Information sources for noun learning

Why are some words easier to learn than others? And what enables the eventual learning of the more difficult words? These questions were addressed for nouns using a paradigm in which adults were exposed to naturalistic maternal input that was manipulated to simulate access to several different information sources, both alone and in combination: observation of the extralinguistic contexts in which the target word was used, the words that co‐occurred with the target word, and the target word's syntactic context. Words that were not accurately identified from observation alone were both abstract (e.g., music) and concrete (e.g., tail). Whether a noun could be learned from observation depended on whether it labeled a basic‐level object category (BLOC). However, the difference between BLOC labels and non‐BLOC labels was eliminated when observation was supplemented with linguistic context. Thus, although BLOC labels can be learned from observation alone, non‐BLOC labels require richer linguistic context. These findings support a model of vocabulary growth in which an important role is played by changes in the information to which learners have access.     

A Bayesian framework for word segmentation: Exploring the effects of context

Since the experiments of Saffran et al. [Saffran, J., Aslin, R., & Newport, E. (1996). Statistical learning in 8-month-old infants. Science, 274, 1926-1928], there has been a great deal of interest in the question of how statistical regularities in the speech stream might be used by infants to begin to identify individual words. In this work, we use computational modeling to explore the effects of different assumptions the learner might make regarding the nature of words - in particular, how these assumptions affect the kinds of words that are segmented from a corpus of transcribed child-directed speech. We develop several models within a Bayesian ideal observer framework, and use them to examine the consequences of assuming either that words are independent units, or units that help to predict other units. We show through empirical and theoretical results that the assumption of independence causes the learner to undersegment the corpus, with many two- and three-word sequences (e.g. what’s that, do you, in the house) misidentified as individual words. In contrast, when the learner assumes that words are predictive, the resulting segmentation is far more accurate. These results indicate that taking context into account is important for a statistical word segmentation strategy to be successful, and raise the possibility that even young infants may be able to exploit more subtle statistical patterns than have usually been considered.     

An algebra of human concept learning

An important element of learning from examples is the extraction of patterns and regularities from data. This paper investigates the structure of patterns in data defined over discrete features, i.e. features with two or more qualitatively distinct values. Any such pattern can be algebraically decomposed into a spectrum of component patterns, each of which is a simpler or more atomic “regularity.” Each component regularity involves a certain number of features, referred to as its degree. Regularities of lower degree represent simpler or more coarse patterns in the original pattern, while regularities of higher degree represent finer or more idiosyncratic patterns. The full spectral breakdown of a pattern into component regularities of minimal degree, referred to as its power series, expresses the original pattern in terms of the regular rules or patterns it obeys, amounting to a kind of “theory” of the pattern. The number of regularities at various degrees necessary to represent the pattern is tabulated in its power spectrum, which expresses how much of a pattern's structure can be explained by regularities of various levels of complexity. A weighted mean of the pattern's spectral power gives a useful numeric summary of its overall complexity, called its algebraic complexity. The basic theory of algebraic decomposition is extended in several ways, including algebraic accounts of the typicality of individual objects within concepts, and estimation of the power series from noisy data. Finally some relations between these algebraic quantities and empirical data are discussed.     

The origins of concepts

Certain of our concepts are innate, but many others are learned. Despite the plausibility of this claim, some have argued that the very idea of concept learning is incoherent. I present a conception of learning that sidesteps the arguments against the possibility of concept learning, and sketch several mechanisms that result in the generation of new primitive concepts. Given the rational considerations that motivate their deployment, I argue that these deserve to be called learning mechanisms. I conclude by replying to the objections that these mechanisms cannot produce genuinely new content and cannot be part of genuinely cognitive explanations.

The role of incremental parsing in syntactically conditioned word learning

In a series of three experiments, we use children’s noun learning as a probe into their syntactic knowledge as well as their ability to deploy this knowledge, investigating how the predictions children make about upcoming syntactic structure change as their knowledge changes. In the first two experiments, we show that children display a developmental change in their ability to use a noun’s syntactic environment as a cue to its meaning. We argue that this pattern arises from children’s reliance on their knowledge of verbs’ subcategorization frame frequencies to guide parsing, coupled with an inability to revise incremental parsing decisions. We show that this analysis is consistent with the syntactic distributions in child-directed speech. In the third experiment, we show that the change arises from predictions based on verbs’ subcategorization frame frequencies.     

The relationship between the shape of the mental number line and familiarity with numbers in 5- to 9-year old children: evidence for a segmented linear model

This experiment aimed to expand previous findings on the development of mental number representation. We tested the hypothesis that children's familiarity with numbers is directly reflected by the shape of their mental number line. This mental number line was expected to be linear as long as numbers lay within the range of numbers children were familiar with. Five- to 9-year-olds (N=78) estimated the positions of numbers on an external number line and additionally completed a counting assessment mirroring their familiarity with numbers. A segmented regression model consisting of two linear segments described number line estimations significantly better than a logarithmic or a simple linear model. Moreover, the change point between the two linear segments, indicating a change of discriminability between numbers, was significantly correlated with children's familiar number range. Findings are discussed in terms of the accumulator model, assuming a linear mental representation with scalar variability.     

Statistical models for the induction and use of selectional preferences

Selectional preferences have a long history in both generative and computational linguistics. However, since the publication of Resnik's dissertation in 1993, a new approach has surfaced in the computational linguistics community. This new line of research combines knowledge represented in a pre‐defined semantic class hierarchy with statistical tools including information theory, statistical modeling, and Bayesian inference. These tools are used to learn selectional preferences from examples in a corpus. Instead of simple sets of semantic classes, selectional preferences are viewed as probability distributions over various entities. We survey research that extends Resnik's initial work, discuss the strengths and weaknesses of each approach, and show how they together form a cohesive line of research.     

The role of discourse context in developing word form representations: a paradoxical relation between reading and learning

To acquire representations of printed words, children must attend to the written form of a word and link this form with the word's pronunciation. When words are read in context, they may be read with less attention to these features, and this can lead to poorer word form retention. Two experiments with young children (ages 5-8 years) confirmed this hypothesis. In our experiments, children attempted to read words they could not previously read, during a self-teaching period, either in context or in isolation. Later they were tested on how well they learned the words as a function of self-teaching condition (isolation or context). Consistent with previous research, children read more words accurately in context than in isolation during self-teaching; however, children had better retention for words learned in isolation. Furthermore, this benefit from learning in isolation was larger for less skilled readers. This effect of poorer word retention when words are learned in context is paradoxical because context has been shown to facilitate word identification. We discuss factors that may influence this effect of context, especially the role of children's skill level and the demands of learning new word representations at the beginning of reading instruction.     

Philosophy and language learning

In this paper, I explore different ways of picturing language learning in philosophy, all of them inspired by Wittgenstein and all of them concerned about scepticism of meaning. I start by outlining the two pictures of children and language learning that emerge from Kripke’s famous reading of Wittgenstein. Next, I explore how social-pragmatic readings, represented by Meredith Williams, attempt to answer the sceptical anxieties. Finally, drawing somewhat on Stanley Cavell, I try to resolve these issues by investigating what characteristically happens to our view of language learning when we do philosophy. The focus throughout is on the relation between the individual (the learning child) and the community (usually represented by the parents), and how that relation is deformed when we operate with a certain philosophical notion of ground.     

The plurality of concepts

Traditionally, theories of concepts in psychology assume that concepts are a single, uniform kind of mental representation. But no single kind of representation can explain all of the empirical data for which concepts are responsible. I argue that the assumption that concepts are uniformly the same kind of mental structure is responsible for these theories’ shortcomings, and outline a pluralist theory of concepts that rejects this assumption. On pluralism, concepts should be thought of as being constituted by multiple representational kinds, with the particular kind of concept used on an occasion being determined by the context. I argue that endorsing pluralism does not lead to eliminativism about concepts as an object of scientific interest.     

The effects of bilingualism on toddlers' executive functioning

Bilingual children have been shown to outperform monolingual children on tasks measuring executive functioning skills. This advantage is usually attributed to bilinguals’ extensive practice in exercising selective attention and cognitive flexibility during language use because both languages are active when one of them is being used. We examined whether this advantage is observed in 24-month-olds who have had much less experience in language production. A battery of executive functioning tasks and the cognitive scale of the Bayley test were administered to 63 monolingual and bilingual children. Native bilingual children performed significantly better than monolingual children on the Stroop task, with no difference between groups on the other tasks, confirming the specificity of bilingual effects to conflict tasks reported in older children. These results demonstrate that bilingual advantages in executive control emerge at an age not previously shown.     

Using Predictability for Lexical Segmentation

This study investigates a strategy based on predictability of consecutive sub‐lexical units in learning to segment a continuous speech stream into lexical units using computational modeling and simulations. Lexical segmentation is one of the early challenges during language acquisition, and it has been studied extensively through psycholinguistic experiments as well as computational methods. However, despite strong empirical evidence, the explicit use of predictability of basic sub‐lexical units in models of segmentation is underexplored. This paper presents an incremental computational model of lexical segmentation for exploring the usefulness of predictability for lexical segmentation. We show that the predictability cue is a strong cue for segmentation. Contrary to earlier reports in the literature, the strategy yields state‐of‐the‐art segmentation performance with an incremental computational model that uses only this particular cue in a cognitively plausible setting. The paper also reports an in‐depth analysis of the model, investigating the conditions affecting the usefulness of the strategy.     

iMinerva: A Mathematical Model of Distributional Statistical Learning

Statistical learning refers to the ability to identify structure in the input based on its statistical properties. For many linguistic structures, the relevant statistical features are distributional: They are related to the frequency and variability of exemplars in the input. These distributional regularities have been suggested to play a role in many different aspects of language learning, including phonetic categories, using phonemic distinctions in word learning, and discovering non‐adjacent relations. On the surface, these different aspects share few commonalities. Despite this, we demonstrate that the same computational framework can account for learning in all of these tasks. These results support two conclusions. The first is that much, and perhaps all, of distributional statistical learning can be explained by the same underlying set of processes. The second is that some aspects of language can be learned due to domain‐general characteristics of memory.     

The paradox of the universal triangle: Concepts,language, and prototypes

For over 300 years, the humble triangle has served as the paradigmatic example of the problem of abstraction. How can we have the idea of a general triangle even though every experience with triangles is with specific ones? Classical cognitive science seemed to provide an answer in symbolic representation. With its easily enumerated necessary and sufficient conditions, the triangle would appear to be an ideal candidate for being represented in a symbolic form. I show that it is not. Across a variety of tasks—drawing, speeded recognition, unspeeded visual judgments, and inference—representations of triangles appear to be graded and context dependent. I show that using the category name “triangle” activates a more prototypical representation than using an arguably coextensive cue, “three-sided polygon”. For example, when asked to draw “triangles” people draw more typical triangles than when asked to draw “three-sided polygons”. Altogether, the results support the view that (even formal) concepts have a graded and flexible structure, which takes on a more prototypical and stable form when activated by category labels.     

The symbolic implications of vowels and of their orthographic representations in two natural languages

Ninety-two Hebrew-speaking subjects judged the magnitude, brightness, and hardness symbolism of orthographic characters designating five vowel phonemes in Hindi and in Japanese. For both languages and all three symbolic dimensions, the figural symbolism of the orthographic characters was found to replicate very closely the sound symbolism of their phonemic referents. The ranking of the five vowel characters in order of increasing magnitude and decreasing brightness and hardness was as follows:i, e, a, u, o. The results were interpreted to suggest that sound patterns and visual patterns tend to carry cross-culturally consistent connotations, and that the symbolic implications of sounds have been embodied in the pattern of orthographic characters in natural languages.This study was supported by a grant from the Human Development Center, The Hebrew University.     

Change one can believe in: Adding learning to computational models of self-regulation

Theories of self-regulation describe motivation as a dynamic process of goal choice and goal striving. To facilitate those processes, individuals learn about themselves and their environment, which is an internal dynamic process. However, the precise nature of the relationship between these learning and motivational processes is not well specified. This article integrates formal models of learning, goal choice, and goal striving using a single information processing structure found in self-regulatory models of motivation. Results from two published studies ( and ) validate the model. In both cases, the integrated model accounts for findings that previous theories of self-regulation could not explain. Discussion focuses on additional tests to validate the model and on the value of incorporating formal models from the cognitive, learning, and motivational literatures to account for behavior in complex settings and over time.     

Lexical effects on compensation for coarticulation: a tale of two systems?

We reply to McQueen's commentary by comparing the parsimony of his account of relevant data and the computational model he favors with the explanation and model we favor. His account requires multiple independent explanations and mechanisms. Ours requires one: lexical feedback.     

The approximate number system is not predictive for symbolic number processing in kindergarteners

The relation between the approximate number system (ANS) and symbolic number processing skills remains unclear. Some theories assume that children acquire the numerical meaning of symbols by mapping them onto the preexisting ANS. Others suggest that in addition to the ANS, children also develop a separate, exact representational system for symbolic number processing. In the current study, we contribute to this debate by investigating whether the nonsymbolic number processing of kindergarteners is predictive for symbolic number processing. Results revealed no association between the accuracy of the kindergarteners on a nonsymbolic number comparison task and their performance on the symbolic comparison task six months later, suggesting that there are two distinct representational systems for the ANS and numerical symbols.     

Concepts as Semantic Pointers: A Framework and Computational Model

The reconciliation of theories of concepts based on prototypes, exemplars, and theory‐like structures is a longstanding problem in cognitive science. In response to this problem, researchers have recently tended to adopt either hybrid theories that combine various kinds of representational structure, or eliminative theories that replace concepts with a more finely grained taxonomy of mental representations. In this paper, we describe an alternative approach involving a single class of mental representations called “semantic pointers.” Semantic pointers are symbol‐like representations that result from the compression and recursive binding of perceptual, lexical, and motor representations, effectively integrating traditional connectionist and symbolic approaches. We present a computational model using semantic pointers that replicates experimental data from categorization studies involving each prior paradigm. We argue that a framework involving semantic pointers can provide a unified account of conceptual phenomena, and we compare our framework to existing alternatives in accounting for the scope, content, recursive combination, and neural implementation of concepts.     

Aging and the use of context in ambiguity resolution: complex changes from simple slowing

Older and younger adults' abilities to use context information rapidly during ambiguity resolution were investigated. In Experiments 1 and 2, younger and older adults heard ambiguous words (e.g., fires) in sentences where the preceding context supported either the less frequent or more frequent meaning of the word. Both age groups showed good context use in offline tasks, but only young adults demonstrated rapid use of context in cross-modal naming. A 3rd experiment demonstrated that younger and older adults had similar knowledge about the contexts used in Experiments 1 and 2. The experiment results were simulated in 2 computational models in which different patterns of context use were shown to emerge from varying a single speed parameter. These results suggest that age-related changes in processing efficiency can modulate context use during language comprehension.