Abstract. In this paper three problems for a connectionist account of language are considered: 1. What is the nature of linguistic representations? 2. How can complex structural relationships such as constituent structure be represented? 3. How can the apparently open-ended nature of language be accommodated by a fixed-resource system? Using a prediction task, a simple recurrent network (SRN) is trained on multiclausal sentences which contain multiply-embedded relative clauses. Principal component analysis of the hidden unit activation patterns reveals that the network solves the task by developing complex distributed representations which encode the relevant grammatical relations and hierarchical constituent structure. Differences between the SRN state representations and the more traditional pushdown store are discussed in the final section.

Abstract: Human agents draw a variety of inferences effortlessly, spontaneously, and with remarkable efficiency — as though these inferences are a reflex response of their cognitive apparatus. Furthermore, these inferences are drawn with reference to a large body of background knowledge. This remarkable human ability seems paradoxical given the results about the complexity of reasoning reported by researchers in artificial intelligence. It also poses a challenge for cognitive science and computational neuroscience: How can a system of simple and slow neuron-like elements represent a large body of systematic knowledge and perform a range of inferences with such speed? We describe a computational model that is a step toward addressing the cognitive science challenge and resolving the artificial intelligence paradox. We show how a connectionist network can encode millions of facts and rules involving n-ary predicates and variables, and perform a class of inferences in a few hundred msec. Efficient reasoning requires the rapid representation and propagation of dynamic bindings. Our model achieves this by i) representing dynamic bindings as the synchronous firing of appropriate nodes, ii) rules as interconnection patterns

We are quickly passing through the historical moment when people work in front of a single computer, dominated by a small CRT and focused on tasks involving only local information. Networked computers are becoming ubiquitous and are playing increasingly significant roles in our lives and in the basic infrastructures of science, business, and social interaction. For human-computer interaction to advance in the new millennium we need to better understand the emerging dynamic of interaction in which the focus task is no longer confined to the desktop but reaches into a complex networked world of information and computer-mediated interactions. We think the theory of distributed cognition has a special role to play in understanding interactions between people and technologies, for its focus has always been on whole environments: what we really do in them and how we coordinate our activity in them. Distributed cognition provides a radical reorientation of how to think about designing and supporting human-computer interaction. As a theory it is specifically tailored to understanding interactions among people and technologies. In this article we propose distributed cognition as a new foundation for human-computer interaction, sketch an integrated research framework, and use selections from our earlier work to suggest how this framework can provide new opportunities in the design of digital work materials.

Cross-cultural research on facial expression and the developments of methods to measure facial expression are briefly summarized. What has been learned about emotion from this work on the face is then elucidated. Four questions about facial expression and emotion are discussed. What information does an expression typically convey? Can there be emotion without facial expression? Can there be a facial expression of emotion without emotion? How do individuals differ in their facial expressions of emotion? In 1965 when 1 began to study facial expression, 1 few thought there was much to be learned. Goldstein (1981) pointed out that a number of famous psychologists—F. and G. Allport, Brunswik, Hull, Lindzey, Maslow, Osgood, Titchner—did only one facial study, which was not what earned them their reputations. Harold Schlosberg was an exception, but he was more interested in how to represent the information derived by those who observed the face than in expression itself. 2 The face was considered a meager source of mostly inaccurate, culturespecific, stereotypical information (Bruner & Tagiuri, 1954). That this contradicted what every layman knew made it all the more attractive. Psychology had exposed the falseness of a folk belief, a counterintuitive finding.

Abstract. To explore issues of developmental structure, physical embodiment, integration of multiple sensory and motor systems, and social interaction, we have constructed an upper-torso humanoid robot called Cog. The robot has twenty-one degrees of freedom and a variety of sensory systems, including visual, auditory, vestibular, kinesthetic, and tactile senses. This chapter gives a background on the methodology that we have used in our investigations, highlights the research issues that have been raised during this project, and provides a summary of both the current state of the project and our long-term goals. We report on a variety of implemented visual-motor routines (smooth-pursuit tracking, saccades, binocular vergence, and vestibular-ocular and opto-kinetic reflexes), orientation behaviors, motor control techniques, and social behaviors (pointing to a visual target, recognizing joint attention through face and eye finding, imitation of head nods, and regulating interaction through expressive feedback). We further outline a number of areas for future research that will be necessary to build a complete embodied system. 1

We report a new phenomenon associated with language comprehension: the action–sentence compatibility effect (ACE). Participants judged whether sentences were sensible by making a response that required moving toward or away from their bodies. When a sentence implied action in one direction (e.g., “Close the drawer ” implies action away from the body), the participants had difficulty making a sensibility judgment requiring a response in the opposite direction. The ACE was demonstrated for three sentences types: imperative sentences, sentences describing the transfer of concrete objects, and sentences describing the transfer of abstract entities, such as “Liz told you the story. ” These data are inconsistent with theories of language comprehension in which meaning is represented as a set of relations among nodes. Instead, the data support an embodied theory of meaning that relates the meaning of sentences to human action. How language conveys meaning remains an open question. The dominant approach is to treat language as a symbol manipulation system: Language conveys meaning by using abstract, amodal, and arbitrary symbols (i.e., words) combined by syntactic rules (e.g., Burgess & Lund, 1997; Chomsky, 1980; Fodor, 2000; Kintsch, 1988; Pinker, 1994). Words are abstract in that the same word, such as “chair, ” is used for big chairs and little chairs, words are amodal in that the same word is used when chairs are spoken about or written about, and words are arbitrarily related to their referents in that the phonemic and orthographic characteristics of a word bear no relationship to the physical or functional characteristics of the word’s referent. An alternative view is that linguistic meaning is

In this paper we discuss various aspects of systematic or conventional polysemy and their formal treatment within an implemented constraint based approach to linguistic representation. We distinguish between two classes of systematic polysemy: constructional polysemy, where a single sense assigned to a lexical entry is contextually specialised, and sense extension, which predictably relates two or more senses. Formally the rst case is treated as instantiation of an underspecied lexical entry and the second by use of lexical rules. The problems of distinguishing between these two classes are discussed in detail. We illustrate how lexical rules can be used both to relate fully conventionalised senses and also applied productively to recognise novel usages and how this process can be controlled to account for semi-productivity by utilising probabilities. 1 Introduction Discussion of polysemy has been central to much recent work on lexical semantics. Most of the arguments for (or again...

Any agent in the real world has to be able to make distinctions between different types of objects, i.e. it must have the competence of categorization. In mobile agents, there is a large variation in proximal sensory stimulation originating from the same object. Therefore, categorization behavior is hard to achieve, and the successes in the past in solving this problem, have been limited. In this paper it is proposed that the problem of categorization in the real world is significantly simplified if it is viewed as one of sensory-motor coordination, rather than one of information processing happening "on the input side". A series of models is presented to illustrate the approach. It is concluded that we should consider replacing the metaphor of information processing for intelligent systems by the one of sensory-motor coordination. But the principle of sensory-motor coordination is more than a metaphor. It offers concrete mechanisms for putting agents to work in the real world. These i...

model meaning as the relations among abstract symbols that are arbitrarily related to what they signify. These symbols are ungrounded in that they are not tied to perceptual experience or action. Because the symbols are ungrounded, they cannot, in principle, capture the meaning of novel situations. In contrast, participants in three experiments found it trivially easy to discriminate between descriptions of sensible novel situations (e.g., using a newspaper to protect one’s face from the wind) and nonsense novel situations (e.g., using a matchbook to protect one’s face from the wind). These results support the Indexical Hypothesis that the meaning of a sentence is constructed by (a) indexing words and phrases to real objects or perceptual, analog symbols; (b) deriving affordances from the objects and symbols; and (c) meshing the affordances under the guidance of syntax. © 2000 Academic Press Key Words: meaning; language; embodiment; computational models; Latent Semantic Analysis; Hyperspace Analogue to Language. Meaning is the most important problem in cognitive psychology. Meaning controls memory and perception. Meaning is the goal of communication. Meaning underlies social activities and culture: To a great degree, what distinguishes human cultures are the meanings they give to natural phenomena, artifacts, and human relations. Yet, rather than being a hotbed of theoretical and empirical investigation, meaning in cognitive psychology has been coopted by a particular approach: Meaning arises

We present a novel methodology for building humanlike artificially intelligent systems. We take as a model the only existing systems which are universally accepted as intelligent: humans. We emphasize building intelligent systems which are not masters of a single domain, but, like humans, are adept at performing a variety of complex tasks in the real world. Using evidence from cognitive science and neuroscience, we suggest four alternative essences of intelligence to those held by classical AI. These are the parallel themes of development, social interaction, embodiment, and integration. Following a methodology based on these themes, we have built a physical humanoid robot. In this paper we present our methodology and the insights it affords for facilitating learning, simplifying the computation underlying rich behavior, and building systems that can scale to more complex tasks in more challenging environments.