November 25th, 1.15pm
Olds-Upton 303

This presentation will use simple mathematical models to examine the interaction between a non-addicted and susceptible population, a population of addicts/pushers, and law enforcement. A description of how this model was developed employing components from the Lotka-Volterra predator prey model, Kermack-McKendrick epidemic model, Richardson arms race model, and Lanchester combat model will also be discussed. A further analysis of the parameters, diffusion terms, and robustness of the model as explored using the program XPP will help shed light on the usefulness of this model in describing the system of drug propagation under the given constraints. Then the presentation will conclude with a look at what can be done to improve the model with further work.

For a pdf version of the talk read http://max.cs.kzoo.edu/~bboven/drug_dynamics.pdf.

November 20th, 4.10pm,
Olds-Upton 207

This talk is motivated in general by the renewed interest in network formation and in particular by possible application of mathematical theories and computer science in the study of social structures such as co-authorship or friendship networks "old-boy" networks, etc.

In the first part of this talk we will introduce the world of networks. We will show that the concept of interconnected entities appear in several context both in our everyday life and in different aspects of science from physics through chemistry, biology, economy to social sciences and humanities.

The second part contains an introduction to a possible mathematical description of networks: a brief overview of real-world graphs. As we are specifically interested in the evolution of social systems we will concentrate on graphs related to them and measures useful to numerically characterize such graphs with special attention paid to the (recently so-called) scale-free graphs.

In the third part of the talk models of network development relevant for studying the formation of different social systems will be examined. We will touch on the Erdõs-Rényi (the random graph), Watts-Strogatz (the small-world graph) and Barabási-Albert (graph with growth and preferential attachment) models describing their algorithms and the properties of networks they generate. Finally we will discuss the model we developed to simulate friendship formation and show results calculated analytically or generated by computer simulations.

Intended audience of this talk are mainly math, computer science, physics, sociology and psychology students and faculty interested in network formation, the description of social networks and applications of graph-theoretical tools and numerical simulations.

Contact persons:	Gábor Csárdi
	Tamás Kiss
	Péter Érdi (tel: 337-5720)

November 13th

October 10th, from 11.30
Olds-Upton 408

October 10th, from 4.00pm
Dewing 103

This talk, intended for a general interdisciplinary academic audience, stresses the "ruthlessly reductionistic" nature of current "mainstream" neuroscience. When learning about (admittedly fascinating) techniques now used in cognitive neuroscience, like functional neuroimaging (fMRI, PET) and neural network modeling, non-neuroscientists can easily get the impression that neuroscience has abandoned its "single-cell" core in favor of "autonomous higher level" investigations. But this impression is mistaken. "Mainstream" neuroscience remains "reductionistic"; and with the now-standard use of sophisticated molecular biological, biochemical, and bioengineering tools (including "gene knock out" and "transgenic" mammals), the discipline continues to unravel the intracellular and molecular mechanisms underlying behavior, and to directly manipulate these mechanisms to generate novel behavioral data. I'll illustrate the nature of this "ruthless reductionism" by focusing on a case study: Eric Kandel and his colleagues' recent work linking the intracellular and molecular-genetic mechanisms of Long-Term Potentiation (LTP), a form of experience-driven synaptic plasticity, to memory consolidation in behaving mammals.

This is an LAC lecture.

Contact persons:	Bob Batsell (tel: 337-7032)
	Péter Érdi (tel: 337-5720)

May 15th, 4.10pm - 5.45pm
Olds/Upton 207

The intention of the new Kalamazoo College program in computer modeling of complex natural and social systems, related to the Henry R. Luce Professorship, is to contribute to the education of both science and non-science students by adopting highly inter- and transdisciplinary approaches.

To promote the building of the bridge between disciplines of natural sciences on one side, and social sciences and humanities on the other side, the Center for Complex Systems Studies (CCSS) has been founded.

Greg Mahler: Introduction Péter Érdi: CCSS: why and how (5 minutes) Richard Gejji: Marriage contract in the Talmud: or how did the Rabbi use game theory? (30 minutes + 5 minutes discussion) Matias Vernengo: The New Economy and the Limits to Growth (30 minutes + 5 minutes discussion) Jan Tobochnik: Closing words

Faculty and students are welcome!
Refreshments are served from 3.45pm.