Four more CEU CDC seminars

Upcoming seminars at the Cognitive Development Center of CEU: --------------------------------------------- Water Schroyens (Psychology, Gent & Leuven) at 10.00am on Monday, 1 March 2010 Title: Meaning and idealization in reasoning towards an interpretation of conditionals: Is there a singular specific meaning or are there a multitude of ephemeral interpretations of natural language connectives, or both? Abstract: The paper investigates the thesis that while the pragmatics of content and context can yield many interpretations, there is an idealized core- meaning for sentential connectives: People do not reason from this core meaning, but can reason towards a corresponding interpretation, i.e., the conditional interpretation akin to the much dis-reputed material implication of classic logic. In reasoning towards this conditional-interpretation of “if A then C”, the utterances are interpreted as meaning that all possibilities except the “A and not-c” contingency are possible. In idealizing towards the conditional interpretation as the core meaning 'if', theorists abstract, simplify and generalize across conditions. Six experiments show that when the context is idealized by taking account of cognitive processing hurdles and auxiliary hypothesis in the mental-models theory of reasoning (e.g., people tend not to throw away semantic information, they start reasoning on the basis of a minimal representation, they are sensitive to the principles of parsimony in positing theoretical entities, … ) people are more likely to reason towards a conditional interpretation. That is, the context induces people to reason towards a more idealized interpretation (which must not be an ideal interpretation). A series of developmental studies additionally indicates that with age (i.e., experience and education) people are more likely to reason towards the conditional interpretation and two individual-differences studies show that people higher in general ability are similarly more likely to reason towards the conditional interpretation. --------------------------------------------- Patrick Shafto (Psychology, Lousville) at 10.00am on Tuesday, 2 March 2010 Title: Computational modeling of pedagogical reasoning Abstract: Much of human learning goes on in social situations. Among these situations, pedagogical situations stand out as potentially the most important. In pedagogical situations, a person (a teacher) chooses data for the purpose of helping another person learn a concept. I will present a computational model of pedagogical data sampling, which formalizes the problem as complementary inferences on the parts of both the teacher and learner. I will present a series of experiments that test the model's predictions about reasoning with adults and children. I will conclude by sketching a larger picture, in which pedagogical sampling is a special case of reasoning about intentionally sampled data, and outline directions of future research in this context. --------------------------------------------- Afra Alishahi (Computational Linguistics, Saarland University) at 16.30pm on Wednesday, 3 March 2010 Title: Modeling different aspects of child language acquisition as a probabilistic process Abstract: I will first present a probabilistic model of word learning by children. A major source of disagreement among the different theories of word learning is whether children are equipped with special mechanisms and biases for word learning, or their general cognitive abilities are adequate for the task. I present a novel computational model of early word learning which learns word meanings as probabilistic associations between words and semantic elements, using an incremental learning mechanism, and drawing only on general cognitive abilities. The computational simulations of the model demonstrate that much about word meanings can be learned from naturally-occurring child-directed utterances (paired with meaning representations), without using any special biases or constraints, and without any explicit developmental changes in the underlying learning mechanism. Furthermore, our model provides explanations for the occasionally contradictory child experimental data, and offers predictions for the behaviour of young word learners in novel situations. Children use their knowledge of word meanings in order to learn the structural properties of the language. I will also present a probabilistic usage-based model of verb argument structure acquisition that can successfully learn abstract knowledge of language from instances of verb usage, and use this knowledge in various language tasks. The model further demonstrates the feasibility of a usage-based account of language learning, and provides concrete explanation for the observed patterns in child language acquisition. --------------------------------------------- József Fiser (Psychology, Brandeis) at 16.30pm on Monday, 8 March 2010 Title: Developing internal representations in the mind to understand the visual world Abstract: Arguably, the mind's internal representations of its environment have a crucial role in the emergence of intelligent behavior, yet there are few concrete proposals about the nature of these internal representations or the way they are acquired. Using the domain of visual recognition, I will present a framework and a combined empirical-computational program that explore these issues. First, I will demonstrate that in everyday perceptual tasks humans process not only the sensory information but also their uncertainty about that information, and they do this in a theoretically optimal manner. As such behavior assumes a probabilistic internal representation of the world, I present evidence that indeed, adults and infants develop such representations when they face a novel visual environment. I will show how this framework can cover not only low-level statistical regularities of events and features, but also more complex abstract “rules” of the sensory world, as well as internal models of higher cognitive functions. Next, I turn to the issue of whether the implementation of such representations and computations is feasible in the brain. I will outline how probabilistic internal representations could be implemented in the cortex in a sampling-based manner, and how this can explain a wide range of puzzling observations such as illusions and dreams, as well as the high level of spontaneous activity in the brain. I provide confirmation of this framework by demonstrating that as young animals grow, the visually evoked and spontaneous activity in their brains becomes statistically similar, indicating how their internal model gets tuned to the structure of their environment. Thus this framework offers a rigorous approach to the age-old question of subjectivity and it provides tools for exploring empirically the structure of internal representations in the mind. --------------------------------------------- Venue: CEU Cognitive Development Center Hattyuhaz 1015 Budapest Hattyu u 14. Level 3 (one level up from the entrance level) Everyone is welcome to attend.