributes, were presented using the "oddball paradigm." ERPs included a sustained negativity which began approximately 100 msec after stimulus onset and peaked at approximately 400 mses (SN4) poststimulus, in response to target as well as to nontarget word stimuli. Semantic effects on SN4 latency were observed only for target utterances. Nonmeaningful word targets were associated with longer SN4 peak latencies as well as slower RTs compared to meaningful word targets. It is suggested that these longer latencies are due to a longer time required for an exhaustive search in permanent memory before categorizing a stimulus as a nonmeaningful word.     

Lexical processing in children and adults during word copying

Copying text may seem trivial, but the task itself is psychologically complex. It involves a series of sequential visual and cognitive processes, which must be co-ordinated; these include visual encoding, mental representation and written production. To investigate the time course of word processing during copying, we recorded eye movements of adults and children as they hand-copied isolated words presented on a classroom board. Longer and lower frequency words extended adults' encoding durations, suggesting whole word encoding. Only children's short word encoding was extended by lower frequency. Though children spent more time encoding long words compared to short words, gaze durations for long words were extended similarly for high- and low-frequency words. This suggested that for long words children used partial word representations and encoded multiple sublexical units rather than single whole words. Piecemeal word representation underpinned copying longer words in children, but reliance on partial word representations was not shown in adult readers.     

Integrating constraints for learning word-referent mappings

Learning word-referent mappings is complex because the word and its referent tend to co-occur with multiple other words and potential referents. Such complexity has led to proposals for a host of constraints on learning, though how these constraints may interact has not yet been investigated in detail. In this paper, we investigated interactions between word co-occurrence constraints and cross-situational statistics in word learning. Analyses of child-directed speech revealed that when both object-referring and non-referring words occurred in the utterance, referring words were more likely to be preceded by a determiner than when the utterance contained only referring words. In a word learning study containing both referring and non-referring words, learning was facilitated when non-referring words contributed grammatical constraints analogous to determiners. The complexity of multi-word utterances provides an opportunity for co-occurrence constraints to contribute to word-referent mapping, and the learning mechanism is able to integrate these multiple sources of information.     

言语产生中音韵编码的计划广度:来自图-词干扰范式的证据

采用图-词干扰范式,要求被试口语产生包含三个名词的句子,同时听觉呈现干扰刺激并操纵其与各个名词的语音相关性,通过与无关刺激相比,与各个名词语音相关的干扰刺激产生的语音效应,探讨句子产生中音韵编码的计划广度,同时检验音韵编码的层级激活假设。结果发现,只有第一个名词上出现了语音促进效应,第二和第三个名词上既无语音促进,也无语音干扰。该结果与层级激活假设相悖,表明句子产生中音韵编码的计划广度为第一个韵律词。     

Infant word segmentation and childhood vocabulary development: a longitudinal analysis

Infants begin to segment novel words from speech by 7.5 months, demonstrating an ability to track, encode and retrieve words in the context of larger units. Although it is presumed that word recognition at this stage is a prerequisite to constructing a vocabulary, the continuity between these stages of development has not yet been empirically demonstrated. The goal of the present study is to investigate whether infant word segmentation skills are indeed related to later lexical development. Two word segmentation tasks, varying in complexity, were administered in infancy and related to childhood outcome measures. Outcome measures consisted of age‐normed productive vocabulary percentiles and a measure of cognitive development. Results demonstrated a strong degree of association between infant word segmentation abilities at 7 months and productive vocabulary size at 24 months. In addition, outcome groups, as defined by median vocabulary size and growth trajectories at 24 months, showed distinct word segmentation abilities as infants. These findings provide the first prospective evidence supporting the predictive validity of infant word segmentation tasks and suggest that they are indeed associated with mature word knowledge. A video abstract of this article can be viewed at http://www.youtube.com/watch?v=jxzLi5oLZQ8 .     

Use of repetition to facilitate spontaneous language acquisition

The role of repetition as a facilitator of spontaneous language acquisition was examined in a 12-month longitudinal study of three children: John, Mindy, and Ashley. Imitation was first defined in a traditional fashion as the exact repetition of a model utterance, within five utterances, without changing the model except to reduce it. During the single-word utterance period, John entered words into his vocabulary by imitating them and then using them spontaneously, while the two girls did not. He also imitated longer utterances than he produced spontaneously while the two girls did not. At the same time, John showed the most rapid language acquisition of the three children. John did not use imitation, as defined, to enter new syntactic-semantic relations into his speech during the two-word stage. However, when the definition of imitation was broadened to include other kinds of repetitions such as repetitions with expansion, either of the child's own productions or of those of others, it was found that repetitions played a significant role in the acquisition of new vocabulary and new syntactic-semantic relations in two-word utterances.This article is based, in part, on a Ph.D. dissertation submitted to the Department of Psychology, University of Denver, in 1976. Portions of the article were presented at the Wisconsin Education Research Association, Oshkosh, Wisconsin, December 1–2, 1977. The research was undertaken while the author was an NIMH predoctoral fellow, 5F01 MH55918-02 MTHL, and was also supported by a Grant Foundation grant for graduate student research to Dr. Marshall Haith and a Spencer Foundation grant to Dr. Kurt Fischer.     

Vocal Features of Conversational Sarcasm: A Comparison of Methods

This study investigated vocal cues that differentiate sarcastic utterances from non-sarcastic utterances. Utterances were drawn from videotapes of participant interviews and arranged on a master tape for analysis. Utterances that were identified as sarcastic by speakers and recognized as sarcastic by listeners were randomly arranged with utterances identified and recognized as non-sarcastic by the same participants. Both sarcastic and non-sarcastic utterances were analyzed by two methods–acoustic analysis and perceptual coding. The acoustic analysis proved slightly more successful than the perceptual coding in discriminating between sarcastic and non-sarcastic utterances. The acoustic analysis indicated that fundamental frequency, frequency range, length of utterance, and total amount of sound significantly discriminated sarcastic from non-sarcastic utterances. The perceptual coding method revealed that pitch range, length of utterance, and total amount of sound significantly discriminated sarcastic from non-sarcastic utterances. Moderate correlations were found between the acoustic and perceptual variables.