Bayesian treatments of animal conditioning start from a generative model that specifies precisely a set of assumptions about the structure of the learning task. Optimal rules for learning are direct mathematical consequences of these assumptions. In terms of Marr’s (1982) levels of analyses, the main task at the computational level

Many evaluations of cognitive models rely on data that have been averaged or aggregated across all experimental subjects, and so fail to consider the possibility that there are important individual di#erences between subjects. Other evaluations are done at the single-subject level, and so fail to benefit from the reduction of noise that data averaging or aggregation potentially provides. To overcome these weaknesses, we develop a general approach to modeling individual di#erences using families of cognitive models, where di#erent groups of subjects are identified as having di#erent psychological behavior. Separate models with separate parameterizations are applied to each group of subjects, and Bayesian model selection is used to determine the appropriate number of groups. We demonstrate the general approach in a concrete and detailed way using the ALCOVE model of category learning, and data from four previously analysed category learning experiments. Meaningful individual di#erences are found for three of the four experiments, and ALCOVE is able to account for this variation through psychologically interpretable di#erences in parameterization.

Abstract not found

A fundamental aspect of human intelligence is the ability to represent and reason about relations. Examples of relational thinking include our ability to appreciate analogies between different objects or events (Gentner, 1983; Holyoak & Thagard, 1995), our ability to apply abstract rules in novel situations (e.g., Smith, Langston & Nisbett, 1992), our ability to understand and learn language (e.g., Kim, Pinker, Prince & Prasada, 1991), our

Classification “rules” in expert and everyday discourse are usually deficient by formal standards, lacking explicit decision procedures and precise terms. The authors argue that a central function of such weak rules is to focus on perceptual learning rather than to provide definitions. In 5 experiments, transfer following learning of family resemblance categories was influenced more by familiar-appearing features than by novel-appearing features equally acceptable under the rule. This occurred both when rules were induced and when rules were given at the beginning of instruction. To model this and other phenomena in categorization, features must be represented on 2 levels: informational and instantiated. These 2 feature levels are crucial to provide broad generalization while reflecting the known peculiarities of a complex world.

A growing number of studies involving nonlinear discrimination problems suggests that stimuli in human associative learning are represented configurally with narrow generalization, such that presentation of stimuli that are even slightly dissimilar to stored configurations weakly activate these configurations. We note, however, that another well-known set of findings in human associative learning, cue-interaction phenomena, suggest relatively broad generalization. Three experiments show that current models of human associative learning, which try to model both non-linear discrimination and cue interaction as the result of one process, fail because they cannot simultaneously account for narrow and broad generalization. The results suggest that human associative learning involves (a) an exemplar-based process with configural stimulus representation and narrow generalization and (b) an adaptive learning process characterized by broad generalization and cue interaction. 2 Stimulus Generalization

available at www.sciencedirect.com www.elsevier.com/locate/brainres

Abstract. We demonstrate that selective attention can improve learning. Considerably fewer samples are needed to learn a source separation problem when the inputs are pre-segmented by the proposed model. The model combines biased-competition model for attention with a habituation mechanism which allows the focus of attention to switch from one object to another. The criteria for segmenting objects are estimated from data and are shown to generalise to new objects.

Research in attention aware systems, i.e. systems that support users in their attentional choices, promises to address many of the problems related to information overload, cyber collaboration, and mobility, by providing features helping users in coping with attentional limitations. However, the design of attention aware systems necessitates a deep understanding of human attentional processes, of the knowledge a system needs to support those processes, and of the manner in which such knowledge may be acquired. Because the conceptualization of such systems requires understanding users ' cognitive states in terms of their past interactions with the environment, the interactivist framework may provide a strong basis for analyzing attention aware systems. This paper briefly introduces the services that attention aware systems may provide, suggests how attention may be modeled within the interactivist framework, and proposes nine questions that may be answered within such framework to gain the knowledge necessary for the creation of attention aware systems.

scaffolding