No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Disfluencies signal reference to novel objects for adults but not children
Abstract
Speech disfluencies can guide the ways in which listeners interpret spoken language. Here, we examined whether three-year-olds, five-year-olds, and adults use filled pauses to anticipate that a speaker is likely to refer to a novel object. Across three experiments, participants were presented with pairs of novel and familiar objects and heard a speaker refer to one of the objects using a fluent (“ Look at the ball/lep! ”) or disfluent (“ Look at thee uh ball/lep! ”) expression. The salience of the speaker's unfamiliarity with the novel referents, and the way in which the speaker referred to the novel referents (i.e., a noun vs. a description) varied across experiments. Three- and five-year-olds successfully identified familiar and novel targets, but only adults’ looking patterns reflected increased looks to novel objects in the presence of a disfluency. Together, these findings demonstrate that adults, but not young children, use filled pauses to anticipate reference to novel objects.
... Additionally, three-year-olds can use disfluencies to predict that a speaker will label an object that is perceptually familiar but novel to the discourse (e.g., a familiar object that has not previously been mentioned; Owens & Graham, 2016). However, while adults can use filled pauses to anticipate reference solely based on object novelty, three-and five-year-olds require that an object be new in the discourse to anticipate reference to a novel object (Owens, Thacker, & Graham, 2018). Together, findings from real-time comprehension studies with monolingual children suggest that their ability to use speech disfluencies predictively emerges around their second birthday, and that they flexibly adapt their predictions based on both speaker and context. ...
... The sequence of each trial followed Kidd et al. (2011), such that novel labels were also discourse-new (see Graham, 2016, andOwens et al., 2018, for evidence that discourse novelty is a driver of disfluency effects in children). There were three presentations of each object label. ...
... Our study provides an important contribution to the literature on children's predictive use of disfluencies, which had previously been limited to studies of monolingual English-learning children (Kidd et al., 2011;Orena & White, 2015;Owens & Graham, 2016;Owens et al., 2018;Thacker, Chambers, & Graham, 2018). We replicated these findings with English monolinguals, and extended them to French monolinguals and French-English bilinguals. ...
Previous research suggests that English monolingual children and adults can use speech disfluencies (e.g., uh) to predict that a speaker will name a novel object. To understand the origins of this ability, we tested 48 32-month-old children (monolingual English, monolingual French, bilingual English–French; Study 1) and 16 adults (bilingual English–French; Study 2). Our design leveraged the distinct realizations of English (uh) versus French (euh) disfluencies. In a preferential-looking paradigm, participants saw familiar–novel object pairs (e.g., doll–rel), labeled in either Fluent (“Look at the doll/rel!”), Disfluent Language-consistent (“Look at thee uh doll/rel!”), or Disfluent Language-inconsistent (“Look at thee euh doll/rel!”) sentences. All participants looked more at the novel object when hearing disfluencies, irrespective of their phonetic realization. These results suggest that listeners from different language backgrounds harness disfluencies to comprehend day-to-day speech, possibly by attending to their lengthening as a signal of speaker uncertainty. Stimuli and data are available at <https://osf.io/qn6px/>.
... As the discourse-new object was also always the novel object in the Kidd et al. (2011) paradigm, a recent series of studies sought to disentangle these factors. These findings demonstrated that 2-and 3-year-old children readily associate filled pauses with upcoming reference to discourse-new (Owens and Graham, 2016) objects but not unfamiliar objects (Owens et al., 2017). This contrasts with studies showing that adults show referential anticipation of novel objects upon hearing a filled pause (Arnold et al., 2007;Owens et al., 2017). ...
... These findings demonstrated that 2-and 3-year-old children readily associate filled pauses with upcoming reference to discourse-new (Owens and Graham, 2016) objects but not unfamiliar objects (Owens et al., 2017). This contrasts with studies showing that adults show referential anticipation of novel objects upon hearing a filled pause (Arnold et al., 2007;Owens et al., 2017). Recently, research has highlighted preschoolers' ability to amend an initial prediction (e.g., that talkers are more likely to refer to a preferred familiar object) when encountering a disfluency (Thacker et al., 2018). ...
... This result in turn extends our understanding of the processing consequences associated with disfluent descriptions. As mentioned earlier, previous findings have shown that adults can use the disfluency-novelty link to infer referential intent, such that they will fixate a novel referent over a familiar referent upon hearing a filled pause (e.g., Owens et al., 2017). Yet both objects in the present study were equally unfamiliar to adult (and child) listeners. ...
An eye-tracking methodology was used to explore adults’ and children’s use of two utterance-based cues to overcome referential uncertainty in real time. Participants were first introduced to two characters with distinct color preferences. These characters then produced fluent (“Look! Look at the blicket.”) or disfluent (“Look! Look at thee, uh, blicket.”) instructions referring to novel objects in a display containing both talker-preferred and talker-dispreferred colored items. Adults (Expt 1, n = 24) directed a greater proportion of looks to talker-preferred objects during the initial portion of the utterance (“Look! Look at…”), reflecting the use of indexical cues for talker identity. However, they immediately reduced consideration of an object bearing the talker’s preferred color when the talker was disfluent, suggesting they infer disfluency would be more likely as a talker describes dispreferred objects. Like adults, 5-year-olds (Expt 2, n = 27) directed more attention to talker-preferred objects during the initial portion of the utterance. Children’s initial predictions, however, were not modulated when disfluency was encountered. Together, these results demonstrate that adults, but not 5-year-olds, can act on information from two talker-produced cues within an utterance, talker preference, and speech disfluencies, to establish reference.
... Like adults, they also modulate their predictive use of disfluencies based on a speaker's characteristics or their pattern of disfluency use (Orena & White, 2015;Thacker, Chambers, & Graham, 2018; but see Thacker, Chambers, & Graham, 2018b for limitations on this ability). However, the types of implicit predictions children make during disfluency may be narrower than for adults, in that they are less likely than adults to anticipate a novel target on the basis of a disfluency (Owens, Thacker, & Graham, 2018). Although disfluencies have been shown to affect children's predictions about upcoming referents, there is no research yet addressing the effects of disfluencies on children's word learning. ...
... However, they may not predict the same range of possible referents that adults do. In particular, they do not appear to predict that a less familiar referent will follow a disfluency (Owens et al., 2018). Therefore, children in our task, presented with familiar and novel objects during training, may not have selectively predicted the novel objects during disfluency, attenuating any potential learning advantage for those items. ...
Disfluencies, such as ‘um’ or ‘uh’, can cause adults to attribute uncertainty to speakers, but may also facilitate speech processing. To understand how these different functions affect children’s learning, we asked whether (dis)fluency affects children’s decision to select information from speakers (an explicit behavior) and their learning of specific words (an implicit behavior). In Experiment 1a, 31 3- to 4-year-olds heard two puppets provide fluent or disfluent descriptions of familiar objects. Each puppet then labeled a different novel object with the same novel word (again, fluently or disfluently). Children more frequently endorsed the object referred to by the fluent speaker. We replicated this finding with a separate group of 4-year-olds in Experiment 1b ( N = 31) and a modified design. In Experiment 2, 62 3- to 4-year-olds were trained on new words, produced following a disfluency or not, and were subsequently tested on their recognition of the words. Children were equally accurate for the two types of words. These results suggest that while children may prefer information from fluent speakers, they learn words equally well regardless of fluency, at least in some contexts.
... Therefore, speech signals convey information about talkers that is often necessary for hearers to fully understand speech. For example, speech signals carry information about talkers' knowledge or confidence (Jiang and Pell, 2017;Pon-Barry and Shieber, 2011), stress level, anxiety, or cognitive load (Chen et al., 2016;Cook, 1969), emotion (Cummins et al., 2015), judgment of current mutual knowledge shared with the listener (Brennan and Williams, 1995;Owens et al., 2018;Smith and Clark, 1993) or intention to introduce new information (Arnold et al., 2003;Barr and Seyfeddinipur, 2010). Because these sources of information cannot easily be represented as units, when speech materials are constructed from discrete units, these sources of information are filtered out. ...
Natural, conversational speech signals contain sources of symbolic and iconic information, both of which are necessary for the full understanding of speech. But speech intelligibility tests, which are generally derived from written language, present only symbolic information sources, including lexical semantics and syntactic structures. Speech intelligibility tests exclude almost all sources of information about talkers, including their communicative intentions and their cognitive states and processes. There is no reason to suspect that either hearing impairment or noise selectively affect perception of only symbolic information. We must therefore conclude that diagnosis of good or poor speech intelligibility on the basis of standard speech tests is based on measurement of only a fraction of the task of speech perception. This paper presents a descriptive comparison of information sources present in three widely used speech intelligibility tests and spontaneous, conversational speech elicited using a referential communication task. The aim of this comparison is to draw attention to the differences in not just the signals, but the tasks of listeners perceiving these different speech signals and to highlight the implications of these differences for the interpretation and generalizability of speech intelligibility test results.
... .") to refer to a novel and discourse-new object as opposed to one that was familiar and discourse-old (Kidd et al., 2011;Morin-Lessard and Byers-Heinlein, 2019;Yoon and Fisher, 2020). This disfluency effect held when the two objects in view differed only in discourse status (both were familiar; Owens and Graham, 2016), but not when they differed only in novelty (neither had been previously named; Owens et al., 2018). Moreover, children's predictive use of disfluencies is not limited to a simple association between disfluency and particular types of referents (e.g., discourse-new referents). ...
Speech disfluencies (e.g., “Point to thee um turtle”) can signal that a speaker is about to refer to something difficult to name. In two experiments, we found evidence that 4-year-olds, like adults, flexibly interpret a particular partner’s disfluency based on their estimate of that partner’s knowledge, derived from the preceding conversation. In entrainment trials, children established partner-specific shared knowledge of names for tangram pictures with one or two adult interlocutors. In each test trial, an adult named one of two visible tangrams either fluently or disfluently while children’s eye-movements were monitored. We manipulated speaker knowledge in the test trials. In Experiment 1, the test-trial speaker was the same speaker from entrainment or a naïve experimenter; in Experiment 2, the test-trial speaker had been one of the child’s partners in entrainment and had seen half of the tangrams (either animal or vehicle tangrams). When hearing disfluent expressions, children looked more at a tangram that was unfamiliar from the speaker’s perspective; this systematic disfluency effect disappeared in Experiment 1 when the speaker was entirely naïve, and depended on each speaker’s entrainment experience in Experiment 2. These findings show that 4-year-olds can keep track of two different partners’ knowledge states, and use this information to determine what should be difficult for a particular partner to name, doing so efficiently enough to guide online interpretation of disfluent speech.
... Recordings of participants' eye movements showed that listeners already anticipated reference to unknown objects (increase in proportion of fixations to unknown objects) when hearing the filler uh in the disfluent condition (i.e., well before hearing the target). Other studies have since shown the same effect in children as young as 2 years of age (Kidd, White, & Aslin, 2011;Orena & White, 2015;Owens & Graham, 2016;Owens, Thacker, & Graham, 2018;Thacker, Chambers, & Graham, 2018a, 2018b. Adult listeners have also been shown to be able to predict other types of complex referents, such as discourse-new (Arnold, Fagnano, & Tanenhaus, 2003;Arnold, Tanenhaus, Altmann, & Fagnano, 2004;Barr & Seyfeddinipur, 2010), compound (Watanabe, Hirose, Den, & Minematsu, 2008), and low-frequency referents (Bosker, Quené, Sanders, & De Jong, 2014a) upon hearing a disfluent filler uh. ...
Disfluencies, like 'uh', have been shown to help listeners anticipate reference to low-frequency words. The associative account of this 'disfluency bias' proposes that listeners learn to associate disfluency with low-frequency referents based on prior exposure to non-arbitrary disfluency distributions (i.e., greater probability of low-frequency words after disfluencies). However, there is limited evidence for listeners actually tracking disfluency distributions online. The present experiments are the first to show that adult listeners, exposed to a typical or more atypical disfluency distribution (i.e., hearing a talker unexpectedly say uh before high-frequency words), flexibly adjust their predictive strategies to the disfluency distribution at hand (e.g., learn to predict high-frequency referents after disfluency). However, when listeners were presented with the same atypical disfluency distribution but produced by a non-native speaker, no adjustment was observed. This suggests pragmatic inferences can modulate distributional learning, revealing the flexibility of, and constraints on, distributional learning in incremental language comprehension.
Listeners utilize speech disfluencies to anticipate novelty in spoken language, but little is known about how speech disfluencies impact novel word learning. We investigated how monolingual and bilingual adults learn novel words under disfluent speech conditions, focusing on fillers such as uh and um. If fillers highlight novelty, they might be an especially potent cue during word learning; however, because fillers also signal uncertainty, listeners may be less willing to learn in a disfluent condition. We also tested whether an effect of fillers on word learning would be moderated by bilingual experience, expecting that bilinguals would be affected differently because their exposure to distributional information within each language is reduced relative to monolinguals. In Experiments 1 and 2, where participants were exposed only to novel words, we found that participants learned words equally well in fluent and disfluent conditions, and that this effect was not moderated by bilingual experience. In Experiment 3, when novel words were embedded within a larger set of known words, we observed a bilingualism by condition interaction, wherein bilinguals benefited from fluency, but monolinguals performed equally well across conditions. These findings suggest that monolinguals' word learning-unlike word processing-may be robust to variations in speaker fluency, but that language experience may moderate the effect of fluency on learning.
Inference using significance testing and Bayes factors is compared and contrasted in five case studies based on real research. The first study illustrates that the methods will often agree, both in motivating researchers to conclude that H1 is supported better than H0, and the other way round, that H0 is better supported than H1. The next four, however, show that the methods will also often disagree. In these cases, the aim of the paper will be to motivate the sensible evidential conclusion, and then see which approach matches those intuitions. Specifically, it is shown that a high-powered non-significant result is consistent with no evidence for H0 over H1 worth mentioning, which a Bayes factor can show, and, conversely, that a low-powered non-significant result is consistent with substantial evidence for H0 over H1, again indicated by Bayesian analyses. The fourth study illustrates that a high-powered significant result may not amount to any evidence for H1 over H0, matching the Bayesian conclusion. Finally, the fifth study illustrates that different theories can be evidentially supported to different degrees by the same data; a fact that P-values cannot reflect but Bayes factors can. It is argued that appropriate conclusions match the Bayesian inferences, but not those based on significance testing, where they disagree.
No scientific conclusion follows automatically from a statistically non-significant result, yet people routinely use non-significant results to guide conclusions about the status of theories (or the effectiveness of practices). To know whether a non-significant result counts against a theory, or if it just indicates data insensitivity, researchers must use one of: power, intervals (such as confidence or credibility intervals), or else an indicator of the relative evidence for one theory over another, such as a Bayes factor. I argue Bayes factors allow theory to be linked to data in a way that overcomes the weaknesses of the other approaches. Specifically, Bayes factors use the data themselves to determine their sensitivity in distinguishing theories (unlike power), and they make use of those aspects of a theory’s predictions that are often easiest to specify (unlike power and intervals, which require specifying the minimal interesting value in order to address theory). Bayes factors provide a coherent approach to determining whether non-significant results support a null hypothesis over a theory, or whether the data are just insensitive. They allow accepting and rejecting the null hypothesis to be put on an equal footing. Concrete examples are provided to indicate the range of application of a simple online Bayes calculator, which reveal both the strengths and weaknesses of Bayes factors.
It is generally accepted that filled pauses ("uh," "er," and "um") indicate time out while the speaker searches for the next word or phrase. It is hypothesized that the more options, the more likely that a speaker will say "uh." The academic disciplines differ in the extent to which their subject matter and mode of thought require a speaker to choose among options. The more formal, structured, and factual the discipline, the fewer the options. It follows that lecturers in the humanities should use more filled pauses during lectures than social scientists and that natural scientists should use fewest of all. Observations of lecturers in 10 academic disciplines indicate that this is the case. That this is due to subject matter rather than to self-selection into disciplines is suggested by observations of this same set of lecturers all speaking on a common subject. In this circumstance, the academic disciplines are identical in the number of filled pauses used. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
The arcsine square root transformation has long been standard procedure when analyzing proportional data in ecology, with applications in data sets containing binomial and non-binomial response variables. Here, we argue that the arcsine transform should not be used in either circumstance. For binomial data, logistic regression has greater interpretability and higher power than analyses of transformed data. However, it is important to check the data for additional unexplained variation, i.e., overdispersion, and to account for it via the inclusion of random effects in the model if found. For non-binomial data, the arcsine transform is undesirable on the grounds of interpretability, and because it can produce nonsensical predictions. The logit transformation is proposed as an alternative approach to address these issues. Examples are presented in both cases to illustrate these advantages, comparing various methods of analyzing proportions including untransformed, arcsine- and logit-transformed linear models and logistic regression (with or without random effects). Simulations demonstrate that logistic regression usually provides a gain in power over other methods.
The ability to infer the referential intentions of speakers is a crucial part of learning a language. Previous research has uncovered various contextual and social cues that children may use to do this. Here we provide the first evidence that children also use speech disfluencies to infer speaker intention. Disfluencies (e.g. filled pauses 'uh' and 'um') occur in predictable locations, such as before infrequent or discourse-new words. We conducted an eye-tracking study to investigate whether young children can make use of this distributional information in order to predict a speaker's intended referent. Our results reveal that young children (ages 2;4 to 2;8) reliably attend to speech disfluencies early in lexical development and are able to use disfluencies in online comprehension to infer speaker intention in advance of object labeling. Our results from two groups of younger children (ages 1;8 to 2;2 and 1;4 to 1;8) suggest that this ability emerges around age 2.
A picture naming experiment in Dutch tested whether disfluencies in speech can arise from difficulties in lexical access. Speakers described networks consisting of line drawings and paths connecting these drawings, and we manipulated picture name agreement. Consistent with our hypothesis, there were more pauses and more self-corrections in the low name agreement condition than the high name agreement condition, but there was no effect on repetitions. We also considered determiner frequency. There were more self-corrections and more repetitions when the picture name required the less frequent (neuter-gender) determiner “het” than the more frequent (common-gender) determiner “de”. These data suggest that difficulties in distinct stages of language production result in distinct patterns of disfluencies.
Speakers often repeat the first word of major constituents, as in, "I uh I wouldn't be surprised at that." Repeats like this divide into four stages: an initial commitment to the constituent (with "I"); the suspension of speech; a hiatus in speaking (filled with "uh"); and a restart of the constituent ("I wouldn't."). An analysis of all repeated articles and pronouns in two large corpora of spontaneous speech shows that the four stages reflect different principles. Speakers are more likely to make a premature commitment, immediately suspending their speech, as both the local constituent and the constituent containing it become more complex. They plan some of these suspensions from the start as preliminary commitments to what they are about to say. And they are more likely to restart a constituent the more their stopping has disrupted its delivery. We argue that the principles governing these stages are general and not specific to repeats.
After reviewing situational and demographic factors that have been argued to affect speakers' disfluency rates, we examined disfluency rates in a corpus of task-oriented conversations (Schober & Carstensen, 2001 ) with variables that might affect fluency rates. These factors included: speakers' ages (young, middle-aged, and older), task roles (director vs. matcher in a referential communication task), difficulty of topic domain (abstract geometric figures vs. photographs of children), relationships between speakers (married vs. strangers), and gender (each pair consisted of a man and a woman). Older speakers produced only slightly higher disfluency rates than young and middle-aged speakers. Overall, disfluency rates were higher both when speakers acted as directors and when they discussed abstract figures, confirming that disfluencies are associated with an increase in planning difficulty. However, fillers (such as uh) were distributed somewhat differently than repeats or restarts, supporting the idea that fillers may be a resource for or a consequence of interpersonal coordination.
This research explored the relationship between sentence disruptions and the length and complexity of sentences spoken by children developing grammar. The study was cross-sectional in design and used samples of naturalistic, conversational interaction between 26 typically developing children (ages 2;6 to 4;0) and a primary caregiver. The active, declarative sentences produced by these children were coded for the presence of disruption, length in morphemes and words, and clausal complexity. The results showed that, for the majority of the children, disrupted sentences tended to be longer and more complex than fluent sentences. The magnitude of the differences in length and complexity was positively correlated with the children's grammatical development, as measured by the Index of Productive Syntax. It was also found that differences between the average complexity of disrupted versus fluent sentences increased with grammatical development even when sentence length was held constant. As grammatical development proceeded, disrupted sentences were more apt to be sentences on the "leading-edge" of the child's production capacity. Although these more advanced grammatical structures are part of the child's grammatical competence, the child cannot produce these sentences without an increased risk of processing difficulty. The results are congruent with proposals concerning the incremental and procedural nature of adult sentence production.
Speakers are often disfluent, for example, saying "theee uh candle" instead of "the candle." Production data show that disfluencies occur more often during references to things that are discourse-new, rather than given. An eyetracking experiment shows that this correlation between disfluency and discourse status affects speech comprehensions. Subjects viewed scenes containing four objects, including two cohort competitors (e.g., camel, candle), and followed spoken instructions to move the objects. The first instruction established one cohort as discourse-given; the other was discourse-new. The second instruction was either fluent or disfluent, and referred to either the given or new cohort. Fluent instructions led to more initial fixations on the given cohort object (replicating Dahan et al., 2002). By contrast, disfluent instructions resulted in more fixations on the new cohort. This shows that discourse-new information can be accessible under some circumstances. More generally, it suggests that disfluency affects core language comprehension processes.
We describe two experiments on signaling and detecting uncertainty in audiovisual speech by adults and children. In the first study, utterances from adult speakers and child speakers (aged 7-8) were elicited and annotated with a set of six audiovisual features. It was found that when adult speakers were uncertain they were more likely to produce fillers, delays, high intonation, eyebrow movements, and "funny faces." The basic picture for the child speakers was somewhat similar, in that the presence of certain audiovisual cues correlated with uncertainty, but the differences were relatively small and less often significant. In the second study both adult and child judges watched responses from adult and child speakers selected from the first study to find out whether they were able to correctly estimate a speakers' level of uncertainty. It was found that both child and adult judges gave more accurate scores for answers from adult speakers than from child speakers and that child judges overall provided less accurate scores than adult judges.
Eye-tracking and gating experiments examined reference comprehension with fluent (Click on the red. . .) and disfluent (Click on [pause] thee uh red . . .) instructions while listeners viewed displays with 2 familiar (e.g., ice cream cones) and 2 unfamiliar objects (e.g., squiggly shapes). Disfluent instructions made unfamiliar objects more expected, which influenced listeners' on-line hypotheses from the onset of the color word. The unfamiliarity bias was sharply reduced by instructions that the speaker had object agnosia, and thus difficulty naming familiar objects (Experiment 2), but was not affected by intermittent sources of speaker distraction (beeps and construction noises; Experiments 3). The authors conclude that listeners can make situation-specific inferences about likely sources of disfluency, but there are some limitations to these attributions.
Speech disfluencies, such as filled pauses (ummm, uhhh), are increasingly recognized as an informative element of the speech stream. Here, we examined whether 2- and 3-year-olds expected that the presence of filled pause would signal reference to objects that are new to a discourse. Children viewed pairs of familiar objects on a screen and heard a speaker refer to one of the objects twice in succession. Next, children heard a critical utterance and were asked to look and point at either the discourse-given (i.e., previously mentioned) or discourse-new (i.e., previously unmentioned) object using a fluent ('Look at the ball!') or disfluent ('Look at thee uh ball!') expression. The results indicated that 3-year-old children, but not 2-year-old children, initially expected the speaker to continue to refer to given information in the critical utterance. Upon hearing a filled pause, however, both 2- and 3-year-old children's looking patterns reflected increased looks to discourse-new objects, although the timing of the effect differed between the age groups. Together, these findings demonstrate that young children have an emerging understanding of the role of filled pauses in speech.
This chapter examines the relative accessibility of different referents to a particular speaker during language comprehension, focusing on disfluency effects and how new information can be accessed. It describes experiments in which people’s eye movements to different physical objects in a scene were recorded while the degree of fluency of pronunciation of the referent in the utterance was manipulated. It shows that the fluency of production of a referent affects interpretation of that referent and that speech “errors” such as disfluent speech can offer insights into how the human production and comprehension of language work. The chapter also discusses the discourse-history approach to accessibility in the context of reference comprehension, along with the expectancy hypothesis of accessibility.
Speech disfluencies can convey information to listeners: Adults and children predict that filled pauses (e.g., uhh) will be followed by referents that are difficult to describe or are new to the discourse. In adults, this is driven partly by an understanding that disfluencies reflect processing difficulties. This experiment examined whether 3½-year-olds' use of disfluencies similarly involves inferences about processing difficulty. Forty children were introduced to either a knowledgeable or a forgetful speaker, who then produced fluent and disfluent utterances. Children exposed to the knowledgeable speaker looked preferentially at novel, discourse-new objects during disfluent utterances. However, children who heard the forgetful speaker did not. These results suggest that, like adults, children modify their expectations about the informativeness of disfluencies on a speaker-specific basis.
Upon hearing a disfluent referring expression, listeners expect the speaker to refer to an object that is previously unmentioned, an object that does not have a straightforward label, or an object that requires a longer description. Two visual-world eye-tracking experiments examined whether listeners directly associate disfluency with these properties of objects, or whether disfluency attribution is more flexible and involves situation-specific inferences. Since in natural situations reference to objects that do not have a straightforward label or that require a longer description is correlated with both production difficulty and with disfluency, we used a mini-artificial lexicon to dissociate difficulty from these properties, building on the fact that recently learned names take longer to produce than existing words in one’s mental lexicon. The results demonstrate that disfluency attribution involves situation-specific inferences; we propose that in new situations listeners spontaneously infer what may cause production difficulty. However, the results show that these situation-specific inferences are limited in scope: listeners assessed difficulty relative to their own experience with the artificial names, and did not adapt to the assumed knowledge of the speaker.
How do infants and young children coordinate information in real time to arrive at sentence meaning from the words and structure of the sentence and from the nonlinguistic context? This volume introduces readers to an emerging field of research, experimental developmental psycholinguistics , and to the four predominant methodologies used to study on-line language processing in children. Authored by key figures in psycholinguistics, neuroscience and developmental psychology, the chapters cover event-related brain potentials, free-viewing eyetracking, looking-while-listening, and reaction-time techniques, also providing a historical backdrop for this line of research. Multiple aspects of experimental design, data collection and data analysis are addressed in detail, alongside surveys of recent important findings about how infants and children process sounds, words, and sentences. Indispensable for students and researchers working in the areas of language acquisition, developmental psychology and developmental neuroscience of language, this volume will also appeal to speech language pathologists and early childhood educators.
Several recent theories have suggested that the production of speech disfluencies is related, in part, to certain aspects of language formulation. One common characteristic of these theories is the proposal that the same basic mechanism is responsible for disfluency in people who stutter as well as in people who do not stutter. To learn more about the connection between language formulation and fluency in nonstuttering speakers, this study examined relationships between syntactic complexity, utterance length, and disfluency in the spontaneous speech of 12 normally fluent children (six girls and six boys, age 44–64 months), who produced 50-utterance spontaneous speech samples during conversations with a speech–language pathologist. On average, disfluent utterances were longer and more syntactically complex than fluent utterances. Discriminant analyses indicated that utterance length in clausal constituents was the most important factor (among those assessed in this study) for predicting the likelihood that an utterance would be disfluent. Analysis of subjects' individual speech samples indicated that the selected aspects of linguistic complexity could only classify fluent and disfluent utterances at better than chance levels for a subgroup of subjects. Findings provide one means of evaluating theories that explain speech disfluencies based on normal language formulation process. Results also emphasize the value of considering individual patterns in children's speech fluency, and highlight the need to consider factors such as the timing of utterances or other contextual factors, in addition to linguistic complexity, in future studies of children's speech fluency.
Studies of filled and silent pauses performed in the last two decades are reviewed in order to determine the significance of pauses for the speaker. Following a brief history, the theoretical implications of pause location are examined and the relevant studies summarized. In addition, the functional significance of pauses is considered in terms of cognitive, affective-state, and social interaction variables.
Thirty male and female children from three chronological age groups were compared on Developmental Sentence Score, total words spoken in a language sample, total disfluency, and eight specific disfluency subtypes. Analyses were performed to determine in differences existed among the three age groups, between sexes, and whether language and disfluency were related. No sex differences in either language or disfluency were found. Total disfluency showed only a slight decrease with an increase in chronological age between 4 and 8 years. This suggests that the major disfluency decrease must occur subsequent to age 8. Although the total disfluency changed minimally between ages 4 and 8, some of the disfluency subtypes shifted considerably. Older children were almost as disfluent as younger children, but the topography of their disfluency was different. No significant relationship was found between the disfluency and language variables studied.
When preschool-aged children are presented with two objects, one familiar and one unfamiliar, and asked for the referent of a novel word, they will consistently map the novel word to the novel object, a tendency called the disambiguation effect. In this study, we examined the relation between vocabulary size and the disambiguation response tendency during late infancy. Sixteen- to 22-month-old infants were presented with a novel object along with two familiar objects and asked to choose the referents of familiar and novel words. The infants who consistently chose the novel object in the presence of a novel word had significantly higher productive vocabularies than those who did not. These two groups, however, did not differ in age or on familiar word trials. These results suggest that emergence of the disambiguation effect in late infancy is related to productive vocabulary size rather than age.
When survey respondents answer survey questions, they can also produce “paradata” (Couper 2000, 2008): behavioral evidence about their response process. The study reported here demonstrates that two kinds of respondent paradata – fluency of speech and gaze direction during answers – identify answers that are likely to be problematic, as measured by changes in answers during the interview or afterward on a post-interview questionnaire. Answers with disfluencies were less reliable both face to face and on the telephone than fluent answers, and particularly diagnostic of unreliability face to face. Interviewers’ responsivity can affect both the prevalence and potential diagnosticity of paradata: both disfluent speech and gaze aversion were more frequent and diagnostic in conversational interviews, where interviewers could provide clarification if respondents requested it or the interviewer judged it was needed, than in strictly standardized interviews where clarification was not provided even if the respondent asked for it.
We have developed a new software application, Eye-gaze Language Integration Analysis (ELIA), which allows for the rapid integration of gaze data with spoken language input (either live or prerecorded). Specifically, ELIA integrates E-Prime output and/or .csv files that include eye-gaze and real-time language information. The process of combining eye movements with real-time speech often involves multiple error-prone steps (e.g., cleaning, transposing, graphing) before a simple time course analysis plot can be viewed or before data can be imported into a statistical package. Some of the advantages of this freely available software include (1) reducing the amount of time spent preparing raw eye-tracking data for analysis; (2) allowing for the quick analysis of pilot data in order to identify issues with experimental design; (3) facilitating the separation of trial types, which allows for the examination of supplementary effects (e.g., order or gender effects); and (4) producing standard output files (i.e., .csv files) that can be read by numerous spreadsheet packages and transferred to any statistical software.
Eye movements to pictures of four objects on a screen were monitored as participants followed a spoken instruction to move one of the objects, e.g., “Pick up the beaker; now put it below the diamond” (Experiment 1) or heard progressively larger gates and tried to identify the referent (Experiment 2). The distractor objects included a cohort competitor with a name that began with the same onset and vowel as the name of the target object (e.g.,beetle), a rhyme competitor (e.g.speaker), and an unrelated competitor (e.g.,carriage). In Experiment 1, there was clear evidence for both cohort and rhyme activation as predicted by continuous mapping models such as TRACE (McClelland and Elman, 1986) and Shortlist (Norris, 1994). Additionally, the time course and probabilities of eye movements closely corresponded to response probabilities derived from TRACE simulations using the Luce choice rule (Luce, 1959). In the gating task, which emphasizes word-initial information, there was clear evidence for multiple activation of cohort members, as measured by judgments and eye movements, but no suggestion of rhyme effects. Given that the same sets of pictures were present during the gating task as in Experiment 1, we conclude that the rhyme effects in Experiment 1 were not an artifact of using a small set of visible alternatives.
People responding to questions are sometimes uncertain, slow, or unable to answer. They handle these problems of self-presentation, we propose, by the way they respond. Twenty-five respondents were each asked 40 factual questions in a conversational setting. Later, they rated for each question their feeling that they would recognize the correct answer, then took a recognition test on all 40 questions. As found previously, the weaker their feeling of knowing, the slower their answers, the faster their nonanswers ("I don′t know"), and the worse their recognition. But further, as proposed, the weaker their feeling of knowing, the more often they answered with rising intonation, used hedges such as "I guess," responded "I don′t know" instead of "I can′t remember," and added "uh" or "um," self-talk, and other face-saving comments. They reliably used "uh" to signal brief delays and "um" longer ones.
In early lexical development, children must learn to map spoken words onto their respective referents. Since multiple objects are typically present when any word is used, a child is charged with the difficult task of inferring the speaker's intended referent. Previous research has uncovered various cues children may use in this task, including contextual and social cues. We investigate a previously unexplored cue for inferring speaker intention during word learning: speech disfluencies. Disfluencies (such as "uh" and "um") occur in highly predictable locations, such as before words that are infrequent and words that have not been previously mentioned. Further, since they occur before such words, they could enable a young word learner to anticipate an upcoming referent. We conducted an eye-tracking study to investigate whether young children can make use of the information contained in disfluencies. Our results demonstrate that young children (ages 2;4 to 2;8) are sensitive to disfluencies. More critically, they show that children appear to use disfluencies predictively as a cue to reference and to speaker intention as the disfluency is occurring. We also examined potential sources of learning about disfluencies in CHILDES (MacWhinney, 2000) and found that disfluencies, though rare, occur regularly and with increasing frequency over time in child-directed and child-produced speech. These results demonstrate that a feature of the speech signal that likely emerges as a by-product of processing demands in adults is attended to relatively early in lexical development and used by young children to infer speaker intention during online comprehension.
Because of its potential importance for word learning, children’s judgment of whether they know names for objects was investigated. In Study 1, judgment accuracy was at or near ceiling in about two-thirds of 4-year-olds, and covaried with judgment of word familiarity and with justifying novel name mapping in terms of avoidance of name overlap. The two judgments and the justification were unrelated to general intelligence, which is evidence that awareness of lexical ignorance is a specific metalinguistic ability in this age group. Preexposure to an unfamiliar object increased the likelihood of it being misjudged as having a known name, but only by the children who scored above the median in word familiarity judgment. Thus, judgment of lexical ignorance may undergo a developmental change from relying solely on whether target information is retrieved to considering additional information. Consistent with this hypothesis, preexposure did not affect -year-olds’ judgments of object name knowledge in Study 2. About one-third in this age group identified more familiar than unfamiliar objects as having known names. These judgments covaried with word familiarity judgments, but were not related to novel name mapping justifications and were related to general intelligence. Thus, awareness of lexical ignorance does not become a unified metalinguistic ability until after years.
Speech disfluencies have different effects on comprehension depending on the type and placement of disfluency. Words following false starts (such as windmill after in the in the eleventh example is um in the a windmill) have longer word monitoring latencies than the same tokens with the false starts excised. The decremental effect seems to be limited to false starts that occur in the middle of sentences or after discourse markers. I suggest it is at these points that the repair process is most burdened by the false start. In contrast, words following repetitions (heart in of a of a heart) do not have longer word monitoring latencies than the same tokens with the repetitions excised. In two experiments, words following spontaneously produced repetitions have faster word monitoring latencies. Two other experiments suggest that this seeming repetition advantage is more likely the result of slowed monitoring after a phonological phrase disruption. Inserting repetitions where they did not occur in a manner that preserved the original phonological phrases resulted in neither an advantage nor a disadvantage or repeating. These studies provide a first glimpse at how speech disfluencies affect understanding, and also provide information about the types of comprehension models that can accommodate the effects of speech disfluencies.
The notion of common ground is important for the production of referring expressions: In order for a referring expression to be felicitous, it has to be based on shared information. But determining what information is shared and what information is privileged may require gathering information from multiple sources, and constantly coordinating and updating them, which might be computationally too intensive to affect the earliest moments of production. Previous work has found that speakers produce overinformative referring expressions, which include privileged names, violating Grice's Maxims, and concluded that this is because they do not mark the distinction between shared and privileged information. We demonstrate that speakers are in fact quite effective in marking this distinction in the form of their utterances. Nonetheless, under certain circumstances, speakers choose to overspecify privileged names.
We investigated the mechanisms by which fillers, such as uh and um, affect memory for discourse. Participants listened to and attempted to recall recorded passages adapted from Alice's Adventures in Wonderland. The type and location of interruptions were manipulated through digital splicing. In Experiment 1, we tested a processing time account of fillers' effects. While fillers facilitated recall, coughs matched in duration to the fillers impaired recall, suggesting that fillers' benefits cannot be attributed to adding processing time. In Experiment 2, fillers' locations were manipulated based on norming data to be either predictive or non-predictive of upcoming material. Fillers facilitated recall in both cases, inconsistent with an account in which listeners predict upcoming material using past experience with the distribution of fillers. Instead, these results suggest an attentional orienting account in which fillers direct attention to the speech stream but do not always result in specific predictions about upcoming material.
When listeners hear a speaker become disfluent, they expect the speaker to refer to something new. What is the mechanism underlying this expectation? In a mouse-tracking experiment, listeners sought to identify images that a speaker was describing. Listeners more strongly expected new referents when they heard a speaker say um than when they heard a matched utterance where the um was replaced by noise. This expectation was speaker-specific: it depended on what was new and old for the current speaker, not just on what was new or old for the listener. This finding suggests that listeners treat fillers as collateral signals.
Research has shown that preschoolers monitor others' prior accuracy and prefer to learn from individuals who have the best track record. We investigated the scope of preschoolers' attributions based on an individual's prior accuracy. Experiment 1 revealed that 5-year-olds (but not 4-year-olds) used an individual's prior accuracy at labelling to predict her knowledge of words and broader facts; they also showed a 'halo effect' predicting she would be more prosocial. Experiment 2 confirmed that, overall, 4-year-olds did not make explicit generalizations of knowledge. These findings suggest that an individual's prior accuracy influences older preschoolers' expectations of that individual's broader knowledge as well as their impressions of how she will behave in social interactions.
In these studies, we examined how a default assumption about word meaning, the mutual exclusivity assumption and an intentional cue, gaze direction, interacted to guide 24-month-olds' object-word mappings. In Expt 1, when the experimenter's gaze was consistent with the mutual exclusivity assumption, novel word mappings were facilitated. When the experimenter's eye-gaze was in conflict with the mutual exclusivity cue, children demonstrated a tendency to rely on the mutual exclusivity assumption rather than follow the experimenter's gaze to map the label to the object. In Expt 2, children relied on the experimenter's gaze direction to successfully map both a first label to a novel object and a second label to a familiar object. Moreover, infants mapped second labels to familiar objects to the same degree that they mapped first labels to novel objects. These findings are discussed with regard to children's use of convergent and divergent cues in indirect word mapping contexts.
This study examined associations between peers' expressive language abilities and children's development of receptive and expressive language among 1,812 four-year olds enrolled in 453 classrooms in 11 states that provide large-scale public pre-kindergarten (pre-k) programs. Higher peer expressive language abilities were positively associated with children's development of receptive and expressive language during pre-k. The positive association between peers' expressive language abilities and children's receptive language development was stronger for children who began pre-k with higher receptive language skills and within classrooms characterized by better classroom management. Implications of these findings for understanding ecological inputs to children's language development and for designing effective pre-k programs are discussed.
In 3 experiments, children's ability to vary their responses on perspective-taking tasks as a function of the other person's age was examined. In Experiment 1,4- and 5-year-olds were shown to be accurate in their judgments about the knowledge of a 6-month-old baby, a 4-year-old child, and an adult. In Experiment 2, 4-year-olds were asked to determine if a baby, child, and adult would be able to identify an object from a restricted view showing either an identifiable part, a small nondescript part, or no part of the object. Children tended to report that the observer would be able to identify the object from an identifiable or nondescript part. Their judgments were not affected by the age of the observer. Experiment 3 replicated the asymmetry in performance on the general knowledge task and the restricted view task and extended these results by testing 4-year-olds, 6-year-olds, and adults using a within-subjects design.
Nearly every recent account of children's word learning has addressed the claim that children are biased to construct mutually exclusive extensions, that is, that they are disposed to keep the set of referents of one word from overlapping with those of others. Three basic positions have been taken--that children have the bias when they first start to learn words, that they never have it, and that they acquire it during early childhood. A review of diary and test evidence as well as the results of four experiments provide strong support for this last view and indicate that the bias develops in the months following the second birthday but does not gain full strength or become accessible to consciousness until sometime after the third birthday. Several studies also show that, after this point, it can still be counteracted by information in input or by a strong belief that something belongs to the extension of a particular word. The full body of evidence is compatible with the view that mutual exclusivity is the default option in children's and adults' procedures for integrating the extensions of new and old words. We present several arguments for the adaptive value of this kind of bias.
An empirical puzzle regarding toddlers' fast mapping motivated the current investigation. Whereas children between 1;10 and 2;1 have shown only a modest rate of mapping novel nouns onto unfamiliar rather than familiar objects, a very high rate has been observed in those between 1;4 and 1;8 (Mervis & Bertrand, 1991). Study 1 examined whether young two-year-olds (N = 40, mean age = 2;1) might map at a higher rate when tested with procedures unique to Mervis & Bertrand's assessment--strong corrective feedback rather than mild positive non-contingent feedback; large sets of test objects rather than pairs; presentation of easier tests first. Only the first variable affected performance in a manner that could solve the puzzle. Unfamiliar kinds were selected at a much higher rate under corrective (0.86) than non-contingent (0.57) feedback. Although nearly every child in the non-contingent group chose correctly on the first trial, many failed to do so thereafter. In Study 2, rather than presenting a test word to the children (N = 16, mean age = 2;2), the experimenter merely asked for 'the one I want'. Unfamiliar kinds were selected much less often than in Study 1, suggesting that at least one lexical principle proposed in the literature underlies the noun mapping preference. Changes over trials in the two studies indicated that the noun mapping preference is quite prevalent, but unless initial choices are strongly reinforced, an increase in the salience of familiar kinds after the first trial lures some children into error. Consistent with this analysis, toddlers in Study 3 (N = 24, mean age = 2:1) who received non-contingent strong acceptance for their noun mapping decisions, selected unfamiliar kinds more often than those who had received non-contingent mild acceptance in Study 1.
Unlabelled:
The purpose of this study was to examine the influence of utterance length and complexity relative to the children's mean length of utterance (MLU) on stuttering-like disfluencies (SLDs) for children who stutter (CWS) and nonstuttering-like disfluencies (nonSLDs) for children who do not stutter (CWNS). Participants were 12 (3;1-5;11, years;months) children: 6 CWS and 6 age-matched (+/-5 months) CWNS, with equal numbers in each talker group (CWS and CWNS) exhibiting MLU from the lower to the upper end of normal limits. Data were based on audio-video recordings of each child in two separate settings (i.e., home and laboratory) during loosely structured, 30-min parent-child conversational interactions and analyzed in terms of each participant's utterance length, MLU, frequency and type of speech disfluency. Results indicate that utterances above children's MLU are more apt to be stuttered or disfluent and that both stuttering-like as well as nonstuttering-like disfluencies are most apt to occur on utterances that are both long and complex. Findings were taken to support the hypothesis that the relative "match" or "mismatch" between linguistic components of an utterance (i.e., utterance length and complexity) and a child's language proficiency (i.e., MLU) influences the frequency of the child's stuttering/speech disfluency.
Educational objectives:
The reader will learn about and be able to: (1) compare different procedures for assessing the relationship among stuttering, length and complexity of utterance, (2) describe the difference between relative and absolute measures of utterance length, (3) discuss the measurement and value of mean length of utterance and its possible contributions to childhood stuttering, and (4) describe how length and complexity influence nonstuttering-like disfluencies of children who stutter as well as the stuttering-like disfluencies of children who do not stutter.
Most research on the rapid mental processes of on-line language processing has been limited to the study of idealized, fluent utterances. Yet speakers are often disfluent, for example, saying "thee, uh, candle" instead of "the candle." By monitoring listeners' eye movements to objects in a display, we demonstrated that the fluency of an article ("thee uh" vs. "the") affects how listeners interpret the following noun. With a fluent article, listeners were biased toward an object that had been mentioned previously, but with a disfluent article, they were biased toward an object that had not been mentioned. These biases were apparent as early as lexical information became available, showing that disfluency affects the basic processes of decoding linguistic input.
Disfluencies ("um," repeats, self-repairs) are prevalent in spontaneous speech, and are relevant to both human speech communication and speech processing by machine. Although disfluencies have commonly been viewed as `noisy' events, results from a large descriptive study indicate that disfluencies show regularities in a number of dimensions [9]. This paper reports selected results on Switchboard and two comparison corpora of spontaneous speech. Results illustrate the systematic distribution of disfluencies, and highlight differences as well as universals across corpora and speakers.
This research characterizes the spontaneous spoken disfluencies typical of human-computer interaction, and presents a predictive model accounting for their occurrence. Data were collected during three empirical studies in which people spoke or wrote to a highly interactive simulated system as they completed service transactions. The studies involved within-subject factorial designs in which the input modality and presentation format were varied. Spoken disfluency rates during human-computer interaction were documented to be substantially lower than rates typically observed during comparable human-human speech. Two separate factors, both associated with increased planning demands, were statistically related to higher disfluency rates: (1) length of utterance, and (2) lack of structure in the presentation format. Regression techniques demonstrated that a linear model based simply on utterance length accounted for over 77% of the variability in spoken disfluencies. Therefore, design methods ca...
Understanding psychology as a science: an introduction to scientific and statistical inference
- Z Dienes
Dienes, Z. (). Understanding psychology as a science: an introduction to scientific and
statistical inference. Basingstoke: Palgrave Macmillan.
Developmental psycholinguistics: on-line methods in children's language processing
- J C Trueswell
Trueswell, J. C. (). Using eye movements as a developmental measure within
psycholinguistics. In I. A. Sekerina, E. M. Fernandez & H. Clahsen (Eds), Developmental
psycholinguistics: on-line methods in children's language processing, -. Philadelphia, PA:
John Benjamins Publishing Company.