Do infants segment words or recurring contiguous patterns?

Eight experiments tested the hypothesis that infants' word segmentation abilities are reducible to familiar sound-pattern parsing regardless of actual word boundaries. This hypothesis was disconfirmed in experiments using the headturn preference procedure: 8.5-month-olds did not mis-segment a consonant-vowel-consonant (CVC) word (e.g., dice) from passages containing the corresponding phonemic pattern across a word boundary (C#VC#; "cold ice"), but they segmented it when the word was really present ("roll dice"). However, they did not segment the real vowel-consonant (VC) word (ice in "cold ice") until 16 months. Yet, at that age, they still did not false alarm on the straddling CVC word. Thus, infants do not simply respond to recurring phonemic patterns. Instead, they are sensitive to both acoustic and allophonic cues to word boundaries. Moreover, there is a sizable developmental gap between consonant- and vowel-initial word segmentation.

Phonotactic cues for segmentation of fluent speech by infants.

There is growing evidence that infants become sensitive to the probabilistic phonotactics of their ambient language sometime during the second half of their first year. The present study investigates whether 9-month-olds make use of phonotactic cues to segment words from fluent speech. Using the Headturn Preference Procedure, we found that infants listened to a CVC stimulus longer when the stimulus previously appeared in a sentential context with good phonotactic cues than when it appeared in one without such cues. The goodness of the phonotactic cues was estimated from the frequency with which the C.C clusters at the onset and offset of a CVC test stimulus (i.e. C.CVC.C) are found within and between words in child-directed speech, with high between-word probability associated with good cues to word boundaries. A similar segmentation result emerged when good phonotactic cues occurred only at the onset (i.e. C.CVC.C) or the offset (i.e. C.CVC.C) of the target words in the utterances. Together, the results suggest that 9-month-olds use probabilistic phonotactics to segment speech into words and that high-probability between-word clusters are interpreted as both word onsets and word offsets.

Cross-language word segmentation by 9-month-olds.

Dutch-learning and English-learning 9-month-olds were tested, using the Headturn Preference Procedure, for their ability to segment Dutch words with strong/weak stress patterns from fluent Dutch speech. This prosodic pattern is highly typical for words of both languages. The infants were familiarized with pairs of words and then tested on four passages, two that included the familiarized words and two that did not. Both the Dutch- and the English-learning infants gave evidence of segmenting the targets from the passages, to an equivalent degree. Thus, English-learning infants are able to extract words from fluent speech in a language that is phonetically different from English. We discuss the possibility that this cross-language segmentation ability is aided by the similarity of the typical rhythmic structure of Dutch and English words.

The role of talker-specific information in word segmentation by infants.

Infants' representations of the sound patterns of words were explored by examining the effects of talker variability on the recognition of words in fluent speech. Infants were familiarized with isolated words (e.g., cup and dog) from 1 talker and then heard 4 passages produced by another talker, 2 of which included the familiarized words. At 7.5 months of age, infants attended longer to passages with the familiar words for materials produced by 2 female talkers or 2 male talkers but not for materials by a male and a female talker. These findings suggest a strong role for talker-voice similarity in infants' ability to generalize word tokens. By 10.5 months, infants could generalize different instances of the same word across talkers of the opposite sex. One implication of the present results is that infants' initial representations of the sound structure of words not only include phonetic information but also indexical properties relating to the vocal characteristics of particular talkers.

Infants' sensitivity to allophonic cues for word segmentation.

A series of four experiments was conducted to determine whether English-learning infants can use allophonic cues to word boundaries to segment words from fluent speech. Infants were familiarized with a pair of two-syllable items, such as nitrates and night rates and then were tested on their ability to detect these same words in fluent speech passages. The presence of allophonic cues to word boundaries did not help 9-month-olds to distinguish one of the familiarized words from an acoustically similar foil. Infants familiarized with nitrates were just as likely to listen to a passage about night rates as they were to listen to one about nitrates. Nevertheless, when the passages contained distributional cues that favored the extraction of the familiarized targets, 9-month-olds were able to segment these items from fluent speech. By the age of 10.5 months, infants were able to rely solely on allophonic cues to locate the familiarized target words in passages. We consider what implications these findings have for understanding how word segmentation skills develop.

Synchronous change and perception of object unity: evidence from adults and infants.

Adults and infants display a robust ability to perceive the unity of a center-occluded object when the visible ends of the object undergo common motion (e.g. Kellman, P.J., Spelke, E.S., 1983. Perception of partly occluded objects in infancy. Cognitive Psychology 15, 483-524). Ecologically oriented accounts of this ability focus on the primary of motion in the perception of segregated objects, but Gestalt theory suggests a broader possibility: observers may perceive object unity by detecting patterns of synchronous change, of which common motion is a special case. We investigated this possibility with observations of adults and 4-month-old infants. Participants viewed a center-occluded object whose visible surfaces were either misaligned or aligned, stationary or moving, and unchanging or synchronously changing in color or brightness in various temporal patterns (e.g. flashing). Both alignment and common motion contributed to adults' perception of object unity, but synchronous color changes did not. For infants, motion was an important determinant of object unity, but other synchronous changes and edge alignment were not. When a stationary object with aligned edges underwent synchronous changes in color or brightness, infants showed high levels of attention to the object, but their perception of its unity appeared to be indeterminate. An inherent preference for fast over slow flash rates, and a novelty preference elicited by a change in rate, both indicated that infants detected the synchronous changes, although they failed to use them as information for object unity. These findings favor ecologically oriented accounts of object perception in which surface motion plays a privileged role.

Narrowing the distance to language: one step at a time.

Infants' earliest attempts at word segmentation appear to be guided by a single source of information (e.g., English-learners initially rely on the predominant stress pattern of words). This initial strategy successfully identifies many potential words in the input, but mis-segments others. However, simply breaking the input into smaller chunks helps learners to identify other possible cues to the location of word boundaries in utterances. Because no one source of information is completely reliable, listeners must eventually rely on multiple cues to segment words. The development of such skills is not critical for developing a native language vocabulary, but also for acquiring the grammatical organization of utterances. Tracking familiar sound patterns, such as function words and grammatical morphemes, may help in learning about syntactic organization. One factor that facilitates learning about the distribution of such elements is sensitivity to boundaries of prosodic phrases. Access to such linguistically-relevant chunks also helps in tracking the distribution of words in the input.

Phonotactic and prosodic effects on word segmentation in infants.

This research examines the issue of speech segmentation in 9-month-old infants. Two cues known to carry probabilistic information about word boundaries were investigated: Phonotactic regularity and prosodic pattern. The stimuli used in four head turn preference experiments were bisyllabic CVC.CVC nonwords bearing primary stress in either the first or the second syllable (strong/weak vs. weak/strong). Stimuli also differed with respect to the phonotactic nature of their cross-syllabic C.C cluster. Clusters had either a low probability of occurring at a word juncture in fluent speech and a high probability of occurring inside of words ("within-word" clusters) or a high probability of occurring at a word juncture and a low probability of occurring inside of words ("between-word" clusters). Our results show that (1) 9-month-olds are sensitive to how phonotactic sequences typically align with word boundaries, (2) altering the stress pattern of the stimuli reverses infants' preference for phonotactic cluster types, (3) the prosodic cue to segmentation is more strongly relied upon than the phonotactic cue, and (4) a preference for high-probability between-word phonotactic sequences can be obtained either by placing stress on the second syllable of the stimuli or by inserting a pause between syllables. The implications of these results are discussed in light of an integrated multiple-cue approach to speech segmentation in infancy.

The beginnings of word segmentation in english-learning infants.

A series of 15 experiments was conducted to explore English-learning infants' capacities to segment bisyllabic words from fluent speech. The studies in Part I focused on 7.5 month olds' abilities to segment words with strong/weak stress patterns from fluent speech. The infants demonstrated an ability to detect strong/weak target words in sentential contexts. Moreover, the findings indicated that the infants were responding to the whole words and not to just their strong syllables. In Part II, a parallel series of studies was conducted examining 7.5 month olds' abilities to segment words with weak/strong stress patterns. In contrast with the results for strong/weak words, 7.5 month olds appeared to missegment weak/strong words. They demonstrated a tendency to treat strong syllables as markers of word onsets. In addition, when weak/strong words co-occurred with a particular following weak syllable (e.g., "guitar is"), 7.5 month olds appeared to misperceive these as strong/weak words (e.g., "taris"). The studies in Part III examined the abilities of 10.5 month olds to segment weak/strong words from fluent speech. These older infants were able to segment weak/strong words correctly from the various contexts in which they appeared. Overall, the findings suggest that English learners may rely heavily on stress cues when they begin to segment words from fluent speech. However, within a few months time, infants learn to integrate multiple sources of information about the likely boundaries of words in fluent speech.

Sensitivity to discontinuous dependencies in language learners: evidence for limitations in processing space.

Five experiments using the Headturn Preference Procedure examined 15- and 18-month-old children's sensitivity to morphosyntactic dependencies in English. In each experiment, the children were exposed to two types of passages. Passages in the experimental condition contained a well-formed English dependency between the auxiliary verb is and a main verb with the ending -ing. Passages in the control condition contained an ungrammatical combination of the modal auxiliary can and a main verb with the ending -ing. In the experiments, the distance between the dependent morphemes was systematically varied by inserting an adverbial of a specified length between the auxiliary and main verbs. The results indicated that 18-month-olds are sensitive to the basic relationship between is and -ing, but that 15-month-olds are not. The 18-month-olds, but not the 15-month-olds, listened significantly longer to the passages with the well-formed English dependency. In addition, the 18-month-olds showed this preference for the well-formed dependency only over a limited domain of 1-3 syllables. Over domains of 4-5 syllables, they showed no significant preference for the experimental over the control passages. These findings indicate that 18-month-olds can track relationships between functor morphemes. Additionally, these findings are consistent with the hypothesis that 18-month-olds are working with a limited processing window, and that they are only picking up relevant dependencies that fall within this window.

Infants' memory for spoken words.

Infants' long-term retention of the sound patterns of words was explored by exposing them to recordings of three children's stories for 10 days during a 2-week period when they were 8 months old. After an interval of 2 weeks, the infants heard lists of words that either occurred frequently or did not occur in the stories. The infants listened significantly longer to the lists of story words. By comparison, a control group of infants who had not been exposed to the stories showed no such preference. The findings suggest that 8-month-olds are beginning to engage in long-term storage of words that occur frequently in speech, which is an important prerequisite for learning language.

The cocktail party effect in infants.

Most speech research with infants occurs in quiet laboratory rooms with no outside distractions. However, in the real world, speech directed to infants often occurs in the presence of other competing acoustic signals. To learn language, infants need to attend to their caregiver's speech even under less than ideal listening conditions. We examined 7.5-month-old infants' abilities to selectively attend to a female talker's voice when a male voice was talking simultaneously. In three experiments, infants heard a target voice repeating isolated words while a distractor voice spoke fluently at one of three different intensities. Subsequently, infants heard passages produced by the target voice containing either the familiar words or novel words. Infants listened longer to the familiar words when the target voice was 10 dB or 5 dB more intense than the distractor, but not when the two voices were equally intense. In a fourth experiment, the assignment of words and passages to the familiarization and testing phases was reversed so that the passages and distractors were presented simultaneously during familiarization, and the infants were tested on the familiar and unfamiliar isolated words. During familiarization, the passages were 10 dB more intense than the distractors. The results suggest that this may be at the limits of what infants at this age can do in separating two different streams of speech. In conclusion, infants have some capacity to extract information from speech even in the face of a competing acoustic voice.

Infants' sensitivity to word boundaries in fluent speech.

Infants' sensitivity to word units in fluent speech was examined by inserting I sec pauses either at boundaries between successive words (Coincident versions) or between syllables within words (Noncoincident versions). In Experiment 1, 24 11-month-olds listened significantly longer to the Coincident versions. In Experiment 2, 24 four-and-a-half- and 24 nine-month-olds did not exhibit the preference for the Coincident versions that the 11-month-olds showed. When the stimuli were low-pass filtered in Experiment 3, 24 11-month-olds showed no preference for the Coincident versions, suggesting they rely on more than prosodic cues. New stimulus materials in Experiment 4 indicated that responses by 24 11-month-olds to the Coincident and Noncoindent versions did not depend solely on prior familiarity with the targets. Two groups of 30 11-month-olds tested in Experiment 5 were as sensitive to groups of 30 11-month-olds tested in Experiment 5 were as sensitive to boundaries for Strong/Weak words as for Weak/Strong words. Taken together, the results suggest that, by 11 months, infants are sensitive to word boundaries in fluent speech, and that this sensitivity depends on more than just prosodic information or prior knowledge of the words.

Infants' detection of the sound patterns of words in fluent speech.

A series of four experiments examined infants' capacities to detect repeated words in fluent speech. In Experiment 1, 7 1/2-month old American infants were familiarized with two different monosyllabic words and subsequently were presented with passages which either included or did not include the familiar target words embedded in sentences. The infants listened significantly longer to the passages containing the familiar target words than to passages containing unfamiliar words. A comparable experiment with 6-month-olds provided no indication that infants at this age detected the target words in the passages. In Experiment 3, a group of 7 1/2-month-olds was familiarized with two different non-word targets which differed in their initial phonetic segment by only one or two phonetic features from words presented in two of the passages. These infants showed no tendency to listen significantly longer to the passages with the similar sounding words, suggesting that the infants may be matching rather detailed information about the items in the familiarization period to words in the test passages. Finally, Experiment 4 demonstrated that even when the 7 1/2-month-olds were initially familiarized with target words in sentential contexts rather than in isolation, they still showed reliable evidence of recognizing these words during the test phase. Taken together, the results of these studies suggest that some ability to detect words in fluent speech contexts is present by 7 1/2 months of age.

Young infants' retention of information about bisyllabic utterances.

This study examined 2-to 3 month-olds' representations of bisyllables. In 3 experiments, infants were familiarized with sets of bisyllables that either did or did not share a common consonant-vowel (CV) syllable. In Experiment 1, infants detected the presence of a new bisyllable in the test phase except when it shared a common initial CV syllable. a modified version of the high-amplitude sucking (HAS) procedure, incorporating a 2-min delay period, tested infants' retention of information about bisyllables in the remaining 2 experiments. In Experiment 2, infants were significantly more likely to retain information about bisyllables that shared the same initial CV syllable. Finally, the authors investigated whether infants simply benefited from the presence of 2 common phonetic segments, regardless of whether these came from the same cv syllable. The results showed that CV syllable organization is important in infants' ability to encode and retain information about bisyllables.

Do English-learning infants use syllable weight to determine stress?

A linguistic factor governing the assignment of English lexical stress is syllable weight. Heavy syllables which have either a long (tense) vowel or are closed with a consonant are heavy and automatically bear stress. Are infants sensitive to this aspect of the English stress system? Previous research by Jusczyk, Cutler, and Redanz (1993) showed that nine-month-olds listened longer to words exhibiting Strong-Weak than Weak-Strong stress pattern. However, they did not investigate the role of syllable weight in this preference. A series of three experiments explored infants' preference for Strong-Weak versus Weak-Strong lists, but systematically manipulated the syllable weight of Strong syllables. The results suggest that syllable weight is not a necessary component of the Strong-Weak preference observed in previous studies. Rather it appears that infants prefer both words that begin with a Strong syllable and Strong syllables that are heavy. Thus, the results suggest that sensitivity to surface linguistic patterns and the principles that underlie them may be independent in early language acquisition.

Two-month-old infants' sensitivity to allophonic differences.

The present study investigated 2-month-olds' abilities to discriminate allophonic differences that are potentially useful in segmenting fluent speech. Experiment 1 investigated infants' sensitivity to the kind of distinction that may signal the presence or absence of a word boundary. When tested with the high-amplitude sucking procedure, infants discriminated pairs of items, such as "nitrate" versus "night rate" and "nikrate" versus "nike rate". By greatly reducing the potential contribution of prosodic differences to these contrasts, Experiment 2 evaluated whether the allophonic differences for /t/ and /r/ were sufficient for infants to distinguish the "nitrate" versus "night rate" pair. Infants distinguished "nitrate" from a cross-spliced version of "night rate," which differed only in the allophones for /t/ and /r/ that it included. Thus, infants appear to possess one of the prerequisite capacities (i.e., the ability to discriminate allophonic distinctions) necessary to use allophonic information in segmenting fluent speech.

Does sentential prosody help infants organize and remember speech information?

Theories that propose a mapping between prosodic and syntactic structures require that prosodic units in fluent speech be perceptually salient for infants. Although previous studies have demonstrated that infants are sensitive to prosodic markers of syntactic units, they do not show that prosodic information really has an impact on how infants encode the speech they hear. Two experiments were conducted to examine whether infants as young as 2 months old might actually use the prosody afforded by sentences to organize and remember spoken information. The results suggest that infants better remember the phonetic properties of (1) words that are prosodically linked together within a single clause as opposed to individual items in a list (Experiment 1); and (2) words that are prosodically linked within a single clausal unit as opposed to spanning two contiguous fragments (Experiment 2). Taken together, the evidence from both experiments suggests that the prosodic organization of speech into clausal units enhances infants' memory for spoken information. These findings are discussed with regard to their implications for theories of language acquisition.

When prosody fails to cue syntactic structure: 9-month-olds' sensitivity to phonological versus syntactic phrases.

According to prosodic bootstrapping accounts of syntax acquisition, language learners use the correlation between syntactic boundaries and prosodic changes (e.g., pausing, vowel lengthening, large increases or decreases in fundamental frequency) to cue the presence and arrangement of syntactic constituents. However, recent linguistic accounts suggest that prosody does not directly reflect syntactic structure but rather is governed by independent prosodic units such as phonological phrases. To examine the implications of this view for the prosodic bootstrapping hypothesis, infants in Experiment 1 were presented with sentences in which pauses were inserted either between the subject noun phrase (NP) and verb or after the verb. Half of the infants heard sentences with lexical NP subjects, in which prosodic structure is consistent with syntactic structure. The other half heard sentences with pronoun subjects, in which prosodic structure does not mirror syntactic structure. In a preferential listening paradigm, infants in the lexical NP condition listened longer to materials containing pauses between the subject and verb, the main syntactic constituents. However, in the pronoun NP condition, infants showed no difference in listening times for the two pause locations. To determine if other sentence types containing pronoun subjects potentially provide information about the syntactic constituency of these elements, infants in Experiment 2 heard yes-no questions with pronoun subjects, in which the prosodic structure reflects the constituency of the subject. Infants listened longer when pauses were inserted between the subject and verb than after the verb. Taken together, our results suggest that the prosodic information in an individual sentence is not always sufficient to assign a syntactic structure. Rather, learners must engage in active inferential processes, using cross-sentence comparisons and other types of information to arrive at the correct syntactic representation.

Pitch and rhythmic patterns affecting infants' sensitivity to musical phrase structure.

The present studies were undertaken to learn more about the nature of the cues that underlie infants' perception of musical phrase structure. Experiment 1 demonstrated that infants in Krumhansl and Jusczyk's (1990) study were responding to the phrase structure instead of to the beginnings and endings of Mozart minuet stimuli. Experiment 2 showed that infants treat musical passages with pauses inserted at phrase boundaries much as they do unaltered versions of the same passages. Experiments 3 and 4 indicated that the direction of change in pitch height and tone duration is critical to obtaining longer orientation times to musical passages that are segmented at phrase boundaries. Finally, Experiment 5 demonstrates that different effects found for the forward and reversed versions of the passages with inserted pauses are not the result of an intrinsic preference for the forward versions.