Bart Verheij: chairman of the symposium
Biographical note: Bart Verheij is a tenured lecturer/researcher (in Dutch: universitair docent) at the Artificial Intelligence department of the University of Groningen, the Netherlands. His research has focused on defeasible argumentation, legal argumentation, argumentation support software and argumentation schemes, often using formalism as a tool of analysis. A recently added direction of research is agent-based social simulation. In 2005 he published the book "Virtual Arguments. On the Design of Argument Assistants for Lawyers and Other Arguers" (T.M.C. Asser Press, The Hague). In 2006 he coedited the volume "Arguing on the Toulmin Model. New Essays in Argument Analysis and Evaluation" (Argumentation Library, Springer, Dordrecht). Further information is available at http://www.ai.rug.nl/~verheij.
Markus Knauff: How our brains reason logically
The aim of the talk is to strengthen links between cognitive brain research and formal logic. The talk will cover three fundamental sorts of logical inferences: reasoning in the propositional calculus, i.e. inferences with the conditional ”if...then”, reasoning in the predicate calculus, i.e. inferences based on quantifiers such as ”all”, ”some”, ”none”, and reasoning with n-place relations. Studies with brain-damaged patients and neuro-imaging experiments indicate that such logical inferences are implemented in overlapping but different bilateral cortical networks, including parts of the fronto-temporal cortex, the posterior parietal cortex, and the visual cortices. These findings show that we do not use a single deterministic strategy for solving logical reasoning problems. This account resolves many disputes about how humans reason logically and why we sometimes deviate from the norms of formal logic.
Biographical note: Markus Knauff is Professor of Experimental Cognitive Psychology at the University of Giessen. He received his PhD from the University of Freiburg in 1996. During his postdoctoral time he worked at the University of Freiburg, Princeton University, and at the Max-Planck-Institute for Biological Cybernetics. From 2003 to 2006 he was Heisenberg fellow of German Research Foundation. His main research interests are the cortical, computational, and cognitive correlates of complex thoughts. In this field, he is specifically interested in the connection between formal logic, mind, and brain.
Relevant reference: Knauff, M. (in press). How our brains reason logically. To appear in Topoi : International Journal of Philosophy, special issue on Logic and Cognitive Science, edited by Johan van Benthem, Helen Hodges, and Wilfrid Hodges. Link: How our brains reason logically
Jennifer Spenader: Bidirectional reasoning and formal models of language acquisition
It is clear that children’s reasoning skills are not developed to the same extent as adults reasoning capabilities, but does this affect their language skills? In this talk I present both experimental and theoretical results from joint work with Petra Hendriks and Erik-Jan Smits on children’s acquisition of pronouns and reflexives in Dutch and present arguments that the errors made by the children can be accounted for by a combination of problems with bidirectional reasoning, where a hearer has to take into account what a speaker did not say when interpreting what the speaker did say.
Biographical note: Jennifer Spenader got her PhD in Computational Linguistics at Stockholm University in 2003 on presuppositions in spoken discourse. She is currently an assistant professor in language and cognition at the University of Groningen in the Artificial Intelligence Department and is working on a NWO-funded VENI grant project entitled ”The Contribution of Contrast in Context”. Her main interests are computational semantics and pragmatics, child language, bidirectional optimality theory and discourse theory.
Relevant reference: Hendriks, P. and J. Spenader (2005/2006). When production precedes comprehension: An optimization approach to the acquisition of pronouns. Language Acquisition. 13(4), 319-348. Link: When production precedes comprehension
Keith Stenning: Anti-psychologism in psychology
Some professors are prone to denigrate the rationality of their students. Others have been willing to defend their students against this charge at the price of denying that humans reason (at least very much). This talk will seek to reprieve the students, and the rest of the human race, though the professors prove to be a harder case for defense. There may be some morals about the relations, past and hopefully future, between logic and psychology.
Biographical note: Keith Stenning has studied human reasoning and especially its relation to logic and the cognitive effects of learning logic. He was chairman of an Expert Group gathered by EC DG Research which proposed some lines of evolutionary cognitive research under the modest title ‘What it Means to be Human’. He is a Distinguished Fellow of the Cognitive Science Society and a Foreign Fellow of the Royal Netherlands National Academy.
Relevant reference: Stenning, K. and van Lambalgen, M. (in press for 2007) Human Reasoning and Cognitive Science. MIT Press. Especially Chapters 3 and 4. Link: Human Reasoning and Cognitive Science.
Michiel van Lambalgen: The logical form of tasks diagnostic for autism
We outline an experiment on autistic cognition currently performed at the F.C. Donders Center in Nijmegen, and the reasoning that led to it. We initially looked at several tasks on which autists score signficantly worse than neurotypical subjects, such as the false belief task, the unexpected contents task, and the box task. These tasks were developed in very different theoretical paradigms, for example the ‘Theory of Mind Deficit’ theory, or the Executive Dysfunction theory. Careful analysis of these tasks show they have a logical form in common, their diverse theoretical origins notwithstanding. We identified a reasoning task (the ‘suppression task’) that embodies this logical form in more or less pure form. Results show that autistic performance on this task differs significantly from that of controls. The reasoning experiment was a prelude to an ERP study currently under way, whose design will be sketched.
Biographical note: Michiel van Lambalgen is Professor of Logic and Cognitive Science at the Department of Philosophy of the University of Amsterdam. His early work was in mathematical logic, with emphasis on the notion of randomness. After a brief spell in AI, working on reasoning with uncertainty (supported by an NWO PIONIER grant), he decided to switch to Cognitive Science. Since then he has mainly been working on the psychology of reasoning (mostly in collaboration with Keith Stenning) and semantics of tense and aspect, with particular emphasis on how these linguistic categories are processed in the brain (in collaboration with Hagoort and Baggio at the F.C.Donders Center for Neuroimaging in Nijmegen).
Relevant reference: Van Lambalgen, M. and H. Smid, Reasoning patterns in autism: rules and exceptions. In: L.A. Perez Miranda and J.M. Larrazabal (eds), Proceedings 8th International Colloquium on Cognitive Science, Dordrecht, Kluwer, 2004. Link: Reasoning patterns in autism.
Paul Vogt: The emergence of compositionality in populations of autonomous agents
If robots would be able to use logic, they need to have symbolic descriptions of the world around them, as well as of their inner states. Preferably, these symbolic descriptions should be compositional, i.e., parts of the description should refer to parts of its meaning. In this presentation, I present research on how populations of (simulated) autonomous robots can develop a set of conventions that contain compositional symbolic descriptions of objects in their world. I show how this set of conventions becomes gradually more and more compositionally structured, and illustrate the effect that population size has on the way such systems emerge.
Biographical note: Paul Vogt received his M.Sc. in Cognitive Science and Engineering (currently Artificial Intelligence) from the University of Groningen. He performed his Master’s research project at the AI Lab of the Vrije Universiteit Brussel, where he also received his Ph.D. on the thesis ‘Lexicon grounding in mobile robots’. He was a post-doc researcher at the Universiteit Maastricht and at the Language Evolution and Computation unit of the University of Edinburgh. Currently, he is a research fellow at the Artificial Intelligence department of the Vrije Universiteit Amsterdam and a guest researcher at the Communication and Information Sciences Department of Tilburg University. His research focuses on language evolution and acquisition, particularly on those aspects related to symbol grounding and compositionality.
Relevant reference: P. Vogt, The emergence of compositional structures in perceptually grounded language games, Artificial Intelligence, vol. 167 (2005), pp. 206-242. Link: The emergence of compositional structures in perceptually grounded language games.