How do people know as much as they do with as little information as they get? The problem takes many forms; learning vocabulary from text is an especially dramatic and convenient case for research. A new general theory of acquired similarity and knowledge representation, latent semantic analysis (LSA), is presented and used to successfully simulate such learning and several other psycholinguistic phenomena. By inducing global knowledge indirectly from local co-occurrence data in a large body of representative text, LSA acquired knowledge about the full vocabulary of English at a comparable rate to schoolchildren. LSA uses no prior linguistic or perceptual similarity knowledge; it is based solely on a general mathematical learning method that achieves powerful inductive effects by extracting the right number of dimensions (e.g., 300) to represent objects and contexts. Relations to other theories, phenomena, and problems are sketched. Prologue "How much do we know at any time? Much more, or so I believe, than we know we know!" —Agatha Christie, The Moving Finger A typical American seventh grader knows the meaning of

The Nature-Nurture controversy has been with us since it was first outlined by Plato and Aristotle. Nobody likes it anymore. All reasonable scholars today agree that genes and environment interact to determine complex cognitive outcomes. So why does the controversy persist? First, it persists because it has practical implications that cannot be postponed (i.e., what can we do to avoid bad outcomes and insure better ones?), a state of emergency that sometimes tempts scholars to stake out claims they cannot defend. Second, the controversy persists because we lack a precise, testable theory of the process by which genes and environment interact. In the absence of a better theory, innateness is often confused with (1) domain specificity (Outcome X is so peculiar that it must be innate), (2) species specificity (we are the only species who do X, so X must lie in the human genome), (3) localization (Outcome X is mediated by a particular part of the brain, so X must be innate), and (4) learnability (we cannot figure out how X could be learned, so X must be innate). We believe that an explicit and plausible theory of interaction is now around the corner, and that many of the classic maneuvers to defend or attack innateness will soon disappear. In the interim, some serious errors can be avoided if we keep these confounded issues apart. That is the major goal of this paper, i.e., not to attack innateness but to clarify what

Abstract not found

The paper proposes a number of models to examine through what mech-anisms a population of autonomous agents could arrive at a repertoire of perceptually grounded categories that is sufficiently shared to allow successful communication. The models are inspired by the main approaches to human categorisation being discussed in the literature: nativism, empiricism, and culturalism. Colour is taken as a case study. Although the paper takes no stance on which position is to be accepted as final truth with respect to hu-man categorisation and naming, it points to theoretical constraints that make each position more or less likely and contains clear suggestions on what the best engineering solution would be. Specifically, it argues that the collective choice of a shared repertoire must integrate multiple constraints, including constraints coming from communication.

This paper develops new economic models to synthesize and interpret the empirical literature on human skill formation. It presents simple formal economic models of child development that capture the essence of recent findings from the empirical literature. The goal of this essay is to provide a theoretical framework for interpreting the evidence from a vast empirical literature, for guiding the next generation of empirical studies, and for formulating policy. Central to our analysis is the concept that childhood has more than one stage, and that policies need to be tailored to each one. We formalize the concepts of self productivity and complementarity of human capital investments and use them to explain the evidence on skill formation. Together, they explain why skill begets skill through a skill multiplier process. Skill formation is a lifecycle process. It starts in the womb and goes on throughout most of the adult life. Families play a

Abstract not found

this paper is on procedures whereby new (e.g., juvenile) members of a population could learn to communicate with the other members by observing their communicative behavior. Two apparently distinct issues are relevant to the evaluation of such learning procedures. First, the procedure must enable the new members to accurately acquire the communication system of the population, even though their observations may be limited, noisy, or otherwise misleading. Second, the learning procedure used by its new members will affect the population's communication system over time. The use of a particular procedure might result in the population's communication increasing in coordination, ultimately yielding a nearly optimally coordinated system. If a learning procedure were to satisfy both criteria, it could explain how learned communication systems are maintained over time, as well as how they are established in the first place.

Human infants begin to acquire their native language in the first months of life. To determine which brain regions support language processing at this young age, we measured with functional magnetic resonance imaging the brain activity evoked by normal and reversed speech in awake and sleeping 3-month-old infants. Left-lateralized brain regions similar to those of adults, including the superior temporal and angular gyri, were already active in infants. Additional activation in right prefrontal cortex was seen only in awake infants processing normal speech. Thus, precursors of adult cortical language areas are already active in infants, well before the onset of speech production. The adult human brain exhibits anatomical and functional specialization for speech processing (1–5). To understand this adult organization, one must ultimately clarify how it emerges in the course of development through a combination of brain maturation

This paper reviews our recent work on applying inductive logic programming to the construction of natural language processing systems. We have developed a system, Chill, that learns a parser from a training corpus of parsed sentences by inducing heuristics that control an initial overly-general shift-reduce parser. Chill learns syntactic parsers as well as ones that translate English database queries directly into executable logical form. The ATIS corpus of airline information queries was used to test the acquisition of syntactic parsers, and Chill performed competitively with recent statistical methods. English queries to a small database on U.S. geography were used to test the acquisition of a complete natural language interface, and the parser that Chill acquired was more accurate than an existing hand-coded system. The paper also includes a discussion of several issues this work has raised regarding the capabilities and testing of ILP systems as well as a summary of our current research directions.

Selective deficits in aphasics patients ’ grammatical production and comprehension are often cited as evidence that syntactic processing is modular and localizable in discrete areas of the brain (e.g., Y. Grodzinsky, 2000). The authors review a large body of experimental evidence suggesting that morphosyntactic deficits can be observed in a number of aphasic and neurologically intact populations. They present new data showing that receptive agrammatism is found not only over a range of aphasic groups, but is also observed in neurologically intact individuals processing under stressful conditions. The authors suggest that these data are most compatible with a domain-general account of language, one that emphasizes the interaction of linguistic distributions with the properties of an associative processor working under normal or suboptimal conditions. The primary purpose of this article is to provide empirical arguments in support of a new view of language deficits and their neural correlates, particularly in the realm of syntax. Selective syntactic deficits are often cited as evidence that the human brain contains a bounded and well-defined faculty or module dedicated exclusively to the representation and/or processing of syntax (Caplan & Waters, 1999; Grodzinsky, 1995a,