An approach to connectionist natural language processing is proposed, which is based on hierarchically organized modular Parallel Distributed Processing (PDP) networks and a central lexicon of distributed input/output representations. The modules communicate using these representations, which are global and publicly available in the system. The representations are developed automatically by all networks while they are learning their processing tasks. The resulting representations reflect the regularities in the subtasks, which facilitates robust processing in the face of noise and damage, supports improved generalization, and provides expectations about possible contexts. The lexicon can be extended by cloning new instances of the items, that is, by generating a number of items with known processing properties and distinct identities. This technique combinatorially increases the processing power of the system. The recurrent FGREP module, together with a central lexicon, is used as a ba...

We present a framework for the rational analysis of elemental causal induction – learning about the existence of a relationship between a single cause and effect – based upon causal graphical models. This framework makes precise the distinction between causal structure and causal strength: the difference between asking whether a causal relationship exists and asking how strong that causal relationship might be. We show that two leading rational models of elemental causal induction, ∆P and causal power, both estimate causal strength, and introduce a new rational model, causal support, that assesses causal structure. Causal support predicts several key phenomena of causal induction that cannot be accounted for by other rational models, which we explore through a series of experiments. These phenomena include the complex interaction between ∆P and the base-rate probability of the effect in the absence of the cause, sample size effects, inferences from incomplete contingency tables, and causal learning from rates. Causal support also provides a better account of a number of existing datasets than either ∆P or causal power.

What counts as human rationality: reasoning processes that embody content-independent formal theories, such as propositional logic, or reasoning processes that are well designed for solving important adaptive problems? Most theories of human reasoning have been based on content-independent formal rationality, whereas adaptive reasoning, ecological or evolutionary, has been little explored. We elaborate and test an evolutionary approach, Cosmides’ (1989) social contract theory, using the Wason selection task. In the first part, we disentangle the theoretical concept of a “social contract” from that of a “cheater-detection algorithm.” We demonstrate that the fact that a rule is perceived as a social contract—or a conditional permission or obligation, as Cheng and Holyoak (1985) proposed—is not sufficient to elicit Cosmides’ striking results, which we replicated. The crucial issue is not semantic (the meaning of the rule), but pragmatic: whether a person is cued into the perspective of a party who can be cheated. In the second part, we distinguish between social contracts with bilateral and unilateral cheating options. Perspective change in contracts with bilateral cheating options turns P & not-Q responses into not-P & Q responses. The results strongly support social contract theory, contradict availability theory, and cannot be accounted for by pragmatic reasoning schema theory, which lacks the pragmatic concepts of perspectives and cheating detection.

Inducing causal relationships from observations is a classic problem in scientific inference, statistics, and machine learning. It is also a central part of human learning, and a task that people perform remarkably well given its notorious difficulties. People can learn causal structure in various settings, from diverse forms of data: observations of the co-occurrence frequencies between causes and effects, interactions between physical objects, or patterns of spatial or temporal coincidence. These different modes of learning are typically thought of as distinct psychological processes and are rarely studied together, but at heart they present the same inductive challenge—identifying the unobservable mechanisms that generate observable relations between variables, objects, or events, given only sparse and limited data. We present a computational-level analysis of this inductive problem and a framework for its solution, which allows us to model all these forms of causal learning in a common language. In this framework, causal induction is the product of domain-general statistical inference guided by domain-specific prior knowledge, in the form of an abstract causal theory. We identify 3 key aspects of abstract prior knowledge—the ontology of entities, properties, and relations that organizes a domain; the plausibility of specific causal relationships; and the functional form of those relationships—and show how they provide the constraints that people need to induce useful causal models from sparse data.

This paper presents a case study of 3- to 9-year-old children's concepts of size, weight, density, matter, and material kind. Our goal was to examine two claims: (1) that individual concepts undergo differentiation during development; and (2) that young children's concepts are embedded in theory-like structures. To make progress on the first issue, we needed to specify in representational terms what an undifferentiated concept is like and in what sense this undifferentiated concept is a parent of the more differentiated concepts. Our strategy was to use a model of conceptual differentiation suggested by the history of science to guide our search for evidence. In this model, undifferentiated concepts, like differentiated concepts, can be analyzed in terms of their component properties, features, or dimensions. The key difference is that an undifferentiated concept unites certain components which will subsequently be analyzed as components of distinct concepts, and that the undifferentiated concept is embedded in a different theoretical structure from the differentiated concepts. In our study, the same group of 78 children (18 3-year-olds, 18 4-year-olds, 18 5-year-olds, 12 6-7-year-olds, and 12 8-9-year-olds) were given a range of tasks probing their understanding of size, weight, and density; a

Parallel distributed processing (PDP) models provide a rich set of resources for exploring issues of nature, nurture and their interaction in cognition development. I present the essential aspects of the PDP(orconnectionist) framework, and I draw parallels between the child as learner and the mechanisms of learning in connectionist systems. The remaining sections discuss some of the implications of this framework for our understanding of the acquisition of knowledge. I point outlhat many lines ofargumenllhal have typically been given in support ofnalivist approaches need to be reconsidered in the light of the characteristics of PDP models of learning and development. The first of these sections points out that connectionist models offer a dramatic advance over classical associationist approaches to learning. The second illustrates how stage-like progressions can be understood in terms of the typical learning trajectories seen in connectionist models. The third section considers the meaning and possible sources of early competence from- a PDP perspective, and the fourth considers how connectionist models may shed light on the fact that some of the structure of human behaviour appears to be imposed by the learner. In all, the chapter amounts to an argument that connectionist models allow us to see ways in which experience might lead to the rich and interesting cognitive structures and developmental progressions that have often been taken as supportive of nativist approaches.

Elicitation is a key task for subjectivist Bayesians. While skeptics hold that it cannot (or perhaps should not) be done, in practice it brings statisticians closer to their clients and subjectmatter-expert colleagues. This paper reviews the state-of-the-art, reflecting the experience of statisticians informed by the fruits of a long line of psychological research into how people represent uncertain information cognitively, and how they respond to questions about that information. In a discussion of the elicitation process, the first issue to address is what it means for an elicitation to be successful, i.e. what criteria should be employed? Our answer is that a successful elicitation faithfully represents the opinion of the person being elicited. It is not necessarily “true ” in some objectivistic sense, and cannot be judged that way. We see elicitation as simply part of the process of statistical modeling. Indeed in a hierarchical model it is ambiguous at which point the likelihood ends and the prior begins. Thus the same kinds of judgment that inform statistical modeling in general also inform elicitation of prior distributions.

Much research on judgment and decision making has focussed on the adequacy of classical rationality as a description of human reasoning. But more recently it has been argued that classical rationality should also be rejected even as normative standards for human reasoning. For example, Gigerenzer and Goldstein (1996) and Gigerenzer and Todd (1999a) argue that reasoning involves ‘‘fast and frugal’ ’ algorithms which are not justified by rational norms, but which succeed in the environment. They provide three lines of argument for this view, based on: (A) the importance of the environment; (B) the existence of cognitive limitations; and (C) the fact that an algorithm with no apparent rational basis, Take-the-Best, succeeds in an judgment task (judging which of two cities is the larger, based on lists of features of each city). We reconsider (A)–(C), arguing that standard patterns of explanation in psychology and the social and biological sciences, use rational norms to explain why simple cognitive algorithms can succeed. We also present new computer simulations that compare Take-the-Best with other cognitive models (which use connectionist, exemplarbased, and decision-tree algorithms). Although Take-the-Best still performs well, it does not perform noticeably better than the other models. We conclude that these results provide no strong reason to prefer Take-the-Best over alternative cognitive models.

How have connectionist models informed the study of development? This paper considers three contributions from specific models. First, connectionist models have proven useful for exploring nonlinear dynamics and emergent properties, and their role in nonlinear developmental trajectories, critical periods and developmental disorders. Second, connectionist models have informed the study of the representations that lead to behavioral dissociations. Third, connectionist models have provided insight into neural mechanisms, and why different brain regions are specialized for different functions. Connectionist and dynamic systems approaches to development have differed, with connectionist approaches focused on learning processes and representations in cognitive tasks, and dynamic systems approaches focused on mathematical characterizations of physical elements of the system and their interactions with the environment. The two approaches also share much in common, such as their emphasis on continuous, nonlinear processes and their broad application to a range of behaviors.

States Government. To the people, especially those in China. This thesis is a study of “learning from the environment. ” As machine learning moves from toy environments towards the real world, the problem of learning autonomously from environments whose structure is not completely defined a priori becomes ever more critical. Three of the major challenges are: (1) how to integrate various high level AI techniques such as exploration, problem solving, learning, and experimentation with low level perceptions and actions so that learning can be accomplished through interactions with the environment; (2) how to acquire new knowledge of the environment and to learn from mistakes autonomously, permitting incorrect information to be identified and corrected; (3) how to create features in such a way that knowledge acquisition is not limited by the initial concept description language provided by the designers of a system. The thesis defines learning from the environment as the process of inferring the laws of the environment that enable the learner to solve problems. The inputs to the learning are goals to be achieved, percepts and actions sensed and effected by the learner, and