WordNet is an on-line lexical reference system whose design is inspired by current

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.

A series of experiments sought to replicate and extend the classic studies of Rosch Heider on the Dani with a comparable group from Papua New Guinea who speak Berinmo, which has 5 basic color terms. Her results have been interpreted as clearly supporting universal color categories. Some results could, however, be interpreted as supporting linguistic relativity. We investigated naming and memory for highly saturated `focal', `non-focal' and low saturation stimuli from around the color space. Recognition of desaturated colors did appear to reflect color vocabulary. When response bias was controlled, there was no recognition advantage for focal stimuli. Pairedassociate learning also failed to show an advantage for focal stimuli. We further examined `Categorical Perception' effects at the boundaries of both English and Berinmo linguistic categories. These were found, in both populations, only for existing linguistic categories. Whilst Berinmo speakers, like those of all other languages hit...

Various revisions of the Berlin and Kay (1969) model of the evolution of basic color term systems have been produced in the last thirty years, motivated by both empirical and theoretical considerations. On the empirical side, new facts about color naming systems have continually come to light, which have demanded adjustments in the descriptive model. On the theoretical side, there has been a sustained effort to find motivation in the vision science literature regarding color appearance for the synchronic and diachronic constraints observed to govern color terminology systems. The present paper continues the pursuit of both of these goals. A new empirical question is addressed with data from the World Color Survey (WCS) and a revised model is proposed, which both responds to recently raised empirical questions and provides new motivation from the field of color vision for the observed constraints on color naming. Color Appearance and the Emergence and Evolution of Basic Color Lexicons...

This study investigated whether different lexicalization patterns of motion events in English and Spanish predict how speakers of these languages perform in non-linguistic tasks. Using 36 motion events, we compared English and Spanish speakers' linguistic descriptions to their performance on two non-linguistic tasks: recognition memory and similarity judgments. We investigated the effect of language processing on non-linguistic performance by varying the nature of the encoding before testing for recognition and similarity. Participants encoded the events while describing them verbally or not. No effect of language was obtained in the recognition memory task after either linguistic or non-linguistic encoding and in the similarity task after non-linguistic encoding. We did find a linguistic effect in the similarity task after verbal encoding, an effect that conformed to languagespecific patterns. Linguistic descriptions directed attention to certain aspects of the events later used to make a non-linguistic judgment. This suggests that linguistic and non-linguistic performance are dissociable, but language-specific regularities made available in the experimental context may mediate the speaker's performance in specific tasks.

conditioning, etc.). Piaget spelled out a list of "formal operations," such as modus ponens, the probability schema, etc., which he regarded as the fundamental deductive and inductive rule schemas necessary to understand the world. The cognitive revolution, from its earliest incarnation in the work of such theorists as George Miller and Herbert Simon, until nearly the end of the 20 century, essentially embraced Piaget's position of extreme formalism and content independence of inferential rules. Cognitive scientists' endorsement of the formalist, universalistic position was undoubtedly encouraged by the analogy between the human mind and the computer: brain = hardware, cognitive procedures = operating principles and factory-installed software (Block, 1995). This analogy both encouraged the universality assumption and discouraged any assumption that cognitive procedures might be alterable. The heuristics and biases movement of Kahneman and Tversky (1974) and their colleagues in social

Languages vary strikingly in how they encode motion events. In some languages (e.g. English), manner of motion is typically encoded within the verb, while direction of motion information appears in modifiers. In other languages (e.g. Greek), the verb usually encodes the direction of motion, while the manner information is often omitted, or encoded in modifiers. We designed two studies to investigate whether these language-specific patterns affect speakers' reasoning about motion. We compared the performance of English and Greek children and adults (a) in nonlinguistic (memory and categorization) tasks involving motion events, and (b) in their linguistic descriptions of these same motion events. Even though the two linguistic groups differed significantly in terms of their linguistic preferences, their performance in the nonlinguistic tasks was identical. More surprisingly, the linguistic descriptions given by subjects within language also failed to correlate consistently with their memory and categorization performance in the relevant regards. For the domain studied, these results are consistent with the view that conceptual development and organization are largely independent of language-specific labeling practices. The discussion emphasizes that the necessarily sketchy nature of language use assures that it will be at best a crude index of thought.

van de academische graad van doctor in de wetenschappen, in het openbaar te verdedigen op vrijdag 8 maart 2002. Acknowledgements I started as a research assistant in the Artificial Intelligence Laboratory in autumn 1996. My first interests were into behavioural robotics and robot ecosystems. As a continuation to my “licentiaats ” thesis I started building a camera system to extend the sensory perception of the lab’s robots (Belpaeme and Birk, 1997a,b; Belpaeme, 1998; Birk and Belpaeme, 1998; Birk et al., 1998, 1999; Belpaeme and Birk, 2001). It was around that time when Luc Steels got interested in the origins of language. His early experiments formed the seed for what is now one of the most important paradigms for exploring linguistic interactions with computer simulations. Luc soon wanted more and had plans to implement a language experiment in the real world, for which I delivered the visual perception (Belpaeme et al., 1998; Belpaeme, 1999). This got me interested in visual features, and my research

A variety of studies have demonstrated that organizing stimuli into categories can affect the way the stimuli are perceived. We explore the influence of categories on perception through one such phenomenon, the perceptual magnet effect, in which discriminability between vowels is reduced near prototypical vowel sounds. We present a Bayesian model to explain why this reduced discriminability might occur: It arises as a consequence of optimally solving the statistical problem of perception in noise. In the optimal solution to this problem, listeners’ perception is biased toward phonetic category means because they use knowledge of these categories to guide their inferences about speakers ’ target productions. Simulations show that model predictions closely correspond to previously published human data, and novel experimental results provide evidence for the predicted link between perceptual warping and noise. The model unifies several previous accounts of the perceptual magnet effect and provides a framework for exploring categorical effects in other domains.

The ability to categorize and use concepts e#ectively is a basic requirementofany intelligent actor. The utility-based approach to categorization is founded on the thesis that categorization is fundamentally in service of action, i.e., the choice of concepts made by an actor is critical to its choice of appropriate actions. This is in contrast to classical and similarity-based approaches which seek logical completeness in concept description with respect to sensory data rather than action-oriented e#ectiveness. Utility-based categorization is normative and not descriptive. It prescribes howanintelligent agent ought to conceptualize to act e#ectively. It provides ideals for categorization, speci#es criteria for the design of e#ective computational agents, and provides a model of ideal competence. A decision-theoretic framework for utilitybased categorization whichinvolves reasoning about alternative categorization models of varying levels of abstraction is proposed. Categorization mode...