Fellows Summer Term 2021

Nathalie Bredella

Ekistics: Computers, simulations and planning processes

This project investigates ekistics, an urban planning approach from the 1960s that dealt with global networks, data collection and visualizations. Greek architect and urban planner Constantinos Doxiadis launched ‘ekistics’, or the science of human settlements, as a field of study in the mid–1950s. He promoted this new scientific venture through The Athens Center of Ekistics, the interdisciplinary Delos conferences and the Ekistics Journal, which brought together a global network of researchers working on expanding urban systems.

Ekistics research extended beyond architecture, urban planning and geography to include, among others, economics, anthropology, media theory, sociology, and politics. Methods of computerized analysis, data collection and representation were vital to the study of ekistics, and provided a way to understand the interdependencies between designs, resources, people and settlements. Access to data, it was believed, enabled researchers to develop planning scenarios. Accordingly, maps and diagrams were key research tools, while computers were used to mathematically model settlements, and computer games simulated urban development.

My project aims to better understand the epistemological consequences of the use of computers in the research of ekistics scholars. I am interested in the interdependencies between physical space and computational data modelling, as well as in how digital systems were conceptualized in relation to natural systems. I will therefore use ekistics as an example to discuss the role data collection and visualization processes played in the early days of networking and ‘smart’ planning.

Hans De Raedt

A Quantum Theory Lego Set

The long-term aim of the MECS fellowship is to write a book that describes the physics of fundamental quantum physics experiments, the epistemological questions that these experiments pose, the role of the simulation as a metaphor for perfected, idealized experiments, and the simulations themselves. In the first phase of the project, we will focus of developing the simulation platform and a first version of a graphical interface that allows non-programmers to perform simulations and visualize the results.

Peter Geimer

Stefan Rieger

Simon Roloff

Britta Schinzel und Martin Warnke

Mediation through Computational NeuroScience

Brain simulations are extremely far from actual brains and their performance. So how does the medium of computer simulation intervene in the process of understanding within neurosciences?

As the human brain normally is not directly accessible by invasive inspection, computer simulation is used in order to bridge the lack of knowledge about higher biological neural networks. Simulation and modeling in the context of neuroscience uses the theory of dynamic systems. Although it was developed for complex systems in physics, it has also found its way into the life sciences in order to analyze the behavior of complex natural systems, i.e. to represent and understand them in a simplified and standardized manner, to predict their behavior both theoretically and experimentally in comparison to reality.

However, with regard to the brain, the usual procedure is not possible: that a theory leads to a result through simulation, which is then either confirmed or falsified in the classical experiment. This is, because too little is known about the human brain, i.e. there is no theory of neurophysiology. In addition, usually experiments cannot be performed through a living brain but under very limited circumstances in vitro. In computational neuroscience, simulations therefore have a different status: they serve as experimental systems on computer models of the brain.

Experiments on computer models can succeed in simulating meaningfully for small layer sections with sufficiently precise layout assumptions. A theory then is not about the properties of the neuron layer in the brain (neither topographically nor functionally), but about the respective simulation. Various experimental results can be integrated into the computer models, as well as anatomical data, connection probabilities of the structural models, and much more. The simulation here is a description medium with which one can connect and compare findings from different areas and scales: it generates theories. It takes the vacant place of the nonexistent neurophysiological theory, it forms material-semiotic material.

A workshop “Neurological Computer-Simulations” collected well known scientists in the field of Computational NeuroScience, Media Theory, STS, Philosophy of Technology and Gender Studies in Technology. The demanding project investigates in the accessibility of brain sciences through simulation methods, the epistemological meaning of the simulation for scientific investigation of the brain, as well as especially in the media theoretic meaning of the nature of neurological observation through simulation.