Software and demos of computer simulations

ECCO's general research into self-organizing and cognitive systems is made more concrete by building computer models of such systems. This allows us to experiment with their behavior, and better understand how they function or solve problems. Here are some examples in the form of interactive demonstrations, websites, or downloadable software.


Knowledge development websites

Martin M. (2009). Teacherplex: from testing to teaching

This experimental website implements the principle of inverted learning: instead of first proposing new information, and then assessing the degree to which the student has assimilated the information, the process starts with a test. This takes the form of a traditional "quiz" or series of multiple-choice questions. Selecting one of the choices (correct or not) produces not only the correct answer, but an explanation of why this is the correct answer. The initial question functions as a challenge that stimulates the learner to think about the issue, thus motivating him/her to study the explanation. Questions can be created and edited by anyone, in the style of a Wiki. Questions and test-takers automatically get assigned a grade level depending on the average number of people answering a question correctly.

Nagarjuna G. (2006-...). GNOWSYS: developing a semantic network of concepts

This website allows anyone to enter concepts and the relationships between them, thus creating a semantic network representation of a domain. The idea is to collaboratively create an extensive knowledge map that will help others to study a domain.

Rodriguez M. (2008-...). kReef: recommendation of relevant papers, activities and authors

This website helps scientists find relevant material for their research, by proposing other researchers, papers, conference etc. that are related to your own interests,. The underlying database uses  a semantic network (labeled graph) representation of people, publications and activities connected by links such as "authored by", "cites", "participated in", etc. Particle swarms explore the network to find the entities most related to a given query.

 


Network models

Rodriguez M. & Heylighen F. (2005). WordScore: a simulation of verbal IQ test performance

This simulation uses propagation of "energy particles" through an associative network of words to find which word is most/least related to a set of given words. It solves IQ-test-like questions of the following form, where * designates the correct answer (at present it gets about 75 % right):

  • Which word of the second list best fits in the first list? dog, cat, bird, fish : bush, pig*, house, car
  • In the following list, which word is the odd one out? dog, cat, car, pig, cow : dog, cat, car*, pig, cow

You can invent new questions yourself and try them out in the web interface to the program (no longer online).

 

Rodriguez, M. (2005). Peerper: automatically discover referees for journal and proposal submissions

Peerper uses scientific coauthorship networks generated from arXiv and CiteSeer e-print servers to create a map of the scientific community. Someone wishing to locate referees for a particular manuscript can submit the manuscript and the references in the manuscript stimulate the network at the nodes refering to the referenced names. Then a spreading-activation algorithm moves the energy throughout the network. Once all the energy in the network has decayed, the most energetic nodes are returned by the system as being the potential referees for the submitted manuscript. You can try Peerper version beta online via the web interface to the program (experimental -> not always online).

Rodriguez, M. & Steinbock, D. (2003-2005). Confluence: Particle-Flow Network Simulation Package for Java

Confluence is a generic network package that has been used to model collective decision-making, artificial IQ-testing, and collaboration in coauthorship networks. It forms the foundation for many of the network software used by ECCO and is constantly being updated and expanded. For a in-depth look at the features of Confluence please visit the Confluence website

Gershenson C.(2001-2004) : RBNLab

A Software Laboratory for studying the properties of different types of random Boolean networks. It can calculate attractor statistics, graphic representation of dynamics, topology graph, etc. Try the simulation Related papers

Van Overwalle F. (2002-2005): Fit2 for Connectionist and Algebraic Modeling

This program allows you to simulate prominent connectionist and algebraic models of learning and induction. You can directly compare the simulation output with real observed data from actual psychological experiments (hence its name FIT). In addition, you can automatically search for the parameter values of the simulated model that best fit with your actual data. More info and download (for Windows).


Multi-agent simulations

Gershenson, C. (2005). Self-Organizing Traffic Lights.

Using simple rules and no direct communication, traffic lights are able to self-organize and adapt to changing traffic conditions, reducing waiting times, number of stopped cars, and increasing average speeds. Try the simulation with your Java-enabled browser. See Also: Gershenson, C. (2005). Self-Organizing Traffic Lights. Submitted

Gershenson C. (2001-2002) : KEB: Knowledge Emerging from Behavior

A virtual laboratory for comparing different types of cognitive architectures, implemented in different animats. There are rule-based, behaviour-based, concept-based, neural network, and Braitenberg-style animats. Simulation and more information See Also: Gershenson, C. (2004) Cognitive Paradigms: Which One is the Best? Cognitive Systems Research 5(2):135-156, June 2004.

Gershenson C.(1999-2001) : ASIA: a Behaviours Virtual Laboratory used for studying Artificial Societies of Intelligent Agents.

This is a virtual laboratory where predator and prey animats try to survive in their environment.Useful for experimenting with virtual creatures, but also for teaching and learning animal behaviour. It includes the following features; Imitation and induction of behaviours. Reflex, reactive, and motivated behaviours. Adaptation of motivation degree, classical conditioning. For more information, please check the ASIA homepage. Simulation, Help, About