Word-object Mapping Research Articles

The quality of caregiver-child interaction is predicted by (caregivers' perception of) their child's interests.

This current study examines the extent to which children's interests and caregivers' sensitivity to their children's interests are associated with the quality of caregiver-child interaction, and subsequent learning. Eighty-one caregiver-child dyads (24-30-month old children) completed an online shared book-reading task where caregivers and children read two e-books with pictures and descriptions of objects from different categories-one previously determined to be of low and one of high interest to the child (with one novel word-object mapping introduced in each book). We also obtained separate behavioural indices of children's interests and children's later recognition of newly introduced word-object mappings. Our findings highlight that the quality of caregiver-child interaction is predicted by children's interests and caregivers' perception of children's interests, although we find only limited overlap between our behavioural indices of children's interests and caregiver perception of children's interests. Neither of these factors predicted later novel word recognition. Thus, while the dynamics between higher quality of caregiver-child interaction, children's interests and learning remain inconclusive, caregivers and children appear to be more attentive, enthusiastic and engaged in reading about topics that (caregivers believe) interest the child. Furthermore, learning in itself seems to be successful, regardless of factors involved, through the mere task of shared book reading.

Not all is forgotten: Children's associative matrices for features of a word learning episode.

Word learning studies traditionally examine the narrow link between words and objects, indifferent to the rich contextual information surrounding objects. This research examined whether children attend to this contextual information and construct an associative matrix of the words, objects, people, and environmental context during word learning. In Experiment 1, preschool-aged children (age: 3;2-5;11 years) were presented with novel words and objects in an animated storybook. Results revealed that children constructed associations beyond words and objects. Specifically, children attended to and had the strongest associations for features of the environmental context but failed to learn word-object associations. Experiment 2 demonstrated that children (age: 3;0-5;8 years) leveraged strong associations for the person and environmental context to support word-object mapping. This work demonstrates that children are especially sensitive to the word learning context and use associative matrices to support word mapping. Indeed, this research suggests associative matrices of the environment may be foundational for children's vocabulary development.

Linking disambiguation and retention in a developmental eye-tracking study with monolingual and multilingual children

The disambiguation effect, also referred to as process of elimination, occurs during word learning, whereby novel words are mapped onto new referents, precluding the application of a novel label to a familiar object. Prior studies showed that the emergence and use of disambiguation can be affected by children’s vocabulary growth and linguistic experience, such as growing up with more than one language. To test this, we investigated (a) whether monolingual and multilingual children disambiguated a novel word–object mapping, (b) whether they retained a trained, previously seen word–object mapping, (c) whether they retained the novel fast-mapped word–object mapping, and (d) whether and how age, English vocabulary size, and language background modulated disambiguation and retention. Lastly, we tested (e) whether children who disambiguated also retained better. Eye-tracking data from 18- to 30-month-old monolingual children (n = 43) and multilingual children (n = 40) were collected. A looking-while-listening paradigm with two objects included two familiar items, one novel item, and one trained item. Mixed-effect models reported that vocabulary size predicted the outcome of mapping and retention better than age. Monolingual children’s accuracy on disambiguation trials was high from the start, whereas multilingual children started to disambiguate later as their vocabulary grew. Only monolingual children performed above chance level on retaining the novel label. Lastly, the use of disambiguation improved retention for monolingual children but not for multilingual children. This research corroborates that disambiguation should be regarded as a mechanism facilitating default fast mapping rather than fully fledged learning. Vocabulary growth leading to an increase in disambiguation supports the notion that the disambiguation effect stems from prior episodes of learning.

Adults Use Cross-Situational Statistics for Word Learning in a Conservative Way

This paper examined how much adults rely on cross-situational information in word learning by comparing their gaze behavior in a word learning task with models of four learning strategies. We manipulated the input type of situations (consecutive versus interleaved) and the co-occurrence frequencies between words and objects so that adult learners could infer correct word-object mappings based on cross-situational information. There are two key findings. First, an exposure-by-exposure analysis of gaze behavior during the word learning procedure revealed that most participants collected sufficient cross-situational information before they developed a preference for one particular word-object mapping, with consecutive as well as interleaved situations. Second, a classification approach in which individual gaze behavior was attributed to different word learning strategies showed that participants relied mostly on a conservative cross-situational learning (XSL) strategy, compared to Associative XSL, Propose-but-Verify, and Random strategies. Adults relied on Conservative XSL when presented with consecutive and interleaved situations, but they shifted toward Associative XSL when presented with interleaved situations.

Mechanisms of Cross-situational Learning: Behavioral and Computational Evidence.

Word learning happens in everyday contexts with many words and many potential referents for those words in view at the same time. It is challenging for young learners to find the correct referent upon hearing an unknown word at the moment. This problem of referential uncertainty has been deemed as the crux of early word learning (Quine, 1960). Recent empirical and computational studies have found support for a statistical solution to the problem termed cross-situational learning. Cross-situational learning allows learners to acquire word meanings across multiple exposures, despite each individual exposure is referentially uncertain. Recent empirical research shows that infants, children and adults rely on cross-situational learning to learn new words (Smith & Yu, 2008; Suanda, Mugwanya, & Namy, 2014; Yu & Smith, 2007). However, researchers have found evidence supporting two very different theoretical accounts of learning mechanisms: Hypothesis Testing (Gleitman, Cassidy, Nappa, Papafragou, & Trueswell, 2005; Markman, 1992) and Associative Learning (Frank, Goodman, & Tenenbaum, 2009; Yu & Smith, 2007). Hypothesis Testing is generally characterized as a form of learning in which a coherent hypothesis regarding a specific word-object mapping is formed often in conceptually constrained ways. The hypothesis will then be either accepted or rejected with additional evidence. However, proponents of the Associative Learning framework often characterize learning as aggregating information over time through implicit associative mechanisms. A learner acquires the meaning of a word when the association between the word and the referent becomes relatively strong. In this chapter, we consider these two psychological theories in the context of cross-situational word-referent learning. By reviewing recent empirical and cognitive modeling studies, our goal is to deepen our understanding of the underlying word learning mechanisms by examining and comparing the two theoretical learning accounts.

Social interaction facilitates word learning in preverbal infants: Word–object mapping and word segmentation

In natural settings, infants learn spoken language with the aid of a caregiver who explicitly provides social signals. Although previous studies have demonstrated that young infants are sensitive to these signals that facilitate language development, the impact of real-life interactions on early word segmentation and word–object mapping remains elusive. We tested whether infants aged 5–6 months and 9–10 months could segment a word from continuous speech and acquire a word–object relation in an ecologically valid setting. In Experiment 1, infants were exposed to a live tutor, while in Experiment 2, another group of infants were exposed to a televised tutor. Results indicate that both younger and older infants were capable of segmenting a word and learning a word–object association only when the stimuli were derived from a live tutor in a natural manner, suggesting that real-life interaction enhances the learning of spoken words in preverbal infants.

Modulation of scene consistency and task demand on language-driven eye movements for audio-visual integration

Previous psycholinguistic studies have demonstrated that people tend to direct fixations toward the visual object to which spoken input refers during language comprehension. However, it is still unclear how the visual scene, especially the semantic consistency between object and background, affects the word-object mapping process during comprehension. Two visual world paradigm experiments were conducted to investigate how the scene consistency dynamically influenced the language-driven eye movements in a speech comprehension and a scene comprehension task. In each trial, participants listened to a spoken sentence while viewing a picture with two critical objects: one is the mentioned target object (e.g., tiger), which was embedded in either a consistent (e.g., field), inconsistent (e.g., sky) or blank background; the other is an unmentioned non-target object (e.g., eagle), which was always consistent with its background. The results showed that the fixation proportion of the inconsistent target was higher than the consistent target, and the task demand can affect the strength and the direction of the inconsistency effect before and after the target had been mentioned. In summary, the spoken language, scene-based knowledge and task demand were intertwined to determine eye movements during audio-visual integration for comprehension.

Posture affects how robots and infants map words to objects.

For infants, the first problem in learning a word is to map the word to its referent; a second problem is to remember that mapping when the word and/or referent are again encountered. Recent infant studies suggest that spatial location plays a key role in how infants solve both problems. Here we provide a new theoretical model and new empirical evidence on how the body – and its momentary posture – may be central to these processes. The present study uses a name-object mapping task in which names are either encountered in the absence of their target (experiments 1–3, 6 & 7), or when their target is present but in a location previously associated with a foil (experiments 4, 5, 8 & 9). A humanoid robot model (experiments 1–5) is used to instantiate and test the hypothesis that body-centric spatial location, and thus the bodies’ momentary posture, is used to centrally bind the multimodal features of heard names and visual objects. The robot model is shown to replicate existing infant data and then to generate novel predictions, which are tested in new infant studies (experiments 6–9). Despite spatial location being task-irrelevant in this second set of experiments, infants use body-centric spatial contingency over temporal contingency to map the name to object. Both infants and the robot remember the name-object mapping even in new spatial locations. However, the robot model shows how this memory can emerge –not from separating bodily information from the word-object mapping as proposed in previous models of the role of space in word-object mapping – but through the body’s momentary disposition in space.

Event-related potentials during word mapping to object shape predict toddlers' vocabulary size.

What role does attention to different object properties play in early vocabulary development? This longitudinal study using event-related potentials in combination with behavioral measures investigated 20- and 24-month-olds' (n = 38; n = 34; overlapping n = 24) ability to use object shape and object part information in word-object mapping. The N400 component was used to measure semantic priming by images containing shape or detail information. At 20 months, the N400 to words primed by object shape varied in topography and amplitude depending on vocabulary size, and these differences predicted productive vocabulary size at 24 months. At 24 months, when most of the children had vocabularies of several hundred words, the relation between vocabulary size and the N400 effect in a shape context was weaker. Detached object parts did not function as word primes regardless of age or vocabulary size, although the part-objects were identified behaviorally. The behavioral measure, however, also showed relatively poor recognition of the part-objects compared to the shape-objects. These three findings provide new support for the link between shape recognition and early vocabulary development.

The role of partial knowledge in statistical word learning.

A critical question about the nature of human learning is whether it is an all-or-none or a gradual, accumulative process. Associative and statistical theories of word learning rely critically on the later assumption: that the process of learning a word's meaning unfolds over time. That is, learning the correct referent for a word involves the accumulation of partial knowledge across multiple instances. Some theories also make an even stronger claim: partial knowledge of one word-object mapping can speed up the acquisition of other word-object mappings. We present three experiments that test and verify these claims by exposing learners to two consecutive blocks of cross-situational learning, in which half of the words and objects in the second block were those that participants failed to learn in Block 1. In line with an accumulative account, Re-exposure to these mis-mapped items accelerated the acquisition of both previously experienced mappings and wholly new word-object mappings. But how does partial knowledge of some words speed the acquisition of others? We consider two hypotheses. First, partial knowledge of a word could reduce the amount of information required for it to reach threshold, and the supra-threshold mapping could subsequently aid in the acquisition of new mappings. Alternatively, partial knowledge of a word's meaning could be useful for disambiguating the meanings of other words even before the threshold of learning is reached. We construct and compare computational models embodying each of these hypotheses and show that the latter provides a better explanation of the empirical data.

Failure to learn from feedback underlies word learning difficulties in toddlers at risk for autism.

Children's assignment of novel words to nameless objects, over objects whose names they know (mutual exclusivity; ME) has been described as a driving force for vocabulary acquisition. Despite their ability to use ME to fast-map words (Preissler & Carey, 2005), children with autism show impaired language acquisition. We aimed to address this puzzle by building on studies showing that correct referent selection using ME does not lead to word learning unless ostensive feedback is provided on the child's object choice (Horst & Samuelson, 2008). We found that although toddlers aged 2;0 at risk for autism can use ME to choose the correct referent of a word, they do not benefit from feedback for long-term retention of the word-object mapping. Further, their difficulty using feedback is associated with their smaller receptive vocabularies. We propose that difficulties learning from social feedback, not lexical principles, limits vocabulary building during development in children at risk for autism.

Word Generalization by a Dog (Canis familiaris): Is Shape Important?

We investigated the presence of a key feature of human word comprehension in a five year old Border Collie: the generalization of a word referring to an object to other objects of the same shape, also known as shape bias. Our first experiment confirmed a solid history of word learning in the dog, thus making it possible for certain object features to have become central in his word comprehension. Using an experimental paradigm originally employed to establish shape bias in children and human adults we taught the dog arbitrary object names (e.g. dax) for novel objects. Two experiments showed that when briefly familiarized with word-object mappings the dog did not generalize object names to object shape but to object size. A fourth experiment showed that when familiarized with a word-object mapping for a longer period of time the dog tended to generalize the word to objects with the same texture. These results show that the dog tested did not display human-like word comprehension, but word generalization and word reference development of a qualitatively different nature compared to humans. We conclude that a shape bias for word generalization in humans is due to the distinct evolutionary history of the human sensory system for object identification and that more research is necessary to confirm qualitative differences in word generalization between humans and dogs.

Statistical speech segmentation and word learning in parallel: scaffolding from child-directed speech.

In order to acquire their native languages, children must learn richly structured systems with regularities at multiple levels. While structure at different levels could be learned serially, e.g., speech segmentation coming before word-object mapping, redundancies across levels make parallel learning more efficient. For instance, a series of syllables is likely to be a word not only because of high transitional probabilities, but also because of a consistently co-occurring object. But additional statistics require additional processing, and thus might not be useful to cognitively constrained learners. We show that the structure of child-directed speech makes simultaneous speech segmentation and word learning tractable for human learners. First, a corpus of child-directed speech was recorded from parents and children engaged in a naturalistic free-play task. Analyses revealed two consistent regularities in the sentence structure of naming events. These regularities were subsequently encoded in an artificial language to which adult participants were exposed in the context of simultaneous statistical speech segmentation and word learning. Either regularity was independently sufficient to support successful learning, but no learning occurred in the absence of both regularities. Thus, the structure of child-directed speech plays an important role in scaffolding speech segmentation and word learning in parallel.

Language-mediated visual orienting behavior in low and high literates.

The influence of formal literacy on spoken language-mediated visual orienting was investigated by using a simple look and listen task which resembles every day behavior. In Experiment 1, high and low literates listened to spoken sentences containing a target word (e.g., “magar,” crocodile) while at the same time looking at a visual display of four objects (a phonological competitor of the target word, e.g., “matar,” peas; a semantic competitor, e.g., “kachuwa,” turtle, and two unrelated distractors). In Experiment 2 the semantic competitor was replaced with another unrelated distractor. Both groups of participants shifted their eye gaze to the semantic competitors (Experiment 1). In both experiments high literates shifted their eye gaze toward phonological competitors as soon as phonological information became available and moved their eyes away as soon as the acoustic information mismatched. Low literates in contrast only used phonological information when semantic matches between spoken word and visual referent were not present (Experiment 2) but in contrast to high literates these phonologically mediated shifts in eye gaze were not closely time-locked to the speech input. These data provide further evidence that in high literates language-mediated shifts in overt attention are co-determined by the type of information in the visual environment, the timing of cascaded processing in the word- and object-recognition systems, and the temporal unfolding of the spoken language. Our findings indicate that low literates exhibit a similar cognitive behavior but instead of participating in a tug-of-war among multiple types of cognitive representations, word–object mapping is achieved primarily at the semantic level. If forced, for instance by a situation in which semantic matches are not present (Experiment 2), low literates may on occasion have to rely on phonological information but do so in a much less proficient manner than their highly literate counterparts.

Rapid fast-mapping abilities in 2-year-olds

Learning a new word consists of two primary tasks that have often been conflated into a single process: referent selection, in which a child must determine the correct referent of a novel label, and referent retention, which is the ability to store this newly formed label–object mapping in memory for later use. In addition, children must be capable of performing these tasks rapidly and repeatedly as they are frequently exposed to novel words during the course of natural conversation. Here we used a preferential pointing task to investigate 2-year-olds’ ( N = 72) ability to infer the referent of a novel noun from a single ambiguous exposure and their ability to retain this mapping over time. Children were asked to identify the referent of a novel label on six critical trials distributed throughout the course of a 10-min study involving many familiar and novel objects. On these critical trials, images of a known object and a novel object (e.g., a ball and a nameless artifact constructed in the laboratory) appeared on two computer screens and a voice asked children to “point at the _____ [e.g., glark].” Following label onset, children were allowed only 3 s during which to infer the correct referent, point at it, and potentially store this new word–object mapping. In a final posttest trial, all previously labeled novel objects appeared and children were asked to point to one of them (e.g., “Can you find the glark?”). To succeed, children needed to have initially mapped the novel labels correctly and retained these mappings over the course of the study. Despite the difficult demands of the current task, children successfully identified the target object on the retention trial. We conclude that 2-year-olds are able to fast map novel nouns during a brief single exposure under ambiguous labeling conditions.

Early referential understanding: Infants' ability to recognize referential acts for what they are.

Previous research has documented infants' ability to follow adults' line-of-regard and pointing gestures by 9-12 months of age, but it has not been clear whether infants appreciate that such cues are a privileged source of information about word reference. Study 1 demonstrated that this is understood by at least 19-20 months: Infants wed referential cues to guide a new word-object mapping even though this required that they ignore competition from temporal contiguity. In Study 2, the effects observed in Study 1 were not obtained with actions that served to enhance the salience of an object nonreferentially (e.g., manipulating without looking), suggesting that infants appreciate that actions that accompany labeling are not necessarily referential